Annual Report 2014–2015 3

5 Foreword

6 Academics contents 7 Imaging Science Undergraduate Program 12 Motion Picture Science Program 15 Imaging Science Graduate Program

20 Research 21 Digital Imaging and Remote Sensing Lab 29 Laboratory for Advanced Instrumentation Research 35 Multidisciplinary Vision Research Laboratory 39 Biomedical and Materials Multimodal Imaging Laboratory 49 Magnetic Resonance Imaging Laboratory 53 Laboratory for Advanced Optical Fabrication, Instrumentation and Metrology (AOFIM) 57 Optics 61 Nanoimaging Research Laboratory 67 Laboratory for Space Weather Alert Technologies 69 Historical Manuscript Restoration 77 Laboratory for Multiwavelength Astrophysics

87 Outreach 4

RIT Imaging Science 2014–2015 Annual Report FOREWORD 5

By David W. Messinger, Ph.D.

On behalf of the faculty, staff, and students of the Chester F. Carlson Center for Imaging Science at RIT, I am pleased to present our Annual Report for the 2014-15 Academic Year. I hope you enjoy reading about all the exciting research and educational activities of the Center over the past year. During the past academic year the Center has continued to be a leader at RIT in research focused on undergraduate and graduate education. The Center was awarded almost $5.5M in externally sponsored research funds during this period! The represents a significant amount and allows us to continue our excellence in broad, multidisciplinary approaches to solving problems with imaging systems. The graduate program has over 100 students pursuing either MS or Ph.D. degrees in Imaging Science. Our undergraduate program also continues to be strong with over 50 undergraduates enrolled in the BS degree program. Additionally, our undergraduate and graduate students, as well as our faculty, continue to be recognized for their accomplishments. In this past year, Rose Rustowicz, who graduated with her BS in May, received a Fulbright Scholarship to travel to Iceland to work with a research team from the University of Iceland to conduct remote sensing at the Hekla volcano. Liz Bondi, who will be a senior in 2015-16, received a very prestigious Goldwater Scholarship. At the faculty level, Jan van Aardt was awarded the RIT Trustees Scholarship Award for his outstanding research portfolio. Also, a journal paper published by Dr. Aaron Gerace and Dr. John Schott in the Journal of Applied Remote Sensing was designated as one of the most downloaded papers published by JARS in the past year. Other research highlights include Dr. Grover Schwartzlander’s work for NASA in using “glitter” as a mirror to formulate images, Dr. Maria Helguera’s work to use ultrasound technology to create tiny blood vessels needed to nourish organs, and Drs. Naval Rao and Hans Schmitthenner’s work in photo-acoustic imaging for prostrate cancer detection. Not to mention our Freshman Imaging Project and Ph.D. graduate lab projects developing novel imaging systems for transient light imaging and crime scene investigation as part of our experiential learning program! Our summer programs continue to be a major success, at both the High School and undergraduate levels. Our NSF Research Experience for Undergraduates program in “Imaging in the Physical Sciences” continues to provide real, hands on research experience across many imaging science applications to undergraduates from around the country. Our High School intern program provides a unique research experience for rising seniors to come into the Center for several weeks during the summer and learn about Imaging Science. Additionally, the High School intern program has once again contributed to the growth of our incoming freshman class, with 3 incoming students in the Fall of 2015 having been interns in the summer of 2014! This has been an exciting year, and I hope you enjoy reading this Annual Report and learning about the activities of the Chester F. Carlson Center for Imaging Science. Please contact us for more information or if you have other ideas for Imaging Science!

David W. Messinger, Ph.D. Professor and Director Chester F. Carlson Center for Imaging Science Rochester Institute of Technology ACADEMICS ACADEMICS6

RIT Imaging Science 2014–2015 Annual Report ACADEMICS 7 IMAGING SCIENCE UNDERGRADUATE PROGRAM Program Coordinator’s Comments By Dr. Maria Helguera

As soon as I came back from sabbatical I stepped back into my responsibilities as the Undergraduate Program Coordinator. Like Carl Salvaggio, I am totally devoted to our program and I am sure that for both of us it is difficult to step aside. I thank him for the seamless transition.

This past year marked the fifth installment of our year long Innovative Fresh- man Experience. A total of 14 students from Imaging Science and Motion Pic- ture Science were enrolled. This year the class was tasked with building a transient imaging system based on the work developed at the University of British ACADEMICSColumbia (UBC). As always, the project seemed overwhelming to begin with. As always, the students proved themselves to be capable of surmounting every challenge along the way. We were fortunate to recruit Dr. Matthias Hullin -the original developer of the system in UBC- to come for a visit. Our students enjoyed a couple of days in his company and had a chance to ask many questions and learn what was needed of them in order to be successful. They presented their working prototype in ImagineRIT in May. They got invited to participate in the Rochester Museum and Science Center’s Year of Light celebration. Thanks to our combined efforts, Carl and I managed to get the first semester of this wonderful experience to be recognized as general science, meaning that university students can now take Imaging Science courses as their general education concentration. We started the academic year with a very small Imaging Science class, six students. However, throughout the year four students switched their majors to Imaging Science and three more intend to pursue a double major in combination with Imaging and Photographic Technology, and Motion Picture Science. As always, we continue to adapt and renew our undergraduate curriculum. We want our students to be competitive in the market and be solidly prepared if they choose to pursue graduate studies. This past year we adapted/created two courses, Probability and Statistics for Imaging is a sophomore year course. As the title implies our students will not only learn the mathematical foundations but will apply concepts to imaging, image analysis, etc. Imaging Systems Analysis is a four credit course in the junior year. This course will teach students about the physical factors that affect the spatial and temporal response properties of optical, electronic, and biological imaging systems, and the mathematical methods that have been developed for describing these properties. We continue to offer our National Science Foundation sponsored Research Experience for Undergraduates (NSF-REU) in Imaging in the Physical Sciences. This year the following student have joined us: Christian Cammarota, from RIT working in optics with Prof. Swartzlander Matthew Portman, from University of Texas, Dallas working in astronomy with Prof. Kastner Merlin Hoffman, from Oberlin College working in MRI with Prof. Hornak Klemens Gowin, from University of experience. Several of them spent at Megan Iafrati 8 Rochester working in optics with Prof. least one semester abroad and all of Rose Rustowicz Qiao them where involved in internships. SPIE Optics and Photonics Scholarship Daniel Morgen, from University of ISC organized a min-symposium on Kevin Sacca Rochester working in optics with Prof. May 19 in which the seniors presented Qiao the results of their capstone projects. Fulbright Fellowship The experience was so successful that Vikie Ngan, from Wellesley College the faculty members of the Under- Rose Rustowicz working in remote sensing with Dr. graduate Curriculum Committee have Barry Goldwater Scholarship Gerace decided to assume responsibility and Laure Morehouse, from Cornell Uni- organize the event in the years to Elizabeth Bondi versity working in devices with Prof. come. Stay tuned; you will be receiving Ninkov an invitation when the time comes. American Society for Photogrammetry This will be an excellent opportunity to and Remote Sensing (ASPRS) Student of Bruce Jones from RIT working in crime meet our talented students. the Year scene with Prof. Dube Overall, the 2014-2015 academic year Elizabeth Bondi Aviriana Follet, from Monroe Commu- was a great success as you will see from United States Geospatial Intelligence nity College working in crime scene our students’ accomplishments below. Foundation (USGIF) GEOINT Hackathon with Prof. Dube We would love to hear from you, wheth- Caroline Trier from Washington Uni- er it is by providing an internship, co- Briana Neuberger versity in St. Luis working in biomedi- op, visiting campus and talking at our Dan Simon cal imaging with Prof. Linte. seminar series on Wednesday afternoon or to the student-run Imaging Science Alpha Sigma Lambda Honorary Society This past year I renewed my connec- Club on Friday afternoon. tions with Monroe Community College Rose Rustowicz and have met with the Optical Adviso- Awards/Recognitions UNYTE—Hitting the Accelerator: Health ry Board. We do have an articulation RIT Outstanding Undergraduate Research Innovation through Data Sci- agreement in place and a few students ence. Best Student Poster Award have already expressed their interest in Scholar Award pursuing a 2 + 2 program. I am looking Elizabeth Bondi Rose Rustowicz forward to continuing this collaboration. RIT College of Science—John Wiley Jones Publications, Conference and Poster Student Body Scholarship Presentations Our program continues to be small Ryan LaClaire Hitchner, A., Vodacek, A. “Accessing with about 45 students. We do have the Accuracy of Cloud Height Mea- students pursuing double majors with RIT Women in Computing Hackathon— surements using Landsat 8 Images”, Industrial Engineering, Applied Math- Most Novel Hack 2014 RIT Undergraduate Research ematics, Motion Picture Science and Victoria Scholl Symposium Imaging and Photographic Technology. Elizabeth Bondi Jermyn, A., O’Dea, C. “Infrared Ob- More and more students are pursu- servations of Low-luminosity Radio ing internships and co-ops as well Chester F. Carlson Center for Imaging Galaxies”, 2014 RIT Undergraduate as benefitting from REU programs all Science—Carlson Scholarship Research Symposium over the country. Some of the places Ryan Connal receiving them include Colorado State Bondi, E., Easton, R. “Spectral Image Nathan Dileas Processing Applied to Manuscripts of University, Ball Aerospace, Exelis, Sara Leary University of Buffalo, Wellman Center Cultural Importance from the Museo Ryan Hartzell del Tesoro del Duomo in Vercelli, Ita- for Photomedicine at Massachusetts Makayla Roof General Hospital, University of Roch- ly”, 2014 RIT Undergraduate Research Dylan Dang Symposium ester, NASA JPL, United Technologies Zach Mulhollan Aerospace Systems, NGA, OGSystems, Kevin Sacca Rustowicz, R., Schultz, M., Pow, J, "Do Boeing, Farm Crisp, Army Research Victoria Scholl Innovative Curricula Actually Improve Lab, Vencore, QVI, several labs in CIS- STEM Retention: A Case Study”, 2014 RIT, etc. Chester F. Carlson Center for Imaging STEMtech Conference, Denver, Colorado, Science—Fitz Scholarship We were lucky to have a very active United States, November (2014). and engaged senior class. Throughout Amy Becker Peck, D. S., Schultz, M., Bachmann, C. the year some of them volunteered in M., Ambeau, B., Harms, J. “Influence the executive committee of the Im- Chester F. Carlson Center for Imaging of density on hyperspectral BRDF sig- aging Science Club (ISC) organizing Science—Hughes Scholarship natures of granular materials,” Proc. weekly meetings, some worked as Matthew Casella SPIE 9472, Algorithms and Technol- teaching assistants (TA), and resident ogies for Multispectral, Hyperspec- assistants (RA). During these past four Chester F. Carlson Center for Imaging tral, and Ultraspectral Imagery XXI, years they really used every opportu- Science Faculty Award Celebrating 94720F (May 21, 2015), http://dx.doi. nity they could to expand their knowl- Excellence org/10.1117/12.2177455. edge and improve their academic

RIT Imaging Science 2014–2015 Annual Report Qiao, J., Travinsky, A., Ding, G., Dang, ject and the individual focal planes Advisor D., Dorrer, C. “A differential wavefront coincide with a lens designed with Zoran Ninkov 9 sensor based on binary pixelated chromatic aberration. The various transmission filters”,2014 NASA Mir- contrast response of the lens-illumi- TITLE ror Technology Workshop, Albuquerque, nation configurations provide a tool Development of Control Hardware for to increase the depth of field of the New Mexico, 18–20 Nov, 2014 the Audine CCD Camera imaging system and discrete contrast Iafrati, M., Rustowicz, R., Helguera, M. responses. An algorithm to decipher Abstract: The Audine CCD camera kit “RAM: Remotely Accessible Micro- the focal axis position and/or com- is a package that allows the user to scope”, UNYTE—Hitting the Accelerator: parison of focal positions of various assemble a camera on his or her own Health Research Innovation through sources and method is proposed. The - giving the user a better understand- Data Science, University of Rochester, algorithm is intended to significantly ing of how the imaging system works, May 28, 2015. increase throughput while maintain- how to repair any potential problems, Díaz, G., Binaee. K., Smith, A., Griffo. ing positional accuracy for various and eliminating any cost associated C., Gopinathan, R. “Evaluation of a sensing methods. with purchasing an already assem- Phantogram Groundplane for the bled camera. The standard interface Study of Visually Guided Walking Students and software used to control the Behavior”, Annual Meeting of the Vision Megan Iafrati camera is outdated, and only works Science Society, St. Pete's Beach, Flori- Rose Rustowicz with old computers running Microsoft da (May 2015) Windows 95. This project developed Advisor a new USB controller interface for Dang, D. N., Linte, C.A., Otani, N.F. the Audine camera that can be used “Reconstruction and Visualization of María Helguera with modern computers, as well as Action Potential Wave Propagation software that can be used to perform Patterns in Cardiac Tissue”. Accepted TITLE basic image capture. The prototype for oral presentation 2015 Computer Remotely Accessible Microscope (RAM) for the system was built using a methods in Biomechanics and Biomed- Abstract: Biologists and other research Texas Instruments TIVA TM4C1294 ical Engineering conference, Montreal, professionals have a need for an inex- micro-controller that provides com- Canada, Sept. 2015. pensive, easy-to-use microscope with munication between the camera and the users computer. The controller re- Senior Projects image processing and analysis capabilities. Although the applications for this ceives a signal from the users comput- Our graduating class worked on their type of microscope may be endless, the er, sends the necessary clock signals senior project as part of the require- motivation for developing this tech- to the sensor, reads the digitized ments for degree completion. Cap- nology comes from the desire to view signal from the camera, and sends stone projects prepare them for a spe- cell samples within an incubator over the collected information back to the cialty within imaging science or for time. In current technologies, the cells computer. The completed package can graduate school in the coming year. must be fixed and destroyed before be used by RIT students who build the As you can see from the list below, viewing under a microscope. Samples CCD camera in the laboratory, and can their interests are wide and varied. must be transported to a microscope be distributed for use by others in the each time they are observed, risking imaging community who build their Student sample contamination and interruption own Audine camera. Daryl Schwarz of cell processes. To mitigate the need for sample transportation, a remotely Student Advisor accessible microscope was developed. Christian Taylor Jie Qiao A successful system requires optics, electronics, and image processing. By Advisor TITLE adapting a Makeblock XY-Plotter kit as Carl Salvaggio Chromatic Autofocus Method for Auto- an XY motor stage, utilizing the optics mated Video Metrology Systems from an Adafruit USB microscope, and TITLE taking advantage of the capabilities Radiometric Calibration of a Modified Abstract: Automated video metrol- of a Raspberry Pi and it’s compatible ogy systems rely on accurate axial camera board, components were inte- DSLR for NDVI positioning of the imaging sensor grated into one fully functional system. Abstract: Silicon CCD detectors to enhance performance of sensing Graphical user interfaces (GUIs) were found in commercial DSLR cameras algorithms, e.g. edge detection. Tra- developed for motor control and image are responsive far out into the Near ditional autofocus methods rely on a processing. Image processing features Infrared wavelengths. These cameras contrast sweep through the imaging were developed specifically for biolo- typically have an Infrared block filter systems depth of field. The sweep gists needs and include field flatten- built in to more closely resemble inherently produces a lag in inspec- ing, adaptive histogram equalization, the human visual system. This paper tion throughout which correlates to and segmentation for cell counting. To proposes a method for radiometric fewer parts per hour. U.S. Patent 7 further improve the capabilities of the calibration in order to extract NIR and 812 971 B provides the intellectual system, future work is anticipated. RED Radiance bands from a Canon property to use an apparatus in which 300D with the IR filter removed. With an illumination source with discrete Student these spectral bands a single expo- wavelengths are incident on an ob- Ryan LaClair sure NDVI measurement can be produced for each pixel in an image for expand upon these earlier results abundances, of a mixed pixel. Linear identification of live green vegetation. that were conducted primarily in the hyperspectral unmixing often relies 10 principal scattering plane and only at on ground truth data to assess the Student minimum and maximum densities. In accuracy of the unmixing model. In Andrew Hitchner the present study, the BRDF of gran- the case where ground truth data do ular composites is compared along not exist, research is done to evaluate Advisor a gradient of densities for optically if the unmixing model follows statis- contrasting materials. Systematic tical logic. This project qualitatively Anthony Vodacek analysis of angular and material evaluated the produced abundances dependence will be used to develop utilizing data of two different spatial TITLE better models for multiple scattering resolutions and two different linear Cloud Height Data from Landsat 8 Ther- effects of the granular materials. The hyperspectral unmixing algorithms. mal Images measurements in this experiment We found the same cover types fell Abstract: One of the major sources of used the recently constructed, labora- in the same groups of under-mixed error in climate models is the lack tory and field-deployable Goniometer and over-mixed. These over- and of information about how clouds of the Rochester Institute of Technol- under-mixtures were attribute to interact with the climate around ogy (GRIT), which measures BRDF for spectral BRDF variation. them. To improve climate models, a geometries covering 360 degrees in method for determining cloud heights azimuth and 65 degrees in zenith. In Student at a high resolution is needed. This contrast to the previous studies lim- Malachi Schultz project presents a proof of concept ited to the principal scattering plane, for determining cloud heights using GRIT provides a full hemispherical Advisor BRDF measurement. the thermal images from the Landsat Charles Backmann 8 satellite. The method involves using Student mutual information image registra- TITLE tion to find the parallax shift between Cicely DiPaulo NASA Algodones Experiment clouds in the Band 10 and Band 11 images, and then uses this shift, along Advisors By examining the shape of the angular distribution of the hyperspectral with newly available look angle data Joel Kastner, Charles Backmann for each band, to determine the cloud reflectance curve, or BDRF, we can get a better understanding of the height. This method is a viable meth- TITLE geophysical properties of sand, such od, however there are some sources 3D Perspective on Visible Light & X-rays of error and uncertainties that need to as density, grain size distribution, from Planetary Nebulae: From the Ant to be addressed and considered before and moisture content. In this research creating an automated method to the Eskimo we are measuring BRDF of varying generate cloud height data. There are millions and millions of sand compositions, moisture content, stars in our sky each with their own density, and grain size distributions Student unique makeup and morphology. in both the field and the controlled conditions of a laboratory where each Doug Scott Peck Planetary nebulae are formed by low to intermediate mass stars that swell geophysical variable can be inde- Advisor up when they reach the end of their pendently controlled. Based on the lives forming cool, red giants, expel- discrepancy between the model and Charles Backmann ling their outer layers and beginning the real world we can refine Hapke’s the transformation into white dwarfs. model to more accurately predict the TITLE To gain insight into 3D perspective of geophysical properties of sand in Influence of Density on Hyperspectral planetary nebulae, we have used HST aerial and satellite imagery. BRDF Signatures of Granular Materials and Chandra data along with SHAPE, Recent hyperspectral measurements a modeling program, to demonstrate Student of composite granular sediments of similarities in morphology. Andrew Smith varying densities have revealed phenomena that contradict what radiative Student Advisor transfer theory would suggest. In Susan Kratzer Gabriel Díaz high-density sands where dominant constituents are translucent and sup- Advisors TITLE plementary, darker grains are present, Jan van Aardt, David Messinger Evaluation of a Phantogram Ground- bidirectional reflectance distribu- plane for the Study of Visually Guided tion function (BRDF) measurements TITLE Walking Behavior of high density sediments showed Assessing Linear Unmixing Algorithms reduced intensity when compared to A phantogram is a two-dimensional lower density counterparts. It is con- and their Performance on Low and High projection that, when looked upon jectured that this is due to diminished Spatial Resolution Hyperspectral Data from the proper angle, appears multiple scattering from the darker Hyperspectral unmixing utilizes pure three-dimensional to an observer. The particles which more optimally fill pixels’ spectra as endmembers (i.e. concept is the same as for the recently pore space as density increases. The each pixel contains only one cov- popular street chalk drawings, where goal of these experiments is to further er type) to unmix the fractions, or the artist appears to draw a large

RIT Imaging Science 2013–2014 Annual Report chasm in the ground or something the Naval Research Laboratory. access to water. They were awarded similar. Our implementation builds Briana Neuberger, a soon to be complimentary registration to GEOINT 11 on this concept by computational- senior at Rochester Institute of Foreword, the pre-GEOINT symposium ly re-rendering the groundplane in Technology (RIT) double majoring science and technology day held June real-time according to the subject’s in imaging science and industrial 22. “Team Agile” members are: physical head position. The result is systems engineering. Neuberger is Nathan Currier, an incoming senior a very realistic illusion, where virtual a SMART scholar and intern with at Colorado State University, Fort objects lying on the ground appear the National Geospatial-Intelligence Collins majoring in computer sci- real to the subject. Agency (NGA). ence as well as an intern with Sting- Two CIS Undergrads On Winning Dan Simon, a rising senior at RIT er Ghaffarian Technologies (SGT). GEOINT Hackathon Team and intern at OGSystems. Jesse Pai, a rising sophomore ma- Paul Warren, a rising junior at Stan- joring in computer engineering at Developers challenged to produce pre- the University of Maryland, College dictive analysis of disease in West Africa ford University majoring in computer science as well as an OGSystems Park. Pai is also an SGT intern. Nearly 30 developers and data sci- intern. Team Agile described their solution entists turned out for the first USGIF as “a lightweight web app” that GEOINT Hackathon June 12-14. “We developed an open-source python library to model the spread of would be ideal for doctors working in disease as it’s carried by contagious environments with limited connectiv- people through a network of nodes ity. Both Currier and Pai are new to and edges using network theory,” geospatial intelligence. Warren said. “The main thing that I liked about Simply put, Team Intern’s library this experience was the fact that I aimed to capture where sick people was forced out of my comfort zone,” travel and why. Pai said. “I was given a new task and objective, and I had to learn a new “Once we came up with results, we set of information and tools in order developed a way to visualize them so to understand the task and produce a they could make meaningful sense,” result. This hackathon allowed me to Mason said. experience what it felt like to devel- Participants were tasked to determine The team attributed Neuberger for op for another part of the computing why certain areas of West Africa were helping “expose the way they think by industry.” unaffected by the Ebola outbreak as figuring out how they think.” well as predict where additional out- The GEOINT Hackathon was spon- breaks might occur. However, this was The second-place team produced what sored by DigitalGlobe, Esri, and their secondary goal—the primary it calls “non-historic” predictive anal- OGSystems, and included judges from goal was to expose their team’s think- ysis and was awarded complimentary USGIF, NGA, DigitalGlobe, Esri, and ing and build in hooks so another GEOINT 2015 registration. “Team Flo OGSystems. team working with another geography Hacks” members are: —See more at: http://trajectorymag- or outbreak could modify the solution Boris Polania of Hollywood, Fla., a azine.com/got-geoint/item/1967-stu- to a new set of conditions. software engineer with post-grad- dent-interns-win-first-geoint-hack- “It was a fabulous event,” said Dr. uate studies in economics who athon.html#sthash.6C1TbIU2.dpuf Darryl Murdock, USGIF’s vice pres- moved to the U.S. from Venezuela ident of professional development. six years ago. Polania recently “The judges learned as much as the helped found small software con- participants did and we can’t wait to sulting firm V/F. host another hackathon.” Armando Umerez of Boca Raton, The first-place team included four Fla., who recently moved back to student interns and was aptly named the U.S. from Venezuela and is also “Team Intern.” Their solution focused a partner in V/F. Umerez is an elec- on travel and revealed an “Ebola tronic engineer with post-graduate superhighway” along the coast of studies in marketing, management, West Africa. They were awarded the and sustainable development. $15,000 grand prize as well as com- “The standard approach of predictive plimentary registration to USGIF’s analytics is to go to old data sets GEOINT 2015 Symposium, to be held and do standard clustering to make next week in Washington, D.C. Team generalizations,” the team said in its Intern members are: presentation. “We didn’t want to do R. Blair Mason, a member of the more of the same, so we did a com- U.S. Naval Academy class of 2016 pletely new approach.” and a double major in computer The third-place team developed an science and aerospace engineering. easy-to-use graphical user interface Mason is currently interning with based on sanitation data such as Motion Picture Science Program of light in flight at Imagine RIT. In fact, (HFR) domain have resulted in a flurry the experience of past MPS students of passionate response from profession- 12 Prof. David Long, Chairperson constructing a low-cost stereoscopic als and the general public alike. The projection system was key to this year’s term ”soap-opera look” has once again 2014-2015 marked another wonder- success. Project management, commu- entered into the public’s vernacular to fully successful year of collaboration nication and research skills gained here describe the percept. Frame rate has between undergraduate students in will form a great foundation for all of clearly proven to be a pivotal aspect RIT’s Motion Picture Science (MPS) these students navigating the remain- of an observers perception of cinema and Imaging Science programs. der of their undergraduate careers. content, and yet the underlying causes In particular, students studying in Upperclassmen also contributed to of this perception are unknown. This MPS as part of the School of Film research hopes to take the first step to- and Animation continue to reap the some impressive research both as members of select research groups wards a veridical approach to modeling benefits of world-class classroom and this response in a high-order manner laboratory opportunities in CIS. On a within CIS and through their capstone engineering projects. Work ranged with respect to motion pictures to help campus with so many imaging-centric inform future decisions in the creative degree programs in the sciences and from high dynamic-range video system design to psychophysical modeling of story telling industry and future human arts outside of Imaging Science, the temporal vision research. rigor and quality of education shared human response to varying framerates in cinematic exhibition. Several papers with our students by CIS faculty gives Matt Donato them a unique advantage in the mar- were submitted for conferences as a re- ketplace. The cinema industry contin- sult of research efforts. On an extra-cur- Towards Standardizing a Reference ues to take advantage of this bright ricular front, members of the student White Chromaticity for High Definition new talent. And we are motivated to chapter of the Society of Motion Picture Television make these bonds even tighter! and Television Engineers designed and built a video system for capturing Emerging high definition displays exhibit Whenever prospective students and live-action virtual reality content and great spectral variance with respect to families from high school come to visit produced a short film that they showed their unique illuminant technologies. me, I’m excited to tell the stories of off at Imagine RIT. The demonstration As a result, their spectral power distri- the strong engineering and research was a hit and a testament to the energy butions differ. When calibrating these opportunities available to MPS students our students can bring to fun side spectrally unique displays, calibration at RIT. And quite of few of them are just projects. The team was even extended instruments claim chromaticity matches a little bit surprised to hear about all we a personal invitation to present their while film industry professionals (nat- do here. Students contemplating MPS work at the 2015 SMPTE Technical urally diverse in their color perception) are quite unique. They are drawn to Conference in Los Angeles. experience chromaticity mismatch- film and to the art of cinema but every es. This was not an issue when early one of them has technical talent and In May, 11 MPS students completed phosphor-based displays in use were a desire to keep their left brains fully their degrees and officially joined the spectrally similar to one another. Both engaged in school. I joke with many of real world. Three will be attending a simulation and psychophysical test them that the motion picture industry graduate school in color science and are developed in an attempt to quantify is a lot like another CIS favorite, NASA. computer science while the rest head differences in color perception between So much technology developed for the off to new careers. Most have landed classic phosphor-based displays and space program spawned products and positions in the motion picture indus- emerging high definition displays. The systems that engineers and scientists try but a few will also be starting with proposed method can be used to effec- continue to take out to the rest of the non-cinema imaging firms. Again, we tively determine a mean visually-cor- world today. For imaging systems, the have always been very proud that the responding chromaticity offset from a motion picture marketplace is that same CIS connection permits our students to given standard white point chromaticity high-tech playground. Cameras, dis- contribute immediately to many differ- for both LED and OLED displays to satisfy plays, computer graphics, computational ent kinds of imaging groups, not just a greater population of observers. While photography—they all have roots in the the movie-making ones. So look out for this offset may be satisfactory for a great- cinematic arts. This is what makes RIT MPS alums in New York and Hollywood er number of observers, a single observer so fun for students in both programs. for sure, but don’t be surprised to find model cannot accurately predict meta- And the collaboration is so infectious them in Milwaukee, Denver and Boston meric matches for an entire population of that we actually now have three students this year, too. diverse observers. This issue is mag- on campus earning both degrees! nified as three-primary color rendering Abstracts taken from sample MPS becomes increasingly monochromatic. There is much to be proud of from Senior Projects the work of MPS and CIS students on campus this year. For the second year, Sean Cooper Carly Cerquone the full freshmen classes from both A Psychophysical Approach to Modeling Green Screen Shootout programs joined to participate in the Temporal Texture in Cinema The purpose of this project is to year-long Freshmen Imaging Project. determine the effectiveness of var- The current cinema industry is on the This year’s team was tasked with build- ious green screen materials (both cusp of innovation, and one of the most ing a transient imaging system capable professional and consumer grade) impactful elements being explored is of capturing light in flight. The team did in multiple difficult-to-composite sit- the deviation from the century-long tra- some amazing work in both engineering uations. In total, five different green dition of 24 fps capture and exhibition. and software development and showed screen materials (three professional the world’s first ever stereo simulation Recent tests into the high frame rate

RIT Imaging Science 2013–2014 Annual Report and 2 consumer), six different lighting scenarios, and three cameras at five quality settings were tested. In the 13 end, it is the hope that this project will be successful in aiding student filmmakers (particularly those in the RIT School of Film and Animation) in their decisions to shoot for visual effects by showing them the pros and cons of the cameras they so often choose, and the materials they should be looking into using. Jordan Westhoff The Cerberus Project As the age of cinema has progressed, technology has become more advanced. Today, systems exist that promise images that are both incredibly large and also exceptionally detailed. Images are now being recorded in 2K, 3K, 4K and 6K resolutions in RAW formats, which has proven to exceed the processing power of the workstations designated to manipulate them. Although processing and GPU power has continued to climb, it is still a major undertaking to import and process the complex images that are generated, often in proprietary format, by each major motion picture camera manufacturer. Executing these workflows has always added an additional level of complexity to the post production workflows for the companies and directors that choose to implement them. The conversions from truly RAW encoded formats to viewable and editable formats in the correct color spaces is a process that is often crip- pling both in complexity and cost. This project attempts to tackle multiple hardware and distributed computing technologies simultaneously with bene- ficial implications for both independent and low-budget filmmakers. The underlying goal of the project aims to provide sound image processing from an imaging science perspective while providing an elegant, scalable hardware solution. This combination builds the foundation of the project and aims to make it mar- ketable for enterprise adaptation. Miscellaneous Awards Elizabeth Pieri • RIT Outstanding Undergraduate Scholar Matt Donato & Victoria Scholl • Society of Motion Pictures and Television Engineers Lou Wolf Scholar ACADEMICS ACADEMICS14

First year graduate students taking a test in one of their core imaging science classes.

RIT Imaging Science 2014–2015 Annual Report ACADEMICS 15 IMAGING SCIENCE GRADUATE PROGRAM Program Coordinator’s Comments By Dr. John Kerekes

The Imaging Science graduate program continues to be highly regarded by our peers and prospective employers. Our graduates remain in high demand and are finding employment in industry, government, nonprofits and universities.

For the second time under the new semester calendar, we took advantage of the extended break in January with faculty and graduate students offering a number of short courses free of charge to our students. These courses ranged from practical introductions to the technological alphabet soup of LaTeX, ACADEMICSMODTRAN, C++, and DIRSIG, to the higher math behind Graph Theory as well as even courses on Entrepreneurship for Scientists and Principles and Tech- niques to Enhance Your Presentations. These sessions were all well attended and enjoyed by many. Also for the second year, we welcomed our new graduate students in the fall with an “Imaging Science Immersion” orientation program. Occurring just prior to the start of classes, we exposed the incoming students to the imaging chain through lectures, demonstrations, and lab tours. The program wrapped up with a social gathering where the new students had an opportunity to get to know the faculty. We look forward to the continued success of our nation’s unique graduate program in Imaging Science. The following is a summary of activities, changes, and student highlights over the 2014-15 academic year. Graduate Program Faculty This was a quiet year of little change in our graduate faculty with the notable The first year graduate students won the exception of the move by Center Director Professor Stefi Baum to a new posi- prestigious 'Sponsor's Award' for their work tion as Dean of the Faculty of Science at the University of Manitoba. Prof. Baum on multispectral crime scene imaging at remains engaged advising our graduate students as a Research Professor here the annual Imagine RIT Innovation and in the Center. Creativity Festival. As of the end of the 2014-15 academic year there are a total of 51 members of the CIS Graduate Program Faculty. Seventeen are tenured, or tenure-track, with the Center as their primary appointment. Another twenty-five have a primary appointment in one of thirteen other departments centers, programs or laboratories with which the Center is affiliated. The Center is the home to nine Research Faculty. There are five Program Allied Faculty who hold positions at other organizations outside of RIT. Curriculum Development With two years now completed under the semester calendar we have just about stopped saying “quarter” when referring to the academic term! Kidding aside, the transition has been smooth and the students seem to be benefiting from the slower pace as well as the additional material included in each course. One minor change to our curriculum was a renaming of the graduate “Digital Image Processing” course to “Image Processing and Computer Vision.” This change reflects the fact that nearly all image processing is digital these days, and that we have been introducing the fall of 2014 included 8 M.S. and Manuscript Cultures, 7, pp. 35-46 16 computer vision topics into the course 17 Ph.D. students. In this group of (February 06, 2015) to expose our students to an increas- students there are two US Air Force of- • Meng, Lingfei; Kerekes, John P., An ingly common application of imaging ficers (1 M.S. and 1 Ph.D.) and one Ca- Analytical Model for Polarimetric Im- science. The course was taught this nadian Forces officer (M.S.). Except for aging Systems, IEEE Transactions on year by Prof. Nathan Cahill, a member the Air Force officer, all Ph.D. students Geoscience and Remote Sensing, 52, of our Graduate Program Faculty from received an assistantship covering full 10, pp. 6615-6626 (October 2014) the School of Mathematical Sciences, tuition and a stipend, while two of the and proved to be very popular. M.S. students also received an assis- • Pahlevan, Nima; Lee, Z; Wei, J; Schaaf, C; Schott, John R.; Berk, New special topics courses were tantship covering tuition and stipend. Ten of the new students were from the A, On-Orbit Radiometric Charac- taught this year by two of the faculty terization of OLI (Landsat-8) for who have recently joined our program. United States, with the balance being international including five from China, Applications in Aquatic Remote Associate Professor Charles “Chip” Sensing, Remote Sensing of Environ- Bachmann taught “Introduction to three from India, two from Nepal, two from Israel and one each from Canada, ment, 154, pp. 272-284 (November Radiative Transfer in Media” in the fall. 30, 2014) This course explored the physics and Iran, and Slovakia. mathematics behind light scattering in Student Awards • Ruane, Garreth J.; Swartzlander, random media and helped prepare the Grover A.; Slussarenko, Sergei; Our graduate students continue to be students for more advanced research Marrucci, Lorenzo; Dennis, Mark recognized through a variety of awards, under the guidance of Prof. Bachmann. R., Nodal areas in coherent beams, scholarships and other forms of recog- Associate Professor Jie Qiao taught Optica, 2, 2, pp. 147-150 (2015) nition. The following is a sampling of “Optical Component, System Design, awards received by imaging science • Ruane, Garreth J.; Watnik, Abbie T.; and Performance Evaluation” in the graduate students during 2014-15. Swartzlander, Grover A., Reduc- spring. This course was centered on ing the risk of laser damage in a the use of the ZEMAX optical modeling • IEEE Western New York Image and focal plane array using linear pu- software and provided the students Signal Processing Workshop Best pil-plane phase elements, Applied practical experience in the design and Remote Sensing Paper: Jie Yang Optics, 54, 2, pp. 210-218 (2015) analysis of optical systems. Both classes were well received by our students. • RIT 2015 Graduate Student Delegate • Uzkent, Burak; Hoffman, Matthew and College of Science Commence- J.; Vodacek, Anthony; Chen, Bin, This was the second year for our new ment Speaker: David Kelbe Feature Matching with an Adaptive yearlong PhD laboratory course led Optical Sensor in a Ground Target • SPIE Travel Scholarship: Garreth by Prof. Roger Dube. Seventeen first- Tracking System, IEEE Sensors Jour- Ruane year PhD students were challenged nal, 15, 1, pp. 510-519 (2015) to design and build a multispectral • Wallonie-Bruxelles International • Vantaram, Sreenath R.; Piramanay- imaging system to document a crime Scholarship for Excellence: agam, Sankaranaryanan; Saber, Eli; scene, with the Imagine RIT festi- Garreth Ruane val held in May as a deadline. The Messinger, David W., Spatial seg- students organized themselves along Student Publications and Presentations mentation of multi/hyperspectral subsystems developing requirements, imagery by fusion of spectral-gra- Imaging science graduate students are dient-textural attributes, SPIE designs, and then building and strongly encouraged and provided op- demonstrating the system, named Journal of Applied Remote Sensing, portunities to broadly disseminate their 9, 1, pp. 1-37 (March 31, 2015) Crime Scene Investigation System research by publishing journal articles, (CRISIS). Their project for the second presenting and publishing at scientific • Zhang, Jiashu; Kerekes, John P., An year in a row won a first place sponsor conferences, and interacting with spon- Adaptive Density-based Model for award at the festival. Congratulations! sors at meetings and workshops. The Extracting Surface Returns from Prof. Dube also continued to refine following is a partial list of publications Photon-counting Laser Altimeter our use of the Matterhorn video authored or coauthored by our graduate Data, IEEE Geoscience and Remote capture system now used to automati- students in 2014-2015. Sensing Letters, 12, 4, pp. 726-730 (April 2015) cally record nearly all graduate course Selected Journal Articles with Gradu- lectures. This system continues to ate Student Authors (student author Selected Conference Proceeding Papers be used not only by our online M.S. underlined) with Graduate Student Authors (student students, but also by our on-campus author underlined) students who have found it useful to • Basnet, Bikash; Vodacek, Anthony, review lectures. Tracking Land Use/Land Cover • Bandyopadhyay Madhurima; van Graduate Student Body Dynamics in Cloud Prone Areas Aardt, Jan; van Leeuwen, Martin, Using Moderate Resolution Satel- Modeling individual trees in an At the beginning of the 2014-15 aca- lite data: A Case Study in Central urban environment using dense dis- demic year there were a total of 117 Africa, Remote Sensing, 7, pp. 6683- crete return LiDAR, Proc. SPIE 9465, graduate students pursuing degrees in 6709 (May 26, 2015) Laser Radar Technology and Applica- Imaging Science. There were 28 resi- tions XX, and Atmospheric Propaga- dent M.S. students, 7 online M.S. stu- • Easton, Jr., Roger L.; Kelbe, David, tion XII, 94650J (May 21, 2015) dents, and 82 Ph.D. students including Statistical Processing of Spectral one who is pursuing the degree online. Imagery to Recover Writings from • Concha, JA; Schott, John R., In-wa- Erased or Damaged Manuscripts, ter component retrieval over Case The incoming class that started in

RIT Imaging Science 2014–2015 Annual Report 2 water using Landsat 8: Initial • Vaidyanathan, Preethi B.; Pelz, Jeff B.; Advisor: Carl Salvaggio results , Geoscience and Remote Alm, Cecilia; Shi, Pengcheng; Haake, • Bin Chen, Multispectral Image Road 17 Sensing Symposium 2014 IEEE Anne, Recurrence quantification Extraction Based Upon Automated International, IGARSS 2014, July 13 analysis reveals eye-movement be- Map Conflation, Advisor: Anthony - 18, 2014, pp. 4458-4461, Quebec, havior differences between experts Vodacek Quebec, Canada (July 13, 2014) and novices, Proceedings of the Symposium on Eye Tracking Research • Monica Cook, Atmospheric Com- • Cui, Zhaoyu; Kerekes, John P.; pensation for a Landsat Land Surface DeAngelis, Chris; Brown, Scott and Applications, Symposium on Eye Tracking Research and Applications Temperature Product, Advisor: D.; Nance, C. Eric, A Comparison John Schott of Real and Simulated Airborne (ETRA), pp. 303-306, Safety Harbor, Hyperspectral Imagery, Western Florida, United States (2015) • Rey Garma, Image Quality Model- New York Image and Signal Process- • Yang, Jie; Kerekes, John P., A Com- ing and Characterization of Nyquist ing Workshop, pp. 19-22, Rochester, bined Approach for Ice Sheet ele- Sampled Framing systems with Op- New York, United States (November vation Extraction From Lidar Point erational Considerations for Remote 07, 2014) Clouds, Western New York Image Sensing, Advisor: John Schott • Dangi Shusil; Ben-Zikri, Kfir; and Signal Processing Workshop, pp. • Shea Hagstrom, Voxel-based LIDAR Schwarz KQ, Cahill Nathan; and 15-18, Rochester, New York, United analysis and Applications, Advisor: Linte, Cristian A., Endocardial States (November 07, 2014) David Messinger left ventricle feature tracking and • Yao, Wei; van Leeuwen, Martin; • David Kelbe, Forest Structure from reconstruction from tri-plane TEE Romanczyk, Paul; Kelbe, David; van Terrestrial Laser Scanning—In Sup- data for computer-assisted image Aardt, Jan, Assessing the Impact of port of Remote Sensing Calibration/ guidance and cardiac function Sub-pixel Vegetation Structure on Validation and Operational Invento- assessment. Proc. SPIE Medical Imaging Spectroscopy Via Simu- ry, Advisor: Jan van Aardt Imaging—Image-guided Procedures, lation, Proc. SPIE 9472, Algorithms Robotic Interventions and Modeling. and Technologies for Multispectral, • Kimberly Kolb, Single Photon Vol. 9415. pp: 941505-1-9. 2015. Hyperspectral, and Ultraspectral Im- Counting Detectors for Low Light agery XXI, 94721K (May 21, 2015) Level Imaging Applications, Advisor: • Dorado-Munoz, Leidy; Messinger, Donald Figer David W., Schrodinger Eigenmaps • Yang, Jie; Kerekes, John P., A Com- for Spectral Target Detection, bined Approach for Ice Sheet ele- • Paul Romanczyk, Extraction of Veg- Algorithms and Technologies for vation Extraction From Lidar Point etation Biophysical Structure from Multispectral, Hyperspectral, and Clouds, Western New York Image Small-footprint Full-waveform Lidar Ultraspectral Imagery XXI, Defense and Signal Processing Workshop, pp. Signals, Advisor: Jan van Aardt Sensing + Security, Algorithms and 15-18, Rochester, New York, United • Katie Salvaggio, A Voxel-Based Ap- Technologies for Multispectral, Hyper- States (November 07, 2014) proach for Imaging Voids in Three-Di- spectral, and Ultraspectral Imagery • Ziemann, Amanda; Messinger, David mensional Point Clouds, Advisor: XXI, 9472, 947211, pp. 1-12, Balti- W., An adaptive locally linear em- Carl Salvaggio more, Maryland, United States (2015) bedding manifold learning approach • Saugata Sinha, Photoacoustic Image • Jin, Can; Bachmann, Charles M., for hyperspectral target detection, Analysis for Cancer Detection and Modeling and Mitigating Noise in Algorithms and Technologies for Multi- Building a Novel Ultrasound Imaging Graph and Manifold Representations spectral, Hyperspectral, and Ultra- System, Advisor: Navalgund Rao of Hyperspectral Imagery, Proc. SPIE spectral Imagery XXI, Defense Sensing • Jiangqin Sun, Temporal Signature 9472, Algorithms and Technologies + Security, Algorithms and Technolo- Modeling and Analysis, Advisor: for Multispectral, Hyperspectral, and gies for Multispectral, Hyperspectral, David Messinger Ultraspectral Imagery XXI, 94720W and Ultraspectral Imagery XXI, 9472, (May 21, 2015) 94720O, pp. 1-15, Baltimore, Mary- • Oesa Weaver, An Analytical Framework for Assessing the Efficacy of Small Travinsky, Anton land, United States (May 21, 2015) • Qiao, Jie; ; Ding, Satellites in Performing Novel Imaging Gaozhan; Dorrer , Christophe, Optical Graduates Missions, Advisor: John Kerekes differentiation wavefront sensor based on binary pixelated trans- During academic year 2014-2015 the • Jiashu Zhang, Analytical Modeling mission filters, Proceedings of SPIE, Center conferred 14 Ph.D. degrees and Performance Assessment of Photonics West, 2015, LASE, 9356, and 13 M.S. degrees. Micropulse Photon-counting Lidar 935608, pp. 1-8, San Francisco, Califor- The following students received a System, Advisor: John Kerekes nia, United States (October 15, 2014) Ph.D. in Imaging Science. • Amanda Ziemann, A Manifold • Sun, Jiangquin; Messinger, David • Madhurima Bandyopadhyay, Quan- Learning Approach to Target W., Streaming analysis of track tifying the Urban Forest Environment Detection in High Resolution Hyper- data from video, Geospatial Infor- Using Dense Discrete Return Lidar spectral Imagery, Advisor: David matics, Fusion, and Motion Video and Aerial Color Imagery for Seg- Messinger Analytics V, Defense Sensing + Secu- mentation and Object-level Biomass The following students received an M.S. rity, Geospatial Informatics, Fusion, Assessment, Advisor: Jan van Aardt and Motion Video Analytics V, 9473, in Imaging Science. 947302, pp. 1-15, Baltimore, Mary- • Tyler Carson, Signature Simulation land, United States (May 21, 2015) and Characterization of Mixed Solids • Viraj Adduru, Ultrasound Guided in the Visible and Thermal Regimes, Robot for Human Liver Biopsy Using High Intensity Focused Ultrasound for of Interplanetary Coronal Mass Ejec- • David Kelbe, Oak Ridge National 18 Hemostasis, Advisor: Maria Helguera tions with Strong Magnetic Fields, Laboratory • Sean Archer, Empirical Measure- Advisor: Roger Dube • Kimberly Kolb, U.S. Army Research ment and Model Validation of • Jordyn Stoddard, Toward Three-Di- Laboratory Infrared Spectra of Contaminated mensional Reconstruction from • Christian Lewis, U.S. Air Force Surfaces, Advisor: John Kerekes Cubesat Imagery: Impacts of Spatial • Matthew Murphy, U.S. Air Force • Kevin Bloechl, A Comparison of Real Resolution and SNR on Point Cloud and simulated Airborne Multisensor Quality, Advisor: David Messinger • Paul Romanczyk, Aerospace Corporation Imagery, Advisor: John Kerekes • Xingchao Yu, Studies of Gas • Saugata Sinha, Visvesvaray Na- • Gregory Fertig, Evaluation of Absorption in Infrared Spectra of tional institute of Technology Carbon-rich AGB Stars, Advisor: MOSFETs for Terahertz Detector • Jordyn Stoddard, U.S. Air Force Arrays, Advisors: Emmett Joel Kastner Ientilucci and Zoran Ninkov • Oesa Weaver, U.S. Air Force The following are post-graduate plans • Jiashu Zhang, Aptina Imaging • Colin Fink, Glint Avoidance and Re- for some of the students who graduated moval in the Maritime Environment, during 2014-2015. • Amanda Ziemann, Los Alamos Advisor: Michael Gartley National Laboratory • Sean Archer, UTC Aerospace Systems • Katherine Grzedzicki, Advisor: Carl Salvaggio • Kevin Bloechl, U.S. Government

• Michael Harris, Supervised Mate- • Bin Chen, OmniVision Technologies rial Classification in Oblique Aerial • Monica Cook, Los Alamos National Imagery Using Gabor Filter Features, Laboratory Advisor: David Messinger • Greg Fertig, U.S. Air Force • Matthew Heimbueger, Advisor: John Schott • Colin Fink, U.S. Air Force • Rey Garma, U.S. Air Force • Christian Lewis, The Development of a Performance Assessment Meth- • Katherine Grzedzicki, National odology for Activity Based Intelli- Geospatial-Intelligence Agency gence: A Study of Spatial, Temporal, • Shea Hagstrom, Johns Hopkins Uni- and Multimodal Considerations, versity Applied Physics Laboratory Advisor: David Messinger • Michael Harris, Exelis • Ming Li, Building Model Reconstruction from Point Clouds Derived from Oblique • Matthew Heimbueger, Laboratory Imagery, Advisor: John Kerekes for Laser Energetics, University of Rochester • Matthew Murphy, Statistical Study

RIT Imaging Science 2014–2015 Annual Report First year CIS graduate student 6. Howett, C. J. A., Spencer, J. R., co-author on a Nature magazine Pearl, J. & Segura, M. High heat 19 cover article flow from Enceladus’ south polar region measured using 10–600 Emily Berkson is co-author cm−1 Cassini/CIRS data. J. Geophys. on article titled “Curtain Res. 116,E03003 (2011) eruptions from Enceladus’ 7. Porco, C. C., DiNino, D. & Nimmo, south-polar terrain” F. How the geysers, tidal stresses, and thermal emission across the south polar terrain of Enceladus are Curtain Eruptions From Enceladus’ related. Astron. J. 148, 45 (2014) South-Polar Terrain 8. Porco, C. C. et al. Cassini imaging Joseph N. Spitale, Terry A. Hurford, science: instrument characteristics Alyssa R. Rhoden, Emily E. Berkson & and anticipated scientific investiga- Symeon S. Platts tions at Saturn. Space Sci. Rev. 115, Nature 521,57–60 (07 May 2015), 363–497 (2004) doi:10.1038/nature14368 9. Roatsch, T. J. et al. in Saturn from Received 05 August 2014, Accepted Cassini-Huygens (eds Dougherty, M. 27 February 2015, Published online Esposito, L. &Krimigis, S.) 763–781 07 May 2015 (Springer, 2009).

Abstract: Observations of the south pole of the Left-hand column, reference images are Saturnian moon Enceladus revealed shown with no overlays. Middle column, large rifts in the south-polar terrain, simulated vertical collimated curtains are informally called ‘tiger stripes’, overlain. Right-hand column, simulated named Alexandria, Baghdad, Cairo curtains with the indicated spreading and and Damascus Sulci. These fractures zenith angles are overlain. On Baghdad and have been shown to be the sources Damascus Sulci, the shape of the shadows of the observed jets of water vapour are more consistent with a spreading curtain and icy particles1, 2, 3, 4 and to ex- than a collimated curtain. In every case, the hibit higher temperatures than the shape of the observed curtain is consistent surrounding terrain5, 6. Subsequent with a spreading with altitude. On Damas- observations have focused on obtain- cus Sulcus, the horizontal shadow right of ing close-up imaging of this region to centre is consistent with a fracture not seen better characterize these emissions. in the base map, but which appears in a Recent work7 examined those newer different Cassini image. data sets and used triangulation of discrete jets3 to produce maps of References: jetting activity at various times. Here 1. Porco, C. C. et al. Cassini observes we show that much of the eruptive the active south pole of Enceladus. activity can be explained by broad, Science 311,1393–1401 (2006) curtain-like eruptions. Optical illu- 2. Hansen, C. J. et al. Enceladus’ water sions in the curtain eruptions result- vapor plume. Science 311, 1422– ing from a combination of viewing 1425 (2006) direction and local fracture geometry 3. Spitale, J. N. & Porco, C. C. Asso- produce image features that were ciation of the jets of Enceladus probably misinterpreted previously with the warmest regions on its as discrete jets. We present maps of south-polar fractures. Nature 449, the total emission along the fractures, 695–697 (2007) rather than just the jet-like compo- 4. Hansen, C. J. et al. Water vapour nent, for five times during an approx- jets inside the plume of gas leaving imately one-year period in 2009 and Enceladus. Nature456, 477–479 2010. An accurate picture of the style, (2008) timing and spatial distribution of the 5. Spencer, J. R. et al. Cassini encoun- south-polar eruptions is crucial to ters Enceladus: background and the evaluating theories for the mecha- discovery of a south polar hot spot. nism controlling the eruptions. Science 311, 1401–1405 (2006) RESEARCH 20RESEARCH

RIT Imaging Science 2014–2015 Annual Report RESEARCH 21 DIGITAL IMAGING AND REMOTE SENSING LAB Laboratory Director’s Comments By Dr. Jan VanAardt

A Year of Exciting Research: Overview The 2014-2015 academic year was full of exciting developments and activities for the faculty, staff, and students of the Digital Imaging and Remote Sensing Laboratory (DIRS) in the Chester F. Carlson Center for Imaging Science...

DIRS secured significant funding from especially NASA, but also from other agencies and institutions to further remote sensing science via our “imaging chain” approach, i.e., projects that encompass systems, algorithms, and applications. DIRS activities spanned a range of notable activities; building upon ongoing RESEARCHresearch in understanding the radiometric behavior of materials through bidirectional reflection distribution function (BRDF) measurements, significant support to the Landsat satellite program for NASA and USGS, and detailed system modeling with our Digital Imaging and Remote Sensing Image Generation (DIRSIG) model plus more. New avenues of research have begun to open looking at imaging applications in the exciting arena of Unmanned Aerial Systems (UAS). Wiedman Professor in Imaging Science, Dr. Chip Bachmann, in his second year at DIRS has gone from strength-to-strength in bi-directional reflectance distribution function (BRDF) modeling projects. Dr. Bachmann again exposed students to exciting field trips while deploying the GRIT (Goniometer at RIT)—more on that later—and is developing the second generation GRIT, which will include significant refinements in terms of processing power, data acquisition, and portability. A new addition to our team, Dr. Matt Montanaro (Imaging Science alum previously from NASA Goddard) together with Dr. Aaron Gerace, furthered the NASA Landsat and related mission support and calibration work started by Dr. John Schott. Dr. Schott is transitioning to retirement status, but still maintains an active role in the group, thus ensuring continuity in project and expertise related to especially the Landsat project legacy. Much of this work is underscored by a stalwart, ongoing project in DIRS, namely the Digital Imaging and Remote Sensing Image Generation (DIRSIG) tool under the leadership of Dr. Scott Brown. As has been the case since the inception of DIRS, our students continue to be highly sought after by government and industry alike, with recent MS and PhD alumni obtaining employment in organizations such as the National Geospa- tial-Intelligence Agency, the National Reconnaissance Office, Exelis, Oak Ridge National Laboratory, Los Alamos National Laboratory, Johns Hopkins Applied Physics Lab, Aerospace Corporation, etc. Discussions also are ongoing with the National Geospatial Intelligence Agency (NGA) and Department of Homeland Security (DHS) to further research collaboration between the various groups and employment opportunities at such institutions. Finally, a variety of DIRS graduate students received prestigious recognition or awards for their research or graduate outputs while at RIT. Dave Kelbe (PhD) was selected as the graduate student commencement speaker for the College of Science, while Jie Yang (PhD) was awarded the Best Remote Sensing Paper during the IEEE Western New York Image and Signal Processing Workshop. Our faculty and staff did their part in maintaining a high profile locally and in- ternationally. Dr. Emmett Ientilucci served as chair for the local IEEE Geoscience and Remote Sensing Society, Dr. John Kerekes served as ad-com member for the same society, but at the international record and relate geophysical measurements level, and a variety of faculty and staff to GRIT BRDF. (Left) The NASA Goddard G-Li- 22 served on review panels and science HT sensor overflies the RIT GRIT while GRIT teams for the NSF, IEEE, NASA, etc. scans a white Spectralon reference panel. Especially notable is that Dr. Matt Fixing a Stray Light Problem on the Montanaro, Ms. Nina Raqueño, Dr. Landsat 8 Thermal Imager John Schott, and Dr. Aaron Gerace won the "Robert H. Goddard Award The Thermal Infrared Sensor (TIRS) on for Exceptional Achievement for a board Landsat-8 suffers from a stray Science Team" from NASA for their light problem that adversely affects contributions to the Landsat Calibra- Figure DIRS1: DIRSIG simulation of a power sta- image products from the instrument. tion/Validation Team, thus flying the tion scene as viewed by three different sensors Work performed at RIT (through a DIRS flag as far as Landsat calibration NASA grant) has led to the devel- activities are concerned. Dr. Schott GRIT: Measuring Light Reflecting Proper- opment of a stray light correction also received the John R. Pecora Team ties of Materials algorithm that drastically reduces the Award from the American Society for The GRIT Laboratory, led by Dr. Bach- stray light artifacts in the final image Photogrammetric Engineering and mann, has had a busy year. In Novem- products (see Figure DIRS3). The Remote Sensing (ASPRS) in November ber 2014, Dr. Bachmann presented algorithm is a result of a major imag- 2014, while Dr. Anthony Vodacek was a keynote address entitled “Hyper- ing science effort by RIT personnel to elected as a Senior Member of the spectral Remote Sensing and Char- understand the physics and geometry IEEE. And then of course, the science acterization of the Coastal Zone,” at of the problem and to derive a prac- itself—by last count and inclusive of the University of Sydney’s Institute of tical solution to be used to correct summer 2014-May 2015—DIRS has Photonics and Optical Science IPOS’14 Landsat images operationally. The RIT published 15 peer-reviewed journal Conference, and an invited talk in Sep- algorithm has been implemented into articles, 25 conference proceedings tember on “Coastal Characaterization the official United States Geological papers, and four peer-reviewed con- from Remote Sensing” at the Universi- Survey (USGS) Landsat ground system ference proceedings papers... not a ty of Houston’s NSF National Center for for testing and will be made opera- bad tally for an active group. Airborne Laser Mapping. tional shortly. An example of stray light correction is shown in Figure The following specifically summarizes In March 2015, Dr. Bachmann and his DIRS4. The details of the algorithm some of key research activities of DIRS students (graduate students Justin have been published in the following Harms and Brittany Ambeau, seniors DIRSIG: Simulating Scenes and Sensors journal article, “Toward an operational Doug Peck and Malachi Schultz, and Dr. stray light correction for the Landsat 8 for Remote Sensing Applications Carl Salvaggio’s graduate student Tyler Thermal Infrared Sensor”, which was DIRSIG is a first-principles, phys- Carson) participated in a multi-institu- published in Applied Optics (Vol. 54, ics-based simulation environment tion calibration and validation experi- pages 3963-3978) by authors Montan- that is used by government agencies, ment, led by South Dakota State Univer- aro, Gerace, and Rohrbach. academic institutions, and companies sity (SDSU) and NASA Goddard in the alike for system development and as- Algodones Dunes in southern California sessment, algorithm development and (Figure DIRS2). Dr. Bachmann and his refinement, and in general, furthering students fielded GRIT, a hyperspectral remote sensing as a science. Dr. Scott goniometer system developed in his Brown and Dr. Adam Goodenough are laboratory, to record calibration and the main developers of DIRSIG and validation data. GRIT measurements continuously are immersed in improv- support the development of improved ing on an already highly-valued tool methods to retrieve geophysical param- Figure DIRS3: special data collection from in the remote sensing community. eters which are important to remote the TIRS instrument in which the moon The National Reconnaissance Office sensing; specifically, observed angular was scanned in the out-of-field area of the (NRO) is supporting the development dependence of spectral signatures or instrument and stray light (ghost) signals were of a next generation DIRSIG (DIRSIG bi-directional reflectance distribution detected on the focal plane arrays. [Image Version 5), which will boast refined function (BRDF). credit: Montanaro, M., Gerace, A., Lunsford, A., radiometry, an improved user inter- & Reuter, D. Stray Light Artifacts in Imagery face, and optimized implementation from the Landsat 8 Thermal Infrared Sensor. protocols. This is an exciting develop- Remote Sensing, 6, 10435-10456 (2014).] ment for us as group, especially given the multitude of graduate student RIT researchers continue to investigate projects and related science initia- this approach as well as others to gain tives that benefit from DIRSIG. Figure a better understanding of the sourc- DIRS1 shows an example DIRSIG es of stray light in TIRS and further rendering of a power plant as viewed refine the stray light correction. These by a panchromatic visible (Pan/VIS), Figure DIRS2. RIT CIS team which participat- investigations address tradeoffs in Near Infrared (Near -IR) , and Long- ed in the NASA Algodones Dunes experiment radiometric accuracy and operational wave Infrared (LWIR) sensors. in March 2015. (Right) CIS mobile lab provid- complexity. One approach considers ed a base of operations for the hyperspectral the use of near contemporaneous data GRIT and geotechnical instruments used to from other satellites such as NOAA’s GOES or NASA’s MODIS to provide high

RIT Imaging Science 2013–2014 Annual Report fidelity estimates of stray light contri- ed this allows us to look at not only the butions to the TIRS signal. state of the planet but also to study processes taking place over time. We 23 don’t have to start a study today and wait 20 or 30 years for the study site to evolve. Rather, we can look back in time and look at what the impact of various forcing functions has been. Figure DIRS5: Array of buoys used for thermal A fundamental critical enabler for calibration of Landsat these types of study is a consistent cal- ibrated data set. The Digital Imaging Measuring Land Surface Temperature Figure DIRS4: Early example of TIRS imagery and Remote Sensing (DIRS) laboratory from Space with RIT’s stray light correction, original (L), at the Rochester Institute of Technolo- corrected (R) . [Image credit: Montanaro, M., gy (RIT) has been helping to calibrate Land Surface Temperature (LST), in a Gerace, A., Lunsford, A., & Reuter, D. Stray Landsat Instruments since the 1980’s. simple sense, is how hot the ground Light Artifacts in Imagery from the Landsat 8 Over that period RIT’s vicarious feels to the touch. People involved in Thermal Infrared Sensor. Remote Sensing, 6, thermal infrared calibration methods weather prediction, climate research, 10435-10456 (2014).] have evolved to provide more accurate or even agriculture could benefit from and less field intensive approaches. a product or archive that provides the Using DIRSIG to better understand the Most recently this has focused on LST at any point in the world at any NASA SOLARIS sensor platform the development of semiautonomous time, dating back as far as possible. NASA Goddard’s SOLARIS (Solar, approaches to thermal calibration that Currently, the LST process has been Lunar for Absolute Reflectance Imag- utilize moored ocean buoys coupled to validated for Landsat 5 for North Amer- ing Spectroradiometer) sensor is the in-water thermodynamic and atmo- ican sites, with a mean error of -0.267 calibration demonstration system for spheric radiative transfer models to Kelvin for cloud free scenes. Validation CLARREO (Climate Absolute Radi- produce top of the atmosphere (TOA) on a global scale and for more recent ance and Refractivity Observatory), sensor reaching radiance estimates. Landsat instruments (7 and 8) are un- a mission that addresses the need to These values can then be compared derway. An example scenario is shown make highly accurate observations to satellite observed radiance values in Figure DIRS6 for a buoy on Lake of long-term climate change trends. and used to evaluate and, if necessary, Ontario. Another goal is to be able The SOLARIS instrument will be correct the instrument calibrations. to provide a confidence metric that designed to support a primary objec- RIT derived calibration approaches will indicate how trustworthy the LST tive of CLARREO, which is to advance have been used to support calibration results are, which will mostly depend the accuracy of absolute calibration of Landsats 4, 5, 7 and 8. upon cloud cover because it often for space borne instruments in the This task continues research into the causes Landsat to predict erroneously reflected solar wavelengths. Dr. Aaron automation of thermal band calibra- low temperatures. Once all these steps Gerace’s work focuses on the devel- tion for the Landsat archive and the are completed and the process contin- opment of a simulated environment operational satellites. It focuses on ues to show low error levels, it will be to facilitate sensor trade studies to increasing the number of NOAA buoys combined with an emissivity product support instrument design and build used in the process, the stability of a produced by NASA JPL so that LST of the SOLARIS sensor. Details of the calibration analysis tool, and appli- maps for every Landsat scene can be results to date have been published in cation of the analysis tool to monitor generated and accessed by the public. the following conference proceedings data from the buoys. paper, “The development of a DIRSIG For more than 15 years RIT has provid- simulation environment to support ined the NASA/USGS Landsat Calibration strument trade studies for the SOLARIS Team with critical data on the their sensor” (Proceedings of SPIE 9472, thermal instrument performance. This Algorithms and Technologies for Multi- year RIT has continued to monitor spectral, Hyperspectral, and Ultraspec- the performance of the Landsat 7 and tral Imagery XXI, May 21, 2015), by 8 instruments using an automated authors Gerace, Goodenough, Montan- buoy calibration process that utilizes Figure DIRS6. An example of a Landsat image aro, Yang, McCorkel, and Ong. NOAA’s existing buoys for water tem- of Lake Ontario with a buoy in the scene (the Calibration and Assessment of the perature truth. The extensive array of red triangle indicates the buoy location). The temperature at the buoy location is estimated Landsat Sensors Thermal Calibration buoys used for calibration is illustrat- ed in Figure DIRS5. using the LST process and is compared to the Automation buoy measurement that was taken at the time The Landsat program represents the Throughout it’s 16 year lifespan, of acquisition. longest continuous record of obser- Landsat 7 thermal band has proven vations of the Earth from space. As a to be a reliable and extremely stable Retrieval of color producing agents in result the Landsat data represents a instrument. The average error between Case 2 waters using Landsat 8 buoy truth and Landsat predicted is unique ongoing opportunity to study This project, sponsored by USGS, takes both the current conditions and the 0.1 Kelvin with a standard deviation of 0.4 Kelvin. advantage of Landsat 8's features evolution of the Earth over Landsat’s for the retrieval of color producing 42 years history. If effectively calibrat- agents (CPAs: chlorophyll-a, colored dissolved organic matter (CDOM) and number of issues still affect the analy- atmospheric, orbit and acquisition ge- sediments (TSS)) over inland and sis of multi-temporal data sets. These ometry effects, as well as, a wide range 24 coastal waters. Landsat 8 provides an are primarily due to the process of of detector element specific variations improved signal-to-noise ratio (SNR) imaging from multiple sensors. Some associated with actual sensor systems and a new spectral coastal aerosol of these undesired or uncompensated (e.g. spectral, gain, and bias variation band in the blue. A look-up-table (LUT) sources of variation includes variation and drift and noise) and spectrum-matching methodology in view angle, variation in the illu- This task includes a wide range of was implemented to simultaneously mination angle, atmospheric effects, land cover types and puts emphasis retrieve CPAs, taking advantage of sensor effects such as Relative Spec- of the impact of spectral and BRDF Landsat 8's new features. A LUT of tral Response (RSR) (for more than variability on analysis of data from a spectral remote-sensing reflectances one sensor). The complex interaction Landsat-Sentinel II Constellation. To (Rrs) with different concentration of of these sources would make this type accomplish this, Hyperion spectral CPAs was produced using the in-water of study extremely difficult with the data is merged with MODIS BRDF data radiative transfer model Hydrolight. real data. The approach is to build a and topographic data to generate A model-based empirical line method synthetic scene (forest canopy) using DIRSIG scenes. The high resolution (MoB-ELM) algorithm was developed DIRSIG model. The anisotropic reflec- spatial data (e.g. DigitalGlobe) will be to atmospherically correct the Landsat tance characteristics of forest canopy used for land cover classification and 8 imagery and allow direct comparison can be defined by its bi-directional texture mapping, while BRDF model with the LUT of Rrs. The retrieval algo- reflectance distribution function. Us- coefficients derived from MODIS BRDF rithm was applied over two Landsat 8 ing DIRSIG, the forest canopy's BRDF product will be used in combination scenes and shows a root mean squared can be measured and modeled using with hyperspectral data for estimat- error (RMSE) as a percentage of range existing canopy reflectance models. ing spectral BRDF model coefficients. of about 10% for Chlorophyll-a and An example of forest scene modeled Once a scene is assembled, Sentinel total suspended solid (TSS), and about in DIRSIG is shown in Figure DIRS8. II and Landsat 7 & 8 sensors can be 5% for colored dissolved organic To understand the significance of the “flown” over the scene and simulat- matter (CDOM) when compared with sources of variation and their relative ed data with varying illumination, ground-truth data. The CPA concentra- magnitude, a factorial design exper- atmospheric and view geometries can tion maps exhibit expected trends of iment will be designed using DIRSIG be acquired. These data can then be low concentrations in clear water and as the experimental engine. Using the used to develop and test algorithms higher concentrations in turbid water. response surface methodology, the designed to normalize/remove the Figure DIRS7 shows a concentration relative sensitivity of the observed differences due to sensor view and map for TSS over the Rochester Embay- reflectance to the sources of variation spectral response differences. Figure ment for the Landsat 8 image acquired will be analyzed. DIRS9 illustrates the the synthetic im- on 09-29-2013. Red areas indicate age process using different datasets areas of elevated TSS concentrations (DEM, texture image, hyperspectral in ponds relative to the open waters of data, MODIS BRDF product, classifica- Lake Ontario. tion map).

Figure DIRS8: Synthetic image of trees modeled DIRSIG as viewed by a sensor from 12 degree zenith when sun was at 23 degree Figure DIRS7: concentration map for TSS over zenith with approximate ground resolution the Rochester Embayment for the Landsat 8 of 10cm. image acquired on 09-29-2013 Sentinel II + Landsat Data Modeling Evaluation of multi-sensor constellations RIT is currently studying the impact Figure DIRS9: The synthetic image generation for long term resource monitoring sensor/acquisition differences be- process using different datasets Moderate resolution remote sensing tween Landsat and the European Space data offers the potential to monitor the Agency’s Sentinel II would have on Next Generation Land Remote Sensing long and short-term trends in the con- analysis of data from such a constel- Systems Engineering Support lation. The current work focuses on dition of the Earth's resources at finer This project assists the Landsat detailed modeling of the spectral and spatial scales and over longer time pe- program office with the modeling and bidirectional reflectance properties riods. With improved calibrations (ra- analysis of various concepts for Land of forest canopies. The modeling of diometric and geometric), free access Imaging using the DIRSIG simulation the canopy and the sensors (including (Landsat, Sentinel, CBERS), and high- tool. The goal is to bridge the gap spatial, spectral and radiometric proper-level products in reflectance units between instrument/satellite con- erties) is performed with the Digital makes it easier for the science commu- cepts and understanding their efficacy Imaging and Remote Sensing Image nity to derive biophysical parameters for various applications. This systems Generation (DIRSIG) model. The model from these remotely sensed data. In engineering support will aid in the allows the inclusion of illumination, spite of all these improvements, a development of system requirements,

RIT Imaging Science 2014–2015 Annual Report analysis of various concepts, and evaluation of issues during hardware development. Figure DIRS10 shows a 25 DIRSIG simulation of Lake Tahoe with additional target patterns inserted to evaluate the effects of design parameters of future Landsat sensors. Figure DIRS12. Examples of our 3D structural assessment work in support of NEON Figure DIRS11. Landsat-8 false color images activities. (L) a TLS point cloud is used by of an irrigation project in western Rwanda Dave Kelbe (PhD) to extract tree stems, during terrace construction. Black lines after which multiple scans are registered to are GPS generated ground truth for project each other based on stem tie points. This boundaries obtained from a project manager. research will aid in forest inventory, calibra- The left image was obtained 13 June 2013 and tion-validation of airborne remote sensing the right image was obtained 1 August 2014. assessments, and to develop virtual scenes, Dark green is trees, light green is crops, and based on real scans, for use in DIRSIG simu- Figure DIRS10: Synthetic DIRSIG image of magenta is soil or other bare land including lations. (R) Paul Romanczyk, a PhD student, Lake Tahoe with patterned targets inserted to bare soil near concentrated dwellings. The is using DIRSIG to better understand the study future Landsat design performance progression of the construction of the terraces structural complexity that can be assessed can be found by comparing the two images. via waveform lidar signals. Out of Africa… using time series of Trees in 3D—our involvement with the pan-sharpened Landsat data for agricul- National Ecological Observatory Network Continuing the DIRS legacy in remote tural development projects in Rwanda (NEON) sensing for improved disaster response One of the oldest application areas Dr. Jan van Aardt and a brain trust of In the 2009-2010 timeframe, DIRS of remote sensing is agricultural graduate students have been exten- developed a NSF Partnerships for monitoring. Estimating crop yields sively involved with NEON in building Innovation (PFI) project with the is often the goal that is achieved data processing chains, investigating intent to bridge the gap between by combining time series data of structural algorithms, and developing disaster response practitioners and remotely sensed vegetation health, calibration-validation tools for NEON’s researchers. RIT was able to respond rainfall, temperatures, etc. in predic- Airborne Observation Platform (AOP) to the 2010 Haiti earthquake based tive models. However, yield mod- waveform light detection and ranging on networks, skill sets, and imaging eling for the futures market is very (lidar) instrument. To this end the group science expertise developed during difficult in many tropical developing has published numerous conference the grant. This PFI project has led to countries with small plot sizes, hilly and peer-reviewed papers, have re- approximately $1.7m of directly- and terrain, seasonal cloudiness, a wide ceived NEON funding for two post-doc- indirectly related spin-off projects, variety of crops, and little access toral researchers at RIT to spearhead among which a NSF Science Master to ancillary data. Here, the goal of this effort, and have three PhD current/ Program (SMP), a Google grant, and development impact monitoring is past projects and two past MS projects a Department of Homeland Security to quantify the impact of the money aligned to the work, based on comple- (DHS) award, among others. The latest spent on development projects for mentary NSF and NASA funding. Past example is our collaboration with understanding the highest impact students William Wu (PhD; processing Rensselaer Polytechnic Institute (RPI; interventions. In Rwanda, Dr. Anthony chain development for waveform lidar; Dr. José Holguin-Veras), where Dr. Vodacek has been collaborating with now with Apple Inc.), Joe McGlinchy Jan van Aardt’s students are working a World Bank group to assess the (MS; extracting structural vegetation on remote sensing assessments of use of remote sensing in monitoring components from waveform lidar; now disaster-impacted infrastructure, and agricultural development projects. with ESRI), and Diane Sarrazin (MS; RPI is using our inputs for trafficability PhD student Bikash Basnet has been species classification based on fusion modeling—the goal is improved access using pan-sharpened Landsat-8 of imaging spectroscopy and lidar data; restoration, or in other words, en- data to observe change at one of the now with Canadian Air Force) have laid abling emergency responders to reach World Bank supported projects. Some the groundwork for our NEON-related people in need via either existing example images of the project site research. This work has been solidified routing options, or by opening new demonstrate the potential for mon- by efforts from Dr. Kerry Cawse-Nichol- routes via optimized selection. PhD itoring these sites (Figure DIRS11). son (post-doctoral researcher; under- student Colin Axel is using high spatial Another World Bank supported proj- standing how waveform lidar signals resolution imagery and discrete return ect in Rwanda is the building of rural propagate through the canopy), Dr. lidar data to automatically extract feeder roads to provide farmers better Martin van Leeuwen (post-doctoral roadways, assess their condition, map access to markets. MS student Donath researcher; extracting leaf properties and quantify debris, and also assess Uwanyirigira, who is from Rwanda, from waveform lidar), Dave Kelbe (PhD; building damage (see Figure DIRS13). will investigate this application of terrestrial lidar algorithms for forest The output maps then serve as inputs pan-sharpened Landsat data for his structure), and Paul Romanczyk (PhD; to RPI’s network modeling research. MS research. simulation of waveform lidar towards This applied research effort has led measuring forest canopy structure). Ex- to multiple research, application, and amples of this structural of 3D research training opportunities. are shown in Figure DIRS12. 26

Figure DIRS15: Plots of the surface-leaving Figure DIRS14: Schematic rendering of three total flux density for a bog (left) and upland dimensional building models generated from (right) site at SNWR. The forested plot had imagery data. twice peak flux density and about 3 times the Figure DIRS13. Examples of disaster total energy compared to the peat bog area. response products that Colin Axel (PhD stu- Fuel Consumption and Carbon cycling This is to be expected considering the fuel dent) is developing using the fusion of high in northern peatland ecosystems: Un- loads and moisture in the different areas of spatial resolution color imagery and discrete derstanding vulnerability to burning, the fireground. lidar sensing. (Top) A building damage as- fuel consumption, and emissions sessment map, developed using lidar metrics Benchmarking LiDAR Fuel Load Esti- Peatland ecosystems represent 3-5% mates, Heat Release and Ground-Mea- 3D Building Model Extraction from of the land surface, but sequester 12- sured Fuel Loads Imagery 30% of soil organic carbon. Current A general and systematic approach of Building on past innovations from trends and climate models predict a evaluating fuel treatment effectiveness the DIRS lab, a current effort is general pattern of decreased water is still to be built and applied as a underway to improve and streamline availability as a likely outcome of guideline for the services in charge of the extraction of three dimensional climate change. These climate chang- conducting these activities in the field. building geometries from a suite of es can lead to increased amplitude The project will addresses a series of different co-registered imaging data of water table variation, including general questions: sources, such as optical, LiDAR, or large mid-summer declines in water graphical maps. These disparate data table height. Although peatlands a. What changes do fuels manage- sources will be displayed in an im- have conventionally been considered ment treatments cause to three mersive virtual reality environment, resistant to wildfire due to their dimensional structure and loading? allowing users to navigate, explore, relatively wet soil conditions, recent studies have shown that the extent b. What defines effective fuel reduc- and even query data on individual tion with regard to fire intensity? buildings on an intuitive and ac- of wildfires in boreal North America, cessible platform like Google Earth. where peat fuels are common, has c. What techniques or methods The ability to co-register different been steadily increasing in recent effectively cause a change in fire imagery sources allows a user to decades. Our role in this research behavior that would increase forest toggle between thermal and optical was to quantify the radiant and con- safety in the wildland urban inter- displays, for example, in order to ex- vective heat release from these fires face using a minimum of resources? tract as much information as possible and to obtain an estimate of combus- RIT’s mission in this 3 year experiment from a scene. The building models tion efficiency by measuring the ratio was to: created with this technique can also of CO2/CO in the fire. provide the necessary input data for o Collect and analyze airborne The last of the prescribed fires infrared and visible imagery using DIRSIG, a program developed here at associated with this project were RIT for generating simulated scenes the WASP data collection system, conducted in the upper peninsula of provide airborne LIDAR before and using different modalities, lighting, Michigan at the 90,000 acre Seney or atmospheric conditions. after the fire to assess changes in National Wildlife Refuge (SNWR). forest structure, fuel loading and This technology promises to aid both Six to ten instrument packages, built arrangement the intelligence and disaster commu- entirely at RIT by undergraduate nities. For example, individual build- students, were deployed on the fires o Measure the radiant and convec- ing characteristics, such as structure mounted on 5.5m tall aluminum tive energy flows at many (~12) elevation and construction material, masts. We are still correlating the locations in the fire ground can be automatically assessed and results of these experiments with o Measure the vertical temperature collected across an entire region. other data inputs to determine the profile in active fire to better These data can be used to improve relationship between fuel moisture, define the convective flows and predictive catastrophe models and sub fuel moisture, fire effects and coupling to fire-atmosphere aid post-disaster response efforts. fire penetration in the boggy areas weather models Figure DIRS14 shows a schemat- and uplands. Figure DIRS15 shows some initial radiant flux density data o Install the equipment in the ic rendering of three-dimensional fireground. building models generated from obtained at two widely different veg- imagery data. Note that the scene etation and fuel loading sites within o Provide ground support during excludes terrain and foliage. The or- the fireground. the fire as part of the ignition ange highlighted structures demon- and control team. strate the capability to automatically The ‘Ex3’ fire this year was planned identify building characteristics. to be of moderate intensity and all the instruments and airborne systems performed well. Additionally,

RIT Imaging Science 2014–2015 Annual Report we constructed and deployed 12 2014–2015 period. There are many fire-triggered video cameras. These exciting developments, ranging from cameras were used by other research- Unmanned Aerial Systems (UAS) and 27 ers to record ember distribution and DIRSIG V.5 activities, to projects spe- pre-frontal ignition, fire spread rate cific to each individual investigator. and behavior, and to observe the rising We look forward to collaborating with- smoke plume. Several mosaic frames in the Center and College of Science, of infrared imagery are shown in across RIT colleges, and with external Figure DIRS16. This is the last planned researchers towards advancing the deployment of the WASP camera sys- field of remote sensing, from sensors… tem, having successfully collected data to algorithms… to applications! for 31 fires. We believe that we have the largest collection of highly time-resolved high spatial resolution infrared imagery in the world, and we plan to exploit this data set in the future.

Figure DIRS16—Two long-wave infrared images of the Ex3 fire taken with the WASP instrument. [L] is early in the fire with active fire (brighter red/white) while [R] is a creep- ing ground fire that resulted from ignitions on several sides of the fire control line.

Unmanned Aerial Systems DIRS is in the forefront of the emerging application area of Unmanned Aerial Systems (UAS). We are witnessing an explosive growth in the use of UAS for personal and commercial applications driven by a confluence of technical capabilities and market forces that rivals the early days of the personal computer and cellular phone. In addition to the technology required for UAS flight operations, imaging systems and information derived from imagery are the most prominent enablers to applications using UAS platforms. Practically every UAS produced carries an imager of some kind for some purpose. DIRS is currently engaged in research for using UAS imaging for a variety of applications including determining the characteristics of construction materials and precision agriculture. For sure there will be more to report on this in next year’s report. In Conclusion Hopefully this summary provided you as reader with a good overview of what we have been up to as faculty, staff, and students within the Digital Imaging and Remote Sensing Labo- ratory here in the Chester F. Carlson Center for Imaging Science during the RESEARCH 28RESEARCH

Professor Zoran Ninkov and his students in their lab.

RIT Imaging Science 2014–2015 Annual Report RESEARCH 29 LABORATORY FOR ADVANCED INSTRUMENTATION RESEARCH (LAIR) Laboratory Director’s Comments By Dr. Zoran Ninkov

The Laboratory for Advanced Instrumentation Research is dedicated to; (a) the development of novel and innovative instruments for gathering data from a wide variety of physical phenomena (b) the training of the next generation of instrument scientists who will occupy positions in government, industry and academia.

LAIR utilizes the excellent infrastructure facilities available at RIT including the Semiconductor and Microsystems Fabrication Laboratory, the Center for Electronics Manufacturing and Assembly, and the Center for Detectors. A wide variety of instruments have been developed at RIT over the last twenty RESEARCHyears including digital radiography systems, liquid crystal filter based imaging systems for airborne (UAV) mine detection, a speckle imaging camera for the WIYN 3.6 meter telescope, a MEMS digital micromirror based multi-object spectrometer, and an X-ray imaging systems for laser fusion research. This research has been funded by NASA, the NSF, NYSTAR and a variety of corpora- tions such as Exelis, ITT, Kodak, Moxtek and ThermoFisher Scientific. Graduate Students 2014-15: Dmitry Vorobiev Greg Fertig Kevan Donlan Chao Zhang Katie Seery Sahil Pravin Bhandari Ross Robinson Bryan Fodness Kyle Ryan Jack Horowitz Anton Travinsky Undergraduate Students 2014-15: Ryan LaClair Lauren Morehouse (REU) Descriptions of some of the current research projects are listed below: 1. Studies of the optical properties of TI DMDs and the development of a multi-object spectrometer The Digital Micromirror Device (DMD) built by Texas Instruments is the device used as the optical slit mask in the RITMOS Multi-Object Spectrometer. RITMOS was designed to record the spectra of multiple stars within the field of view. The instrument has been improved, with newly written software and a new imaging camera. The 2010 Astronomy Decadal Survey's leading suggestion for space instrumentation is a wide field IR Space Telescope which will require a multi-object spectrograph to accomplish its science goals. Other space based missions requiring multi-object spectroscopy capability have been proposed, including for the ultraviolet. There have been four key aspects of the performance of DMDs that have been questioned for use in a MOS for space. We have attempted to address each of these. (1) To assess the light scattering prop- wavelengths that the device can be used 3. The effect of IPC on Astronomical 30 erties of DMDs. a spot scanning system for. We are currently working on removing Imaging Systems has been assembled that accurately these windows and repacking the devices The effect of interpixel capacitance translates a spot of light across the with windows that are transmissive in the (IPC) on images captured by infrared DMD and measures the scattered light ultraviolet. Initially we are using magne- sensors was first identified by a PhD across the mirror, at the central via, and sium fluoride and HEM Sapphire as the student at RIT, Drew Moore. Now that at the edges of the individual mirrors. replacement window material A DMD with this effect has been characterized, part of the protective package removed is research has focused on investigating shown below. The window material on the how IPC affects photometry. IPC acts right is kapton and that on the left is a thin as a smoothing filter, by spreading commercial pellicle. out the signal of each pixel into the neighboring pixels and also affects the normal assumptions about the relationship between noise and signal. Astronomers commonly use a method of photometry called aperture photometry which is compromised (2) For use in the infrared it is required 2. Enhancing Focal Plane Array Quantum by IPC effects. For isolated stars the that DMDs operate at cooled tempera- Efficiency with Quantum Dots effect is small. Continuing research tures. The test configuration seen below will explorer IPC effects on diffraction There are many interesting things to see limited imagery, such as on the James in the laboratory at RIT showed that nor- in the ultraviolet (UV). Lithography for mal operation of these devices was able Webb Space Telescope, as well as in integrated circuit production is exposed crowded fields. to carried out to a temperature of 130K. with 193nm light, honey bees' view of This was the limit of how cold the DMD flowers include the UV region and analytical could be cooled by the configuration and instruments use UV emissions to identify did not reflect a failure on the DMD. materials. Current silicon CMOS or CCD based detectors used in standard digital cameras do a poor job of recording UV images. The ability to detect UV light may be improved by switching to exotic materials or by polishing the detector until it is so thin that it is flexible and almost transparent. Shows the smoothing features of an IPC Both of those options are very expensive to coupling at 10% on star and image noise fabricate. A different approach is to apply (dark dominated). a coating of nanometer-scale materials to the surface of a detector chip to convert the 4. Imaging Polarimetry incoming UV light is to visible light which is more readily recorded by standard detec- Imaging polarimeters utilizing the (3) The radiation hardness of the DMD. tor chips. This research has developed a division-of-focal technique present Tests were conducted using the Law- method of coating detector arrays with nano unique challenges during the data re- rence Berkeley National Laboratory 88” materials and applied it to improve the abil- duction process. Because an image is Cyclotron to irradiated the DMDs with ity of detectors to record UV and blue light. formed directly on the polarizing optic, high energy protons. The tests showed each pixel “sees” a different part of that the DMDs worked well when the scene; this problem is analogous to exposed to a dose equivalent to that the challenges in color restoration that found at an L2 orbit over a period of five arise with the use of Bayer filters. years. A picture of the test configuration Although polarization is an inherent at the end of the proton beam line is property of light, the vast majority of shown in the figure. Further radiation light sensors (including bolometers, tests using heavy ions are planned at the semiconductor devices and photograph- Texas A&M Cyclotron facility. ic emulsions) are only able to measure Quantum Dot coated detector in aluminum the intensity of incident radiation. A mask under UV illumination. The active area polarimeter measures the polarization of is 15mmx15mm the electromagnetic field by converting differences in polarization into differences in intensity. The microgrid polarizer array (MGPA) divides the focal plane into an array of superpixels. Each sub-pixel samples the electric field along a different direction, polarizing the light that passes through it and modulating the intensity (4) The DMDs are supplied by Texas according to the polarization of the light Instruments with a protective borosilicate TEM image of thin quantum dot film showing and the orientation of the polarizer. glass window. This glass limits the range of well distributed individual dots on an array.

RIT Imaging Science 2014–2015 Annual Report We are actively looking at techniques Chip Description Keithley. The enclosure is mounted on for hybridizing microgrid polarizer The chip used in these experiments XYZ and rotation stages for alignment 31 arrays to commercial CID, CCD and was a custom designed and fabricated purposes. The Keithley is commanded CMOS arrays. in a 0.35 μm silicon CMOS process us- via a MATLAB serial interface for ap- ing the MOSIS facility. On the chip are plying bias sweeps and relaying data. four test imager arrays and five test The radiation source is a ≈ 5 mW 188 transistors. These ‘test’ transistors GHz Gunn diode from Virginia Diodes. can be connected directly to outputs An illustration of the experimental for characterization without clocking setup is shown below. electronics. Our work has focused on characterizing the response from these five test transistors. The figure below shows a micrograph of the test chip with the test transistors located on the bottom edge.

A polarimeter using an MGPA. The light passing through a single polarizer is collect- Schematic of laboratory experimental setup ed by a single pixel. Patent (5) THz Imaging Robinson R. and Ninkov Z. [2010] A silicon CMOS based array purposed Enhancing Focal Plane Array Quantum for the terahertz regime has prom- Efficiency with Quantum Dots ising applications for many fields including security screening, manu- US Patent Application Serial Number facturing process monitoring, com- 12/655,350 munications, and medicine. Current PCT International Patent Application systems mainly consist of bulky tech- Number PCT/US10/62159 nology, including large pulsed laser systems and are primarily laboratory Current generation MOSIS devices. Five test Sample Publications based setups. A silicon CMOS based structures are seen along the bottom. 1. Optical simulation of terahertz technology was chosen in order to eventually develop a compact, porta- antenna using finite difference time ble, practical imaging system. A large domain method amount of recent research has been Chao Zhang (Carlson Center for Imag- conducted regarding the detection ing Sci., Rochester Inst. of Technol., of terahertz using silicon MOSFETs. Rochester, NY, United States); Ninkov, The THz focal plane technology being Z.; Fertig, G.; Kremens, R.; Sacco, A.; tested is uncooled and employs direct Newman, D.; Fourspring, K.; Lee, P.; overdamped, plasmonic detection Ignjatovic, Z.; Pipher, J.; McMurtry, with silicon CMOS MOSFETs that are C.; Dayalu, J. Source: Proceedings of each coupled to an individual mi- the SPIE, v 9483, p 94830D (11 pp.), cro-antennae. 2015 2. Transmission imaging measurements at 188 GHz with 0.35μm CMOS MOSIS test chip on fan-out board mounted technology in test housing. Sacco, A.P. (Exelis Geospatial Syst., Rochester, NY, United States); New- Test Description man, J.D.; Lee, P.P.K.; Fourspring, K.D.; The transistors were biased using a Osborn, J.H.; Fiete, R.D.; Bocko, M.V.; Keithley 2602 Source Measurement Ignjatovic, Z.; Pipher, J.L.; McMurt- Unit (SMU) which connects to the ry, C.W.; Xi-Cheng Zhang; Dayalu, J.; test enclosure via low noise shielded Seery, K.; Chao Zhang; Bhandari, S.; A photo of experimental setup is shown twisted pair cables. The enclosure Ninkov, Z. Source: Proceedings of the above. The Gunn Diode source is on the creates a Faraday cage around the SPIE, v 9483, p 94830U (7 pp.), 2015 left(1), followed by the shutter(2), and the fan-out board and test chip, and 3. Design, fabrication and characteri- test enclosure(3). The enclosure is mounted the connections are fed through the zation of a polarization-sensitive focal on XYZ and rotation stages for alignment of box with feed-through capacitors to plane array the MOSFET of interest with the source. Re- reduce as much RF noise as possible. Vorobiev, Dmitry (Rochester Insti- sponse is viewed in real-time for alignment A removable high resistivity silicon tute of Technology, Center for Imag- with a source measurement instrument, or a window on the front of the enclosure ing Science, 54 Lomb Memorial Dr, lock-in amplifier (Drain bias current, Id not precedes a high speed shutter which Rochester; NY, United States); Ninkov, available when using the lock-in. is controlled via digital I/O from the Zoran Source: Proceedings of SPIE - The International Society for Optical Judith L.; Sacco, Andrew P.; Zhang, 32 Engineering, v 9403, 2015 Chao Source: Proceedings of SPIE - 4. Design, fabrication and characteri- The International Society for Optical zation of a polarization-sensitive focal Engineering, v 9102, 2014, Terahertz plane array Physics, Devices, and Systems VIII: Advanced Applications in Industry Vorobiev, D. (Center for Imaging Sci., and Defense Rochester Inst. of Technol., Rochester, NY, United States); Ninkov, Z. Source: 7. THz imaging Si MOSFET system Proceedings of the SPIE, v 9403, p design at 215 GHz 94030A (12 pp.), 2015 Sacco, Andrew P. (Exelis Geospatial 5.Polarization in a snap: Imaging po- Systems, Rochester NY 14606, United larimetry with micropolarizer arrays States); Newman, J. Daniel; Lee, Paul P. K.; Fourspring, Kenneth D.; Osborn, Vorobiev, Dmitry (Center for Imag- John H.; Fiete, Robert D.; Bocko, Mark ing Science, Rochester Institute of F.; Ignjatovic, Zeljko; Pipher, Judith L.; Technology, 54 Lomb Memorial Drive, McMurtry, Craig W.; Zhang, Xi-Cheng; Rochester NY, 14623, United States); Dayalu, Jagannath; Fertig, Gregory J.; Ninkov, Zoran; Gartley, Michael Zhang, Chao; Ninkov, Zoran Source: Source: Proceedings of SPIE - The Proceedings of SPIE - The Interna- International Society for Optical Engi- tional Society for Optical Engineering, neering, v 9099, 2014, Polarization: v 9102, 2014, Terahertz Physics, Measurement, Analysis, and Remote Devices, and Systems VIII: Advanced Sensing XI Applications in Industry and Defense 6. T-ray detection in 0.35-μm CMOS technology

Fertig, Gregory J. (Rochester Insti- tute of Technology, Carlson Center for Imaging Science, Rochester, NY 14623, United States); Ninkov, Zoran; Bocko, Mark F.; Dayalu, Jagannath; Fourspring, Kenny D.; Ignjatovic, Zeljko; Lee, Paul P. K.; McMurtry, Craig W.; Newman, J. Daniel; Pipher,

RIT Imaging Science 2014–2015 Annual Report 33 RESEARCH 34RESEARCH

RIT Imaging Science 2013–2014 Annual Report RESEARCH 35 MULTIDISCIPLINARY VISION RESEARCH LABORATORY Laboratory Director’s Comments By Dr. Jeff Pelz

The MVRL continues to grow as Dr. Christopher Kanan joins the group this Fall. The Kanan Lab focuses on the development and application of brain-inspired algorithms for machine perception.

The goal is to identify the algorithms that underlie human perception and to leverage these algorithms for the purpose of computational image understanding and fine-grained object recognition. A central theme of Dr. Kanan's research has been studying how humans move their eyes during goal-directed activities. RESEARCHBy analyzing eye movements, he can estimate various parameters of the user’s mental state, and use this information to predict where the user is most likely to look next. In addition, Dr. Kanan has created computer vision algorithms that make simulated eye movements to improve their ability to recognize, detect, and track objects in images. Much of Dr. Kanan's recent work has involved applications of deep hierarchical neural networks to problems in computer vision. With CIS Ph.D. student Mohammed Yousefhussien, he has been developing the Smooth Pursuit tracking algorithm, which uses deep learning with saliency maps to track multiple targets of interest in videos. Saliency maps are a topologically organized maps of what is interesting in a scene and are a model for human attention. Their algorithm’s performance rivals or surpasses state-of-the-art tracking algorithms at tracking targets in video captured by an aerial vehicle. Dr. Kanan is also engineering new techniques to allow machines to reason about images. Computer vision has often assumed that to understand an image means to attribute each pixel in an image to particular object category. In contrast, humans selectively process their visual environment in a way that is heavily influenced by the current task and goals. This observation has motivated Dr. Kanan, in collaboration with CIS Ph.D. student Kushal Kafle, to focus upon the study of visual question and answer systems. These systems take as input a text based question (the goal) and an image. In return, the algorithm generates a text-based answer to the question. Prior to coming to CIS, Dr. Kanan was a member of the Maritime and Aerial Perception group at NASA's Jet Propulsion Laboratory/California Institute of Technology, where he helped build vision systems for autonomous robots. He earned a Ph.D. in computer science at the University of California, San Diego, working with Dr. Gary Cottrell at the intersection of computer and cognitive science. He received an M.S. in Computer Science at the University of Southern California, working with Dr. Michael Arbib, professor of computer science, bio- Most algorithms for object recognition logical sciences, biomedical engineering, neuroscience, and psychology. process the image in one-shot. Instead, Dr. Kanan adopted an active approach in which a simulated eye looks at multiple regions of the image. During each fixation, the model extracts foveated features from a local region of the image. As more fixations are acquired, the model's confidence in what it is observing increases. websites that utilize ASL cancer health 36 videos for Deaf consumers, with goals of maximizing effectiveness of dissemination for health-related purposes. R01 grant: This project plan builds on Dr. Kushalnagar’s prior research on quality of life outcomes with deaf and hard of hearing (DHH) population. This project will be conducted by a research team that includes: an early career investigator with expertise in quality of life outcomes in DHH population, a senior investigator with expertise in patient reported outcome measures (the creator of the FACIT fatigue QoL measure), a psychometrician with experience in conducting IRT and psy- chometric analyses for patient reported outcome measures, an audiologist with expertise in (re)habilitative audi- ology and hearing related outcomes, and an ASL translation expert. Results from this study will provide a better Each circle represents a filter that is applied understanding of the mechanisms that to regions fixated during simulated visual may contribute to poor quality of life search. These filters are not engineered, frequently observed for individuals but result from the application of machine with hearing loss. learning techniques to natural images. These artificial neurons share properties with Publications neurons in the human primary visual cortex: Gabor-like filters with red-green, blue-yellow, AHP Morice, GJ Diaz, BR Fajen, N and dark-light opponency. Basilio, G Montagne (2015). An affordance-based approach to visually Newly funded research aims to investi- guided overtaking, Ecological Psy- gate national trends in health-related chology 27 (1), 1-25 Internet usage and patient-reported Bona, S., Herbert, A., Toneatto, C., outcomes among deaf and hard of hear- Silvanto, J., & Cattaneo, Z. (2014). ing individuals The causal role of the lateral occip- Dr. Poorna Kushalnagar is a Research ital complex in visual mirror sym- Associate Professor in CIS. She was re- metry detection and grouping: an cently awarded R15 and an R01 grants, fMRIguided TMS study. Cortex, 51, totaling around 2 million dollars over 46-55. http://dx.doi.org/10.1016/j. 3 to 5 years. Both grant projects will cortex.2013.11.004 include developing and testing visually Cattaneo, Z., Bona, S., Bauer, C., accessible applications in American Silvanto, J., Herbert, A. M., Vecci, T., & Sign Language and English that can be Merabet, L. B. (2014). Symmetry de- used by people from diverse back- tection in visual impairment: Behav- grounds. Both projects will include a ioral evidence and neural correlates. large team of Deaf/HH undergraduate Symmetry, 6, 427-433. doi:10.3390/ student researchers in psychology, sym6020427 information technology, and science. Collaborating institutions include Gal- Cattaneo, Z., Bona, S., Monegato, laudet University, Institute for Public M., Pece, A., Vecchi, T., Herbert, A. of accessible technology and services M., & Merabet, L. B. (2014). Visu- Health and Medicine at Northwestern will provide important insights on University Feinberg School of Medi- al symmetry perception in early understanding the trends in Deaf peo- onset monocular blindness. Visu- cine, and University of Arkansas for ple's use of the Internet for health-re- Medical Sciences. al Cognition, 22, 963-974. DOI: lated purposes and improving health 10.1080/13506285.2014.938712 R15 grant: National Cancer Institute’s communication models that will lead Health Information National Trends to better personal and public health Easterbrooks, SR, Lederberg, AR, Antia, Survey (NCI-HINTS) is heavily depen- within the underserved deaf popula- S., Schick, B., Kushalnagar, P., Webb, dent on English, which present serious tion. The research plan builds on Dr. M., Branum-Martin, L., & Connor, CM language barriers to Deaf patients who Poorna Kushalnagar's prior research (2015). Reading among Diverse DHH use American Sign Language (ASL). on health communication with Deaf Learners: What, How and For Whom? The availability of HINTS in ASL and signers. Expected findings will support American Annals of the Deaf. English and such that is valid for users improvement and refinement of health

RIT Imaging Science 2013–2014 Annual Report Guo X., Li R., Alm C.O., Yu Q., Pelz J., Kushalnagar, P., Bruce, S.*, Sutton, T., celia Ovesdotter Alm, Reynold Bailey, Shi P., Haake A. Infusing Perceptual & Leigh, IW (under review). Early Life Joe Geigel, and Linwei Wang, “Face- 37 Expertise and Domain Knowledge into Communicative Difficulties, Gender, speech sensor fusion for non-invasive a Human-Centered Image Retrieval and Adulthood Depression. stress detection”, 1st Joint Confer- System: A Prototype Application. Kushalnagar, P., Naturale, J., Palud- ence on Facial Analysis, Animation Proc. of ETRA 2014, 275-278 neviciene, R., Smith, SR, Werfel, E.*, and Audio-Visual Speech Processing Guo X., Yu Q., Alm C., Calvelli C., Pelz J., Doolittle, R., Jacobs, S., DeCaro, J. (FAAVSP 2015), to appear. Shi P., Haake A. From Spoken Narratives (2015). Health Websites: Accessibil- Wang, P., Cottrell, G., Kanan, C. (to to Domain Knowledge: Mining Linguis- ity and Usability for American Sign appear) Modeling the Object Recog- tic Data for Medical Image Understand- Language Users. Journal of Health nition Pathway: A Deep Hierarchical ing. Artificial Intelligence in Medicine, Communication. Model Using Gnostic Fields. In: Proc. Elsevier. 62(2), 79–90, 2014. Rantanen, E. M., Alm, C. O., Worrell, 36th Annual Conference of the Cognitive Guo X., Yu Q., Li R., Alm C.O., Haake T., Valentage, N., & Iuliucci, N. (2014, Science Society. A.R. Fusing Multimodal Human Expert June). Linguistic analysis of clinical Will Paul, Cecelia Ovesdotter Alm, Data to Uncover Hidden Semantics. communications: A novel method for Reynold Bailey, Joe Geigel, and Linwei GazeIn ’14 Proc. 7th Workshop on Eye study of diagnostic errors.Proceedings Wang, “Stressed out: What speech Gaze in Intelligent Human Machine of the International Symposium of Human tells us about stress”, Interspeach Interaction: Eye-Gaze & Multimodali- Factors and Ergonomics in Healthcare, 2015, to appear. ty. ACM New York, NY, (21-26), 2014. 3(1), 207-213. SAGE Publications. Guo X., Yu Q., Li R., Alm C.O., Haake Schenkel, L. S., Towne, T. L., Herbert, A.R. Fusing Multimodal Human Expert A. M., Pelz, J. B. (2015) Errors in Data to Uncover Hidden Semantics. identifying emotion in body postures GazeIn ’14 Proc. 7th Workshop on Eye and facial expressions among youth Gaze in Intelligent Human Machine with pediatric bipolar disorder. 17th Interaction: Eye-Gaze & Multimodali- International Conference on Bipolar ty. ACM New York, NY, (21-26), 2014. Disorder, Toronto. Hochberg L., Alm C. , Rantanan E., Schenkel, L. S., Towne, T. L., Herbert, Yu Q, DeLong C., Haake A. Towards A. M., Pelz, J. B. (2015) Errors in automatic annotation of clinical identifying emotion in body postures decision-making style (pp. 129-138). and facial expressions among youth Proceedings of LAW VIII—The 8th with pediatric bipolar disorder. 17th Linguistic Annotation Workshop at International Conference on Bipolar COLING 2014, Dublin, Ireland. Disorder, Toronto. Kanan, C. (2014) Fine-grained object Smith, SR., Kushalnagar, P., & Hauser, recognition with Gnostic Fields. In PC. (2015). Deaf adolescents’ learning Applications of Computer Vision (WACV), of cardiovascular health information: 2014 IEEE Winter Conference on (pp. Sources and access challenges. Journal 23-30). IEEE. of Deaf Studies and Deaf Education. Kanan, C., Bseiso, D. N., Ray, N. A., Srinivas Sridharan, Reynold Bailey, Hsiao, J. H., & Cottrell, G. W. (2015). "Automatic Target Prediction and Humans have idiosyncratic and Subtle Gaze Guidance for Improved task-specific scanpaths for judging Spatial Information Recall", 12th ACM faces. Vision research, 108, 67-76. Symposium on Applied Perception Kanan, C., Ray, N. A., Bseiso, D. N., (SAP 2015), to appear. Hsiao, J. H., & Cottrell, G. W. (2014). Sutton, T., & Kushalnagar, P. (2015). Predicting an observer's task using Attention Bias for Signed Emotion multi-fixation pattern analysis. In Valence. Poster Presentation at the Proceedings of the symposium on eye annual American Psychological Asso- tracking research and applications (pp. ciation meeting, Toronto, CA 287-290). ACM. Vaidyanathan P., Pelz J., Alm C.O., Shi Khosla, D., Huber, D. J., & Kanan, C. P., and Haake A. Recurrence quanti- (2014). A neuromorphic system for fication analysis reveals differences visual object recognition. Biologically In- in eye-movement behavior between spired Cognitive Architectures, 8, 33-45. expert and novice dermatologists. Kushalnagar R. & Kushalnagar, P. Proc. of ETRA 2014, 303-306. (2014). Live and collaborative gaze Vaidyanathan, P., Prud’hommeaux, E., review for deaf and hard of hearing Alm, C. O., Pelz, J. B., & Haake, A. R. students. Proceeding from the Inter- (2015). Alignment of Eye Movements national Conference on Computers and Spoken Language for Semantic Helping People with Special Needs. Image Understanding. IWCS 2015, 76. Paris, France. Vasudev Bethamcherla, Will Paul, Ce- RESEARCH 38RESEARCH

RIT Imaging Science 2014–2015 Annual Report RESEARCH 39 BIOMEDICAL AND MATERIALS MULTIMODAL IMAGING LABORATORY Laboratory Director’s Comments By Dr. Maria Helguera

Mission: To develop innovative ways to visualize, analyze, and characterize biological tissues and synthetic materials by means of multimodal medical imaging devices.

Staff Dr. María Helguera is the principal investigator. The lab facilities are used as RESEARCHwell by Dr. N.A.H.K. Rao. Research in the lab was conducted and supported by a number of students: • Amy Becker, Imaging Science, “Structure Analyses of Artificial Tissues”. In collaboration with Biomedical Engineering Department, University of Rochester • Megan Iafrati, Rose Rustowicz, BS Imaging Science, “Remotely Accessible Microscope”. • Kfir Ben-Zikri, MS Candidate Imaging Science,“Image-Based Quantification and Analysis of Longitudinal Lung Nodule Deformations”. Sponsored by NIH. • Viraj Adduru, PhD Candidate Imaging Science, “Lesion Segmentation and Relapse Prediction Using Longitudinal MS Data”. Sponsored by Geisinger Health System. • Golnaz Jalalahmadi, PhD Candidate Imaging Science, “Design of a Com- prehensive Protocol for the Diagnosis and Prevention of Abdominal Aortic Aneurisms”. In collaboration with Geisinger Health System. Research Projects “Structure Analyses of Artificial Tissues”, Amy Becker, María Helguera in collaboration with Eric Comeau and Dr. Diane Dalecki, Biomedical Engineering Department, University of Rochester. A texture analysis software tool was used to quantify the structures in engineered tissues. This was done using a three dimensional Gray-Scale Co-occurrence Matrix (GLCM) and a Gray-Level Run Length Matrix (GLRLM) analyses. The software tool was modified to output nine orientations of the GLCM analysis. Test images were designed and used to determine the optimal parameters from this analysis. It was determined that Contrast, Homogeneity, Long Run Emphasis, Gray-Level Non-Uniformity, and Run Percentage were the most indicative parameters from the analyses. Ultrasound standing wave fields have been used to organize living endothelial cells in a collagen gel, with cells collecting at the nodes of the standing wave. The frequency and the pressure of the ultrasound wave determine the structure of the tissue that is grown. For this project, vasculature (blood vessels) were grown. When endothelial cells were placed in a collagen gel, an ultrasound standing wave field was applied to organize the cells into distinct bands. When the banded cells were set in an incubator and allowed to grow, vessel sprouts formed in between the bands. The “Remotely Accessible Microscope”, user selects the desired position by sprouts connect the bands to create Megan Iaftrati, Rose Rustowicz and the click of the mouse. The motors 40 a complex structure of vessels. Engi- María Helguera. This project won the move the microscope/camera unit to neered tissues must be irrigated to 2014-2015 CIS Faculty Award Celebrat- the desired position while a window ensure viability upon implantation. ing Excellence and the UNYTE - Hitting displays the field of view. Fine posi- Samples were exposed to 1 MHz ultra- the Accelerator: Health Research Innova- tion adjustments can be done using sound standing wave fields of different tion through Data Science. Best Student the arrows in the keyboard. This GUI pressures. The pressures used in this Poster Award. also controls the illumination system, project were 0.1 MPa, 0.2 MPa, and We designed and built a portable, low mounted on the microscope/camera 0.3 MPa. It has been observed that the cost, WiFi accessible microscope (20X unit, to be turned on-off to avoid pressure of the ultrasound wave affects to 200X) that fits inside an incubator photo-bleaching and photo-toxicity. the final structure of the bands and the for live cell screening in real time. A screen shot of the GUI is shown in sprouts. Representative examples are The system is based on a Raspberry Figure 4. shown in Figure 1, showing the effects Pi microcomputer and can hold two of 0.1 MPa in the left, 0.2 MPa in the cell culture plates for simultaneous middle, and 0.3 MPa in the right. scanning and time-lapse studies.

Figure 1. Band formation and sprouting after exposure to 1 MHz ultrasound standing wave fields with 0.1 MPa in the left, 0.2 MPa in the middle and 0.3 MPa in the right. Images taken with 5x magnification

Low Gray-Level Non-Uniformity Low Gray-Level Non-Uniformity

Figure 2. Microscopy representative images for which GLNU ranged from lowest to highest value.

Synthetic images reproducing some The system can be adapted to per- band and sprout formations were creat- form 2D or 3D scans, still images or ed to validate the experimental results. video, gray scale or color. With such Among the parameters for the GLCM a system in place there is no need to analysis, it was concluded that manipulate the samples and the risk homogeneity and contrast were the of contamination is greatly reduced. A two most indicative. Homogeneity block diagram of the system is shown and contrast are essentially opposite in Figure 3. Figure 4. Screen shot of the control GUI measures. Homogeneity is a mea- showing the field-of-view in the selected sure of smoothness and contrast is a tray-well. measure of business in an image. For Image processing algorithms to the images taken at 5x magnification, enhance, segment and count cells are the homogeneity increased with an available through another GUI. The increase in pressure used to create GUI displays the image to be pro- the bands. Figure 3. Block diagram showing the exter- cessed and a drop down menu pro- Among the parameters for the GL- nal computer controlling the system. vides options for enhancing contrast, RLM analysis, it was concluded that Ad hoc control and image process- sharpening, reducing noise, flattening Gray-Level Non-Uniformity (GLNU) ing software have been created in the illumination field, etc., as well as was best characterizing the texture, Python. The control graphical user a default setting. This preprocessing as is shown in Figure 2. interface (GUI) shows an image of stage is used to precondition the the cell culture plates on which the image. A dialog window asks the user whether he/she is satisfied with the

RIT Imaging Science 2014–2015 Annual Report image. If the answer is yes, segment- impractical and risky, however, to ing and counting of cells will start. biopsy every lesion that may be found This quantitative analysis stage runs in a computed tomography (CT) scan. 41 behind the scenes without any further To mitigate this risk, longitudinal input from the user. The resulting scans are collected to determine the segmented image highlighting the growth rates of individual lesions in cells will be displayed and the count order to determine if they are consis- of cells will be shown. tent with growth rates of known types Other quantitative analyses protocols of malignancies. Ground glass opacity may include gray-scale textural and (GGO) lesions are especially difficult volumetric analyses, cell tracking, etc. to analyze, as they feature a variety The system is easily customized for of non-solid components that may be different applications. influenced by lung deformation. Four cases are shown in Figure 5. Compen- To reduce contaminants and over-heat- sating for background lung deformation ing, the power supply, power distri- across longitudinal scans (via image bution circuits, and motor drivers are registration) is vital to enable accurate located outside of the incubator. estimation of lesion growth rate. We The students involved in the project demonstrate on four clinical cases that are seen in the pictures below dis- registration over a region of interest cussing their project and proudly (ROI) centered around the lesion yields showing their certificates. Megan lower target registration errors (TRE) Iafrati is on the top and Rose Rusto- than registration over the entire lung. wicz on the bottom. Results are shown in Figure 6.

Figure 5. Four longitudinal clinical cases are shown. Earlier (prior) images top row and current images bottom row. Results of these experiments suggest that registration based on lesion ROIs has the potential to more accurately account for deformations in GGO lesions than registration based on lung ROIs. In 3 of 4 cases for lesion ROIs, the median TRE was less than 1mm. This is less than the variability in the fiducial point locations selected by the radiologist. Registration based on lesion ROIs had better performance in 3 of 4 cases, suggesting that a limited ROI enables Figure 6. Target Registration Errors (TRE) registration algorithms to converge to (mm) for Cases 1–4 after different registra- “Image-Based Quantification and the correct global minimum. One pos- tion algorithms were applied to lung versus Analysis of Longitudinal Lung Nodule sible reason for this is that the use of lesion ROIs Deformations”, Kfir Ben-Zikri, María a larger ROI may increase the number “Lesion Segmentation and Relapse Helguera and Nathan Cahill, School of local maxima in the similarity mea- Prediction Using Longitudinal MS of Mathematics, RIT, in collaboration sure. It was found that affine registra- Data”, Viraj Adduru and María Hel- with Kitware, and Mark Niethammer, tion outperformed rigid registration guera in collaboration with Andrew University of North Carolina. in all four cases, both when lung and Michael, Geisinger Health System. lesion ROIs are used. Lung cancer is the leading cause Multiple sclerosis (MS) is a chronic of cancer related-deaths in the US, By visually inspecting the registered disease of the central nervous sys- accounting for 28% of all cancer-re- lesions, it appears that registration with tem characterized by inflammation, lated mortality in 2012, and having lesion ROIs involved alignment of both demyelination, and axonal degenera- a 16% five-year survival rate. If lung the lung and nodule boundaries, while tion. The onset is with a neurological cancer is diagnosed at an early stage registration with lung ROIs only in- disturbance, known as clinically iso- when the cancer is still localized in volved alignment of the lung boundary. lated syndrome (CIS) and the patients small lesions or nodules, the five-year experience unpredictable episodes survival rate improves to 52%. It is of clinical relapses and remissions T2) MS brain MR images. The prepro- volume, and biomechanical factors such 42 followed by continuous progression cessing steps include Brain extraction as peak wall stress are the factors which of disability over time in most instanc- (removal of non-brain tissue), image might influence the growing weakness es, leading to clinically-definite MS cropping (removal of background in the aortic wall as well as the rupture (CDMS). However, about 20% of CIS voxels), bias correction (correcting for risk. Therefore, surgery only based on patients do not convert to MS after two changes in magnetic intensity), and the AAA’s diameter might end up in decades, even if they have an abnor- intensity standardization (scaling voxel remarkable clinical and financial efforts mal brain scan at onset. Magnetic intensities to a standard range). The which for a notable portion of patients Resonance Imaging (MRI) is one of the pipeline is developed in python using seem to be unwarranted with high risk of diagnostic imaging methods for MS. Nipype libraries which provided tools for death after all. MRI’s of MS patients primarily exhibit package integration (allowed usage of The overall goal of this work is to devel- focal and less often diffuse WM lesions various neuroimaging tools in one place) op an inclusive model that incorporates in the brain and spinal cord. Typically and parallel-processing on a multi-core geometrical and mechanical factors, as the images are visually assessed for processor. This simplified the algorithm well as physiological and demographic qualitative analysis and lesions are execution and reduced the computation patterns. With such a methodology in manually marked if a quantitative time significantly. place, data for new patients will be analysis is required. Manual marking The preprocessed images free of non- assessed to evaluate the probability of of lesion-boundaries is a very time brain tissues and inter-subject intensity rupture. After all, there is an increasing consuming and has a high inter and variability are divided into test, training need for more reliable and accurate intra-subject variability. and validation sets. 3D patches con- methods for the prediction of the rupture For the past two decades many segmen- taining lesions from training sets are risk and the treatment of AAA. tation algorithms have been developed selected from each channel (T1, T2 and to detect MS lesions in structural MRI FLAIR). The machine learning algorithm Selected Publications and images but no approach is widely pre- will be trained with these patches and Conference Presentations ferred due to poor generalization capa- evaluated for its performance using the 1. Mercado, K.P. *, Helguera, M., bility across images acquired from var- validation and test sets. Care will be tak- Hocking, D.C., Dalecki, D. “Non- ious sites. This is mainly because of the en to balance the number of lesion to the invasive Quantitative Imaging inter-subject and inter-site variability in healthy tissue voxels to prevent skewing. of Collagen Microsctructure in tissue intensities. Moreover very limited Three-Dimensional Hydrogels studies are available on the prediction “Design of a Comprehensive Protocol using High Frequency Quantitative of the relapse and used simple machine for the Diagnosis and Prevention of Ultrasound”, Tissue Engineering, learning like SVM on the patient demo- Abdominal Aortic Aneurisms”, Golnaz Part C: Methods. DOI: 10.1089/ten. graphics and basic lesion properties like Jalalahmadi and María Helguera, TEC.2014.0527, 12/2014. size and distance from center. However in collaboration with Aalpen Patel, the structural features of the lesions Geisinger Health System. 2. Mercado, K.P., Langdon, J., Helguera, and their multimodal intensities along Abdominal aortic aneurysm (AAA) is M., McAleavey, S.A., Hocking, D.C., with temporal lesion progression can an abnormal dilation which causes an Dalecki, D., “Scholte wave genera- be exploited to learn much more robust expansion at least 1.5 times of the nor- tion during single tracking location machine learning algorithms for lesion mal diameter. It is known as the 13th acoustic radiation force impulse segmentation and relapse prediction. reason for death in the United States imaging of engineered tissues”, sub- The purpose of this project is to explore and the reason for mortality for 2-4% mitted to Journal of the Acoustical better lesion segmentations methods in elderly. In 2009, the Society for Vas- Society of America Express Letters, for MS and use machine learning to cular Surgery (SVS) proposed for more March 2015. Accepted. predict the occurrence of relapse. To investigations toward the rupture risk 3. Ben-Zikri, K.*, Helguera, M., Fetzer, achieve these goals we will use the large factors and the treatment methods for D., Chittajallu, D., Aylward, S., dataset of longitudinal MRI scans and abdominal aortic aneurysms. Niethammer, M., Cahill, N. “Lon- demographics of the patients collected The main reason for AAA is not known gitudinal Registration of Ground over time using Geisinger’s unique data yet completely but the growing weak- Glass Opacity Lessions in CT resources and information technology ness in the aortic wall has been intro- Scans”, International Symposium capabilities. The main aims of the proj- duced as the main reason which might on Biomedical Imaging (ISBI), ect are: be affected by different types of factors. April, 2015 1. Observe the data and design For decades, the maximal diameter 4. Rustowicz, R.**, Iafrati, M. **, Hel- preprocessing pipelines for noise (Dmax) was considered as the main guera, M., “Remotely Accessible Micro- removal and normalization. factor for this growth and in the process scope (RAM)”, UNYTE, May 2015. 2. Extract useful features by using of clinical follow up and needs for sur- Winner of Best Student Poster Award. the information from prepro- gery or repair for the patients. However, recent studies demonstrated that the * Graduate Student, ** Undergradu- cessed images, patient demo- ate Student graphics and clinical exams. Dmax is not a sufficient factor for the rupture since patients with smaller 3. Investigate machine learning diameter have gone through the rupture Grants and Research Funding methods for relapse prediction. by a range of 1% of rupture per year. It Funding was provided by NIH, and We have developed automatic image was indicated that demographic factors FINCyT (Peru). processing pipelines for preprocess- like age and gender, geometrical- and ing the multimodal (FLAIR, T1 and shape-related factors like diameter and

RIT Imaging Science 2014–2015 Annual Report Additional research in the Biomedical Imaging Laboratory is dedicated to: (a) The development of a new imaging modality called Photoacoustic (PA) Imaging. Focus is on implementing novel 43 methodologies for PA imaging. (b) Building internal and external collaborations to assist the biomedical research community in addressing cancer diagnosis, prognosis and disease management challenges. This is being done by pragmatically using imaging technologies, image analysis and development of cancer cell targeted contrast agents as biomarkers for molecular imaging. We have made significant progress in each of these two major areas. Both were highly interdisciplinary in nature with scientists representation departments other than Center for Imaging Science at RIT and University of Rochester. We were successful in generating over a million dollars in research grants last year. Three major grants where the BIL personnel were directly involved are listed below: (1) R15 EB 019726-01 –(PI-Rao) Agency: NIH $436,000 02/01/2015 –01/31/2018 Project title: Development of Photoacoustic and ultrasound transrectal probe for prostate biopsy. (2) R15 CA 192148-01- (PI: Schmitthenner, Co-I: Rao) Agency: NIH $440,000 09/01/2014 - 08/31/2017 Project title: Targeted Molecular Agents for Photoacoustic Imaging of Prostate Cancer (3) CDMRP PCRP Exploration Hypothesis Award (PI:Dogra, Co-I: Rao) Agency: DOD $100,000 04/01/2014 –03/31/2015 Project Title: In vivo photoacoustic imaging of prostate cancer using targeted contrast agent

Student Participation: Students and to first evaluate and optimize lens research associates involved in the performance was adopted by simply projects were, Saugata Sinha (Ph.D. changing the medium velocity from completed), Viraj Adduru (MS com- that of light to sound in water. A pleted), Zichao Han (Ph.D. student), material commonly available for three Arjun R. Rajanna (EE MS student), dimensional printing was chosen with Jacob Wirth (Ph.D. student), Mathew refractive index of 0.58. The bi- con- Casella, Michaela Piel, Amy Becker cave lens diameter was 32 mm, radii (BS student), Nicholas Brown, Pedro of curvature 33 mm and the working Vallejo, Tess Jacobs (BS optics U of focal length f ∼ 40 mm. Fig.1 shows R), Bhargava Chinni (former RIT MS the ray tracing at 4f configuration graduate, now research associate, U and Fig.2 is the modulation transfer of R), Shalini Singh, Ph.D. ( Research function (MTF) for on-axis point in red Fig. 2. Singlet lens MTF for on- and off- axis Associate, U of R). and extreme off-axis point in blue. points. Clearly we are close to the diffraction A comprehensive two dimensional (2D) Project-1: Advancement in Acoustic limited performance on-axis with a simulation model (see Fig.3) of the 4f con- Lens Design Methodology for PA MTF value of 0.3 at 0.7 cycles/mm but figuration was developed that incorporates significantly worse for off-axis points. Imaging Camera realistic PA signal generation, acoustic Optimization efforts demonstrated Collaborators: B. Chinni, W. Knox, wave propagation from source to transduc- that this is the best we can do using a V. Dogra, J. Bentley (U of R) and R. er element including the focusing through singlet lens system. Ptucha (RIT) the lens and the transducer frequency In last year’s report we described the response. Snapshot of the propagating PA realization of an innovative acoustic signal wave-front at different times after lens focusing based camera for PA the laser pulse firing is shown in Fig.3. PA imaging. A large study encompassing point source is located at pixel number more than 100 freshly excised human 113. A diverging wave-front is shown near prostate and thyroid tissue samples pixel number 350. At a later time the lens was conducted. Preliminary data located at pixel number 512 has started to analysis of PA signal from histology focus the wavefront shown near pixel num- proven areas designated as normal ber 700. The final image plane (the sensor or cancer showed better than 85% position) is located at pixel number 900. accuracy for cancer diagnosis. Further Focused waves are detected at each array studies with neural network based elements as a radio- frequency (RF) time data analysis are in progress. signal (A-line data). Only the slowly varying Fig. 1. Ray tracing through singlet lens in envelope of RF signal is used for generating The image quality of PA camera can Code V optical design software. 2D display of the PSF as a B-scan image in be improved with improvement in Fig. 4. The middle and the right panel show acoustic lens design. An innovative how the PSF might change when the point idea to use optical lens design soft- source is moved off-axis to 2.5 mm and 5 ware called Code V from Synopsys mm distance respectively. Fig. 3. Photoacoustic wave simulation using 44 MATLAB software. minimal detectable tumor size; and ii) endogenous molecules have little specificity for cancer. In order to improve depth penetration and image quality, exogenous chromophores can be em- ployed to enhance sensitivity relative to endogenous agents. This approach has been demonstrated in the use of a near infrared fluorescent (NIRF) dye, IR800CW conjugated to a peptide that targets the neutropilin-1 receptor for the PAI of breast cancer. Excitation of cells can be optimized by use of a laser tuned to the maximum wavelength of NIR dyes with absorptivity factors two to three orders of magnitude greater than those of endogenous agents. Unlike fluorescence imaging with light as the input and backscattered radiation as the output, PA imaging uses light as the input source for excitation (a) On axis (b) 2.5mm off axis (c) 5.0mm off axis but detection and image formation use ultrasound waves generated by the Fig. 4. Simulation results: PSF slices for on Project-2: Development of targeted tissue. Since ultrasonic waves scatter and off axis points, where the horizontal and Contrast Agents for Prostate Cancer much less than light in the tissue, vertical axis respectively corresponds to the PA Imaging PAI can produce higher resolution distance from the center of the pattern and imaging deep in tissue, compared sensor in the center (in mm). to fluorescent detection. For greater An interdisciplinary team has suc- tissue depth penetration and sensi- cessfully integrated ray tracing based tivity, PAI utilizes dyes that absorb in optical lens design with full wave the ‘biological NIR window’ between theoretic simulation of the resulting 700-900 nm. The optimal NIR window PSF. Using this methodology we have is designed to circumvent the strong fabricated and tested a first gener- absorbance of Hb, HbO2 and H2O ation prototype PA imaging camera as shown in Fig. 5. NIR dyes related that can take 1 cm×1cm 2D B-scan to ICG including Cy7, Alexa750 and images in l second and 2D slices of IR800CW are ideal and offer a dramat- C-scan images in less than 2 minutes. ic increase in sensitivity. The innovative element of the system Identification of IRDye800CW as an is a single lens acoustic focusing exogenous contrast agent for PAI. To element. A complete system simula- Fig. 6. Spectral analysis of near-infrared dyes. quantify the PA signal generated with tion modeling, design and fabrication Contrast agents in a cuvette have been exam- tunable laser excitation in the 700-1000 methodology has been described. ined for a photoacoustic signal using the laser nm range of endogenous or exogenous Improvement in image quality using from 700 nm to 1000 nm wavelength. dye components, algorithms have been optical lens design methodologies are developed to de-convolute the individ- currently under investigation. Collaborators: B. Chinni, S. Singh, K. ual chromophore PA images. Five dyes Nastiuk, J. Krolewski, V. Dogra (U of (IRDye800CW, AlexaFluor750, Cy7- R), H. Schmitthenner (RIT) NHS-ester, Cy7-sulfo and Dylight800) were tested using our acoustic lens Limitations of endogenous contrast based device at 100 micromolar (µM) agents for prostate cancer diagnosis. concentration. A water bath setup (Fig. While pre-clinical testing of prostate 6) was used to determine the photo- resection slices indicates that Pho- acoustic spectra of these five dyes. Each toacoustic Imaging (PAI) has higher contrast agent was interrogated in the sensitivity and specificity than TRUS, NIR region in a five step-interval range imaging depth is limited, and chro- and PA signals recorded using our PAI mophores such as de-oxyhemoglobin probe. These recorded PA signals were (dHb) and oxyhemoglobin (HbO2) have further processed to find the absorption two limitations: i) their small absorp- maxima for each exogenous contrast tivity factor (extinction coefficient) agent across the NIR region. The PA Fig. 5. Biological imaging near-infrared leads to weak PA signals, limiting the absorption spectra of each dye have window. depth of tumor detection as well as the been plotted as shown in the Fig. 7.

RIT Imaging Science 2014–2015 Annual Report 45

Fig. 9. Sensitivity of IRDye800CW using our PAI device. IRDye800 concentration varied from 90 nM to 50 µM. Each red square Fig. 7. PA spectra of five contrast agents. Fig. 8. Optical absorption spectra of five represents the corresponding IRDye800CW IRdye800CW (blue); AlexaFluor750 (red); contrast agents. IRdye800CW (blue); Al- concentration and its PA signal intensity. Cy7-NHS-ester (green); Cy7-sulfo (orange); exaFluor750 (purple); Cy7-NHS-ester (red); Dylight800 (purple). Cy7-sulfo (orange); Dylight800 (green). three with (PSMA-) PC3 cells stained with dye-aptamer, and three with As expected from the reported peak Sensitivity of IRDye800CW using our PAI (PSMA+) C4-2 cells stained with dye-ap- intensities (lambda max) in the ultravi- device. The IRDye800CW was diluted tamer, all in a multi-well plate. Initially, olet-visible spectra, the Alexafluor 750 from 50 µM to 90 nM concentrations and the gel phantoms were imaged using has a peak PA signal when irradiated at photoacoustic signal measured to obtain a commercially available fluorescence 750 nm, Cy7-NHS-ester at 760 nm, Cy7- the sensitivity of the IRDye800CW. Each imaging system (IVIS spectrum) at 770 sulfo at 755 nm, Dylight800 at 785 nm red square in the plot represents the nm to confirm the dye aptamer binding and the dye IR800CW has a peak signal corresponding IRDye800CW concen- to PSMA (Fig. 10B, D). Once the bind- at 775 nm. When the targeted molecular tration on the x axis and its PA signal ing was confirmed, the phantoms were imaging agents (TMIAs) are injected intensity on the y axis. PA signal from removed from the multi-well plate and into the blood stream and PA imaging is the dimethyl sulphoxide and de-ionized placed on the imaging stage as shown in conducted during wash out period, the water has been subtracted from each the photograph (Fig. 10A). The PAI scan- image signal will be the linear sum of recorded value and PA signal from the ning procedure (of the yellow square signal from all the underlying dominant dye alone has been plotted. Using our area) was similar to that used in the ex chromophores (dHb, HbO2 and NIR dye). current PAI system, 90 nanomolar (nM) vivo imaging of tissue slices, using the It is important to create separate images was determined to be the threshold for prototype instrument. Fig. 10C shows of individual chromophores to obtain signal above noise (Fig. 9). a PA C-scan image of the gel phantoms. the signature related to only the specific PAI of PCa cells using IRDye800 conjugat- The PA signal intensity was increased biomarker. This task requires input data ed with A10.3 aptamer bonded to PSMA. in the ROI surrounding the C4-2 gel from the PA signal spectrum of endog- We have received a 1-year seed grant phantoms, demonstrating the presence enous or exogenous dye components from Department of Defense Prostate of IRdye800CW conjugated with A10.3 and algorithms have been developed to cancer research program to develop and aptamer bound to cell surface PSMA. de-convolve the individual chromophore test PAI imaging agents. As shown in Image processing algorithms using images to determine i) their sensitivity Fig. 7, we identified a NIR dye (IRDye MATLAB software were used to reduce by comparing their PA yield (PA signal 800CW), which generates a strong PA the spatial variability and enhance the per micro molar solution), and ii) their signal, conjugated it to the PSMA-direct- PA signal intensity throughout the im- spectra shape. Among the five contrast ed A10-3 aptamer and initiated testing age. Each gel type area was considered agents/dyes we investigated, IRDye- in cell cultures. Prostate cancer cells as the ROI for obtaining the averaged 800CW was chosen based on photo- lacking (PC3) and presenting (C4-2) the PA intensity (in future experiments we acoustic spectrum analysis as deter- PSMA cell surface protein were used will determine the signal for individual mined by the highest intensity achieved for testing the PA signal from IRDye- gels, in order to assess reproducibility). relative to the other agents and because 800CW conjugated to the A10.3 aptamer The average PA intensity of the C42 gels the peak absorption is well separated (dye-aptamer) that binds to PSMA. These was ~4 fold higher than either empty from endogenous tissue constituents, two PCa cell lines were incubated with phantoms or PC3 cell containing phan- as shown in Figure 7. NIR fluorescence 4 µM dye-conjugated aptamer, washed toms, and was highly correlated with the imaging was also performed for the five thoroughly and centrifuged. The pellets fluorescence data obtained by scanning contrast agents/dyes and the optical ab- are mixed with agarose, producing a uni- in an IVIS spectrum (Fig. 10D). sorption data correlated to PA spectros- form gel phantom for imaging. As shown copy data as shown in Figure 8. in Fig. 10, we prepared a total 9 agarose Fig. 10. PAI of PCa cells embedded inside aga- gel phantoms: three with agarose-only; rose gels. (A) Digital photograph of imaging platform, with 3 gel types (3 of each, in cylinders). The yellow boxed area was scanned. (B) NIRF (770 nm) IVIS image of the scanned area (C) Photoacoustic image of the scanned area. (D) Average NIRF signal for each cylinder (blue columns) +/-SE, and smoothed averaged PA intensity within an ROI defined by the 3 cylinders of each gel type. RIT scientist collaborates with UR professor at RIT’s Chester F. Carlson 46 colleagues to advance artificial Center for Imaging Science. “Ultra- tissue development sound is a clean way of doing things in the sense that we are not deliv- ering any harmful radiation. We can manipulate the cells and image them without hurting the samples.” Ultrasound standing wave fields are generated in a tissue-culture plate containing endothelial cells embedded in collagen. Endothelial cells Maria Helguera contributes ultra- form the insides of blood vessels. sound techniques, image processing to Pressure from ultrasound standing NIH-funded project wave fields nudges the cells into pre- determined positions. The frequency and intensity of the waves organize the cells and control the density and spacing of cell bands. Samples are then polymerized in an incubator and locked into place. Their close positioning encourages cells to signal to each other and sprout three-dimensional blood vessels. The beginnings of artificial vascular Tissue constructs are visualized with networks created with ultrasound waves. multiphoton microscopy, a technique The frequency and intensity of the waves that captures specimens in three-di- organize the cells into position. mensional sections. Circulating oxygen-rich blood through “Distinct and interesting forma- artificial organs and tissues is a tions can be seen depending on bioengineering conundrum without the frequency and intensity of the an easy answer. The problem is in the ultrasound stationary wave fields,” plumbing. Biomedical teams around Helguera said. “We decided to quan- the world are working on different titatively analyze these three-dimen- solutions to the problem of creating a sional data sets to extract parame- synthetic vascular system. ters characteristic of each exposure Scientists at Rochester Institute of regime.” Technology and the University of Imaging science Ph.D. student Mo- Rochester are looking to ultrasound hammed Yousefhussien developed an technology to create tiny blood image-processing tool for evaluating vessels needed to nourish organs and the structures of the blood sprouts. tissues grown for reconstructive and Third-year imaging science student surgical applications. Amy Becker is modifying the tool to Developing complex vascular systems capture details that will help manipu- with high-frequency ultrasound waves late the sprouts’ growth. Determining is the goal of RIT imaging scientist the preferred direction in which the Maria Helguera ’99 (Ph.D., imaging vessels branch outward will lead to science), and UR’s Diane Dalecki, networks resembling the vascular professor of biomedical engineering, system within an organ. and Denise Hocking, associate profes- “The goal is to design a quantitative sor of pharmacology and physiology. protocol that will allow us to create The UR-led project is funded by the a more complicated structure that is National Institute of Biomedical Im- closer to a real system,” Helguera said. aging and Bioengineering, part of the National Institutes of Health. Helguera provides the team an expertise in ultrasound imaging and image processing through high-frequency ultrasound techniques and quantitative analysis of microscopy images of the tissue samples. “We can use ultrasound in every stage of the process while creating artificial tissue,” said Helguera, an associate

RIT Imaging Science 2014–2015 Annual Report 47 RESEARCH 48RESEARCH

RIT Imaging Science 2013–2014 Annual Report RESEARCH 49 MAGNETIC RESONANCE LABORATORY Laboratory Director's Comments by Dr. Joseph Hornak

The RIT Magnetic Resonance Laboratory is a research and development laboratory devoted to solving real world problems with magnetic resonance. This past year we continued our work on low frequency electron paramagnetic resonance (LFEPR) spectroscopy and imaging which is highlighted in this report.

LFEPR-Imaging One addition to our LFEPR spectrometer this past year was a 3 mm surface coil. A surface coil is a radio frequency transmitter and receiver antenna that can detect the electron paramagnetic resonance (EPR) signal from a surface RESEARCHadjacent to it. This coil enables us to measure non-invasively and non-destruc- tively the EPR and electron magnetic resonance (EMR) signal from paramagnetic and ferri/ferromagnetic materials on a flat surface such as in a painting or illuminated manuscript. By rastering signal bearing surface under the surface coil we are able to image the spatial distribution of the paramagnetic or ferri/ ferromagnetic material on the surface. By placing the surface coil over a specific point on the surface we can record an EPR spectrum of the paramagnetic material in the region and allow spectral identification of a material. We demonstrate this ability by imaging the distribution of EPR signal from the #72 Arial font letters LFEPR stenciled on a vinyl sheet with 2,2-diphenyl-1-pic- rylhydrazyl (DPPH) in epoxy, and the EMR signal from the same letters electro- photographically printed on paper. DPPH has a single, narrow peak in the EPR spectrum, while toner contains a broad non-descript signal. Fixing the magnetic field on the peak and rastering the lettered surface under the surface coil allowed us to produce the following images of the DPPH and toner. Many pigments used in paintings Dr. Hans Schmitthenner, 5. J. Schmitthenner, I. Evans, J.P. 50 and illuminated manuscripts contain a Research Scientist in Hornak, Modular Synthesis of Pep- paramagnetic or ferri/ferromagnetic the Center for Imaging tide-based Single and Multi-modal materials and possess either an EPR Science and Lecturer in Targeted Molecular Imaging Agents. or EMR signal. A non-destructive and the Department of World Molecular Imaging Congress, non-invasive technique which can Chemistry, is working on Honolulu, HI, September 2015. determine the composition and spatial targeted contrast agents distribution of a pigment may be of for magnetic resonance imaging. Publications use to curators and collectors who Lauren Switala has been 1. Y. Qiu, WC.E. Kwok, J.P. Hornak, A need to determine the authenticity and working in the lab for the method of switching the signal in provenance of a work of art, and to art past two years on a an MRI phantom based on trace ion historians who may like to determine variety of projects currents. J. Magn. Reson. 245: 171- the temporal and geographic changes ranging from instrument 176 (2014). in the use and availability of pigments. design and interfacing to 2. Y. Qiu, W. Yao, W-C.E. Kwok, J.P. Staff News examining the LFEPR Hornak, A Circuit for Synchronizing signal from items with cultural heri- External Stimuli and Events to the Amy Becker, a second tage significance. She graduated this Pulse Sequence of a Clinical Mag- year BS Imaging Student year from RIT Magna Cum Laude with a netic Resonance Scanner. Concepts at RIT, has worked on BS in Chemistry. in Magn. Reson. 44B:50-52 (2014). measuring the hydrogen Dr. Nicholas Zumbulyadis, NMR relaxation rates for 3. W.J. Ryan, N. Zumbulyadis, J.P. a retired Research the lipid-water phantom Hornak, The Potential of Low Scientist from Eastman and the targeted contrast Frequency EPR Spectroscopy in Kodak and an expert on agent projects. Studying Pottery Artifacts and Pig- Meissen's Blue and White ments. MRS Proceedings, mrsf13- Wyatt Brown, a third Porcelain continued 1656-pp03-03 doi:10.1557/ year BS Chemistry major working with the lab on opl.2014.708 (2015). at RIT, is working on the identification of ceramics by LFEPR. LFEPR of spectra of paramagnetic complexes Conference Presentations as a function of firing 1. N. Zumbulyadis, W.J. Ryan, L. temperature and instru- Switala, J.P. Hornak, Low Frequency ment characterization. EPR: A Novel, Non-invasive, Spec- Merlin Hoffman, a fourth troscopic Approach to Studying year BS Physics major at Ceramic Objects. Gordon Research Oberlin College, was a Conference: Scientific Methods in summer REU student in Cultural Heritage Research. Newry, the lab. He worked on ME, Jul 27-Aug 4, 2014. the LFEPR spectra of 2. L. Switala, E.I. Hornak, W.J. Ryan, paramagnetic complexes N. Zumbulyadis, J.P. Hornak, A Low as a function of firing temperature Frequency Electron Paramagnetic and instrument characterization. Resonance Spectroscopy Study Of Emma Hornak is a first year The Firing Temperature Of Redart high school student at the Clay. Rochester Academy of Science Harley School. She studied 41th Annual Fall Scientific Paper the LFEPR spectra of iron in Session, Rochester, NY November RedArt clay as a function of 2014. firing temperature. 3. N. Zumbulyadis, L. Switala, W.J. Dr. Joseph Hornak, Ryan, J.P. Hornak, Low Field EPR Professor of Chemistry, in Cultural Heritage Science: the Materials Science and Non-Destructive Characterization Engineering, & Imaging of Large Intact Objects and Spin Science finished his Dynamics at Low Fields. 56th second year as Chairman Experimental NMR Conference, of RIT’s Graduate Council. Asilomar, CA, April 2015. William Ryan, an adjunct 4. L. Switala, E.I. Hornak, W.J. Ryan, faculty member in the N. Zumbulyadis, J.P. Hornak, A Low Department of Chemis- Frequency Electron Paramagnetic try, is working on a Resonance Spectroscopy Study Of LabView interface for the The Firing Temperature Of Redart lab’s low frequency Clay. 60th Rochester Section of the electron paramagnetic American Chemical Society Under- resonance (LFEPR) spectrometer. graduate Research Symposium, Geneseo, NY, 2015.

RIT Imaging Science 2014–2015 Annual Report 51 RESEARCH 52RESEARCH

RIT Imaging Science 2014–2015 Annual Report RESEARCH 53 LABORATORY FOR ADVANCED OPTICAL FABRICATION, INSTRUMENTATION AND METROLOGY (AOFIM) Laboratory Director’s Comments By: Director Dr. Jie Qiao

The laboratory for Advanced Optical Fabrication, Instrumentation and Metrology (AOFIM) at the Chester Center for Imaging Science conducts research in the areas of...

(1) Ultrafast Lasers for advanced optics fabrication including integrated photonics and freeform optics (2) Optical Metrology for phase imaging and wavefront sensing, which can be used for charactering astronomical telescopes, laser beams or retina imaging. (3) Coherent phasing of segmented large-scale gratings and adaptive optics RESEARCHfor next-generation telescopes or laser systems. The vision for the AOFIM Laboratory is to conduct frontier research in the aforemen- tioned three areas and provide coherent, integrated, and closed-looped research and educational experiences to directly engage undergraduate and graduate students to work on a diverse set of real-world projects in Optical Science and Engineering. The mission is to train young researchers on how to frame research problems, acquire relevant knowledge, test hypotheses, demonstrate concepts, and provide solutions. Dr. Qiao, Associate Professor, leads the AOFIM Laboratory where two PhD students, three undergraduate students, one high school intern, one visiting scholar, and one post-graduate researcher have been conducting their research projects during the past academic year. Research projects: 1. Ultrafast Lasers for advanced optics fabrication including integrated photonics and freeform optics Ultrafast lasers are transforming the way that optics are being manufactured. They provide a more cost effective, time effective, and environmentally friend- ly solution for fabricating integrated photonics, freeform optics, and micro optics, and optoelectronic packaging. AOFIM Lab conducts fundamental research on the theoretical modeling and experimental demonstration of ultrafast lasers and matter interaction. This project investigates the phenomena, mechanism, systems, and metrology of non- thermal ultrafast-laser welding, polishing, and cutting of various optical materials, which will allow for the connection of different optics/glass materials and enable more compact, and better integrated photonics circuits with different functions.

Project team Prof. Qiao was working with Lauren Taylor (Ph.D) and Daniel Morgen (NSF REU student) on a laser beam delivery system for ultrafast lasers for advanced optics and integrated photonics fabrication.

2. Optical Differentiation wavefront sensor (OWDS) for freeform metrol- 54 ogy and phase imaging High-resolution wavefront sensors are of great interest for laser engineering and astronomy. The optical differentiation wavefront sensor allows for high signal-to-noise ratio broadband characterization of the Third-year PhD student Lauren Taylor spatial phase of optical waves. When presented her research on laser polishing a filter with a field transmission that experiments at the 2015 University Tech- is linear with respect to a spatial nology Showcase by CEIS. Lauren has also coordinate is located in the far field been working on the theoretical interaction of the optical wave, the spatially modeling to determine the optimum laser resolved wavefront slope along that processing parameters. coordinate can be recovered from the near field of the filtered wave. The complete spatially resolved wavefront is recovered from a set of two Matthieu Chalifour, a high school intern from orthogonal wavefront-slopes maps. Brighton High School, worked on “The Anal- Experimental reconstruction of the ysis and Comparison of Single to Noise Ratio wavefront of a HeNe laser with Cr-on- for Shack-Hardmann and ODWS sensors in glass pixelated binary transmission the summer of 2015. He presented his work filters has been demonstrated. The at a group meeting. trade-off between pixel size, filter size, beam parameters, and wave- NSF-REU student Daniel Morgen, rising se- front reconstruction accuracy has nior from the Institute of Optics, University been studied. of Rochester conducted summer research on “Optical design and performance analysis of The PI Dr. Qiao has provided an a trepanning system for laser material pro- invited talk “Performance analysis of cessing” at the AOFIM lab and he presented an optical differentiation wavefront the work to his fellow REU students. sensor and its application to the metrology of freeform optics” at the Im- Dr. Jun Qiao, Pro- aging and Applied Optics Congress in Prof. Qiao, Zachary Mulhollan (CIS under- fessor of College of June, 2015 in Arlington, VA. Materials Science graduate ), and Dr. Aaron Schweinsberg (post-graduate research assistant) were and Metallurgy, the Project team University of Science working on the experimental demonstration and Technology of the ODWA sensor Liaoning, China, has joined AOFIM lab as a visiting scholar since May 2015. He brings in material science expertise to this ultrafast lasers and material interaction project and he will be conducting collaborative research in the following academic year.

Zachary Mulhollan, a rising senior at the Students across levels have active interac- Chester Center for Imaging Science, RIT tions: conducted summer research “the Modeling and Tolerance Analysis of an Optical Differ- Matthieu Chalifour (high school intern) and entiation Wavefront Sensor” at the Klemens Gowin (NSF-REU student) were work- AOFIM lab. He presented his work at a ing together on a numerical model MATLAB. group meeting. The ODWS showed excellent self-con- sistency using two sets of orthogonal NSF-REU student Klemens Gowin, a rising filters, achieving a difference of less junior from the Institute of Optics, University than 1/20th of optical wave (one of Rochester conducted summer research on wave: 633nm), as shown in Figure 1. “Freeform Surface Generation Using a Deform- able Mirror and Grating Compressor Modeling” at the AOFIM lab. He also did a literature review on Freeform Metrology and presented his work to his fellow REU students.

RIT Imaging Science 2014–2015 Annual Report Journal Publication (2014 - 2015) (CEIS), University of Rochester, (1) C. Dorrer, A. Consentino, D. Irwin, Rochester, NY, 16 April, 2015 (poster) 55 J.Qiao, J. Zuegel, OPCPA Front End and GRANTS Contrast Optimization for the OMEGA EP Kilojoule, Picosecond Laser," ac- (1) PI, NYSTAR CEIS, Development cepted to appear in Journal of Optics, & Investigation of an Integrated 2015, Article reference: JOPT-101628 Laser-based Optics Polishing Sys- Figure 1. Left: measured wavefront using a tem, July 01, 2014 - July 30, 2015, pair of far-field filters with orthogonal linear Conference Publication / Presen- $30,000, Awarded transmissions in horizontal and vertical tation (proceeding / summary / (2) PI, NYSTAR CEIS, Ultrafast laser directions. Middle: measured wavefront using abstract) (2014 - 2015) Polishing for Additive Manufactur- a pair of far-field filters with linear transmis- (1) (Invited) J. Qiao, A. Travinsky, ing, July 01, 2014 - June 30, 2015, sions in two orthogonal 45-degree directions. and C. Dorrer, "Performance Analysis $30,000, Awarded Right: The difference of the two sets of mea- of an Optical Differentiation Wave- surements is less than 1/20th of optical wave. (3) PI, OptiPro Systems LLC, Investi- front Sensor and its Applications to gation of an Integrated Laser-based The progress of this work has also the Metrology of Freeform Optics," Optics Polishing, July 01, 2014 - June been presented at the 2015 SPIE Pho- in Imaging and Applied Optics 2015, 30, 2015, $30,000 cash plus $30,000 tonics West conference (Jan. 2015 San OSA Technical Digest (online) (Opti- equipment, Awarded Francisco, CA) and 2015 CLEO/OSA cal Society of America, 2015), paper FTh2B.2. (abstract only) (4) PI, the Dean's Research Initiation conference (May 2015, San Jose, CA). Grant (COS), An innovative optical dif- 3. Coherent phasing of segmented (2) J. Qiao, A. Travinsky, and C. Dorrer, ferentiation wavefront sensor based large-scale gratings for next-gener- "Optical differentiation wavefront sen- on binary pixelated transmission ation telescopes or laser systems. sor based on binary pixelated transmis- filters, August 01, 2014 - August 01, sion filters," in CLEO: 2015, OSA Tech- 2015, $15,000, Awarded This line of research effort builds on nical Digest (online) (Optical Society of Dr. Qiao’s experience and publica- America, 2015), Paper STu2N.1. (5) PI, ADVANCE RIT/National Sci- tion credentials in the high-energy ence Foundation (NSF), Advance RIT laser field. As part of this effort, Dr. (3) J. Qiao, A. Travinsky, O. Ding, D. Connect Grant (RIT Internal Grant), Qiao has finished the design and Dang, C. Dorrer, Optical differentia- $9,600, Awarded tion wavefront sensor based on binary modeling of an innovative deform- (6) PI, NASA-National Aeronautics able–grating which can be used for pixelated transmission filters, SPIE Proceedings, Paper 9356-6, SPIE Pho- and Space Administration, Astrophys- both lasers and astronomical tele- ics Research and Analysis (APRA) scopes. She has submitted a journal tonics West, San Francisco, CA, 7–12 Feb., 2015 Program: Investigation and Demon- paper to Optics Express. As the PI, stration for the Coherent Phasing Dr. Qiao Dr. Qiao has conceived the (4) J.Qiao, X.Liu, and J. Papa “Actua- of Transmissive Diffractive Optical concepts and developed the proposal tor-Position Optimization and Holo- Elements (DOE) for Large Lightweight “Investigation and Demonstration graphic Correction for Large-Scale Space Telescopes, submitted on 20 for the Coherent Phasing of Trans- Deformable Gratings”, 2014, Inter- March, 2015, $999,720, Pending missive Diffractive Optical Elements national Ultrahigh Intensity Lasers (DOE) for Large Lightweight Space Conference, Paper Tu2-P65, Goa, In- Telescopes”. This proposal was sub- dia, 12–17 October, 2014 (one-page mitted to the NASA APRA program abstract) in March 2015. The objective of the (5) J.Qiao, X.Liu, “Design and Actua- proposal is to apply transmissive tor-Position Optimization for a Large- diffractive optical elements (DOE) Scale Adaptive Grating”, Frontiers into UV telescope systems and to in Optics, Paper FTu2C4, Tucson, study the coherent phasing of DOE Arizona, 19–23 October, 2014 subapertures to support continued development of lightweight space Additional Conference Presentations telescopes providing large collecting area and high angular resolution. (6) J. Qiao, A. Travinsky, G. Ding, D. The proposed work include modeling Dang, C. Dorrer, “A differential wave- and laboratory demonstration of the front sensor based on binary pixelat- coarse and fine phasing of a nominal ed transmission filters”, 2014 SPIE segmented powered DOE telescope. NASA Mirror Technology Workshop, This proposal includes collabora- Albuquerque, New Mexico, 18–20 tions with co-PI Dr. Joel Kastner, Nov, 2014 (poster) Exelis Inc. (now Harris Corp.), NASA (7) L. Taylor, J. Qiao, Ultrafast laser Goddard Space Flight Center, Akti- polishing offers flexible, high-preci- wave LLC, and Patchwork Optical sion finishing for conventional and Consulting LLC. additive manufacturing, University Technology Showcase, Center for Emerging and Innovative Sciences RESEARCH 56RESEARCH

PhD candidate Garreth Ruane with NASA Administrator Charles Bolden.

RIT Imaging Science 2014–2015 Annual Report RESEARCH 57 OPTICS

Laboratory Director’s comments by Dr. Grover Swartzlander

The laboratory enjoyed a productive year developing diverse areas related to imaging and optical physics.

The work was supported by the National Science Foundation, NASA, and others. The research spanned the areas of exoplanet research, radiation pressure for future space propulsion, computational imaging, and future space telescopes. Prof. Swartzlander leads a team of three graduate students (listed by RESEARCHseniority: Alexandra Artusio-Glimpse, Garreth Ruane, and Xiaopeng Peng), one visiting scholar (Dr. Lingyu Wan, Guangxi University), one undergraduate student (Christian Cammarota), various RIT collaborator faculty (Dr. Alan Raisanen, CAST; Dr. Michelle Chabot, SoPA; Dr. Mario Gomes, KGCE). External collaborators included scientists from Jet Propulsion Laboratory, US Naval Re- search Laboratory, NASA Marshall Space Flight Center, University of Liege, and University of Bristol. He also continues as Editor-in-Chief of the Journal of the Optical Society of America—B, which will begin celebrating a 100 year history Alexandra Artusio-Glimpse in the coming year. Research highlights includes a collaboration with the Jet Propulsion Laboratory to explore the shaping of a cloud of small reflective bodies into a large space telescope. Using computational imaging techniques, the RIT team demonstrated imaging using craft store glitter. The team continues to explore how space based lasers may be used to orient the reflective optics with non-contact forces. A separate collaboration with a European Space Agency group based in Belgium is developing a diamond optical vortex coronagraph to help astronomers image exoplanets with large ground based telescopes. An additional collaboration with NRL seeks to use computational imaging schemes to protect optical sensors from damaging laser beams. Projects (additional RIT team members) Christian Cammarota - Radiation Pressure Torsion Oscillator (Artusio-Glimpse, Raisanen, Chabot, Gomes) - Optical Vortex Scatterometer (Ruane, Wan) - Optical Vortex Coronagraph (Ruane) - Granular Space Telescope (Peng, Ruane, Artusio-Glimpse) - Radiation Protection (Ruane) Publicity Highlights: - National Public Radio All Things Considered, Joe’s Big Ideas - YouTube, Light Matters - The Economist, Space Telescopes, A Glittering Prize - RIT Press Release Xiaopeng Peng - JPL Press Release 58

This image shows white light reflected off of a glitter mirror onto a camera sensor. Researchers tested this in a laboratory as part of the concept of "Orbiting Rainbows," a low-cost solution for space telescope mirrors. Credit: G. Swartzlander/Rochester Institute of Technology

Patent Awarded: ceedings (2014) nique for extracting spatial - Optical lift apparatuses and - “An Overview of Solar Sail Pro- information from unresolved methods thereof (2014) pulsion within NASA,” Johnson, sources,” Ruane, Kanburapa, Swartzlander, Artusio-Glimpse, Han, Swartzlander, arXiv (2014) Recent Publications: chapter in Advances in Solar Glitter Cloud May Serve as - “Nodal areas in coherent Sailing (2014) Space Mirror beams,” Ruane, Swartzlander, - “Thrust Efficiency on an Ide- What does glitter have to do with Slussarenko, Marrucci, Dennis, alized Deformable Sail,” Ar- finding stars and planets outside Optica (2015) tusio-Glimpse, Swartzlander, our solar system? Space telescopes - “Reducing the risk of laser dam- chapter in Advances in Solar may one day make use of glitter-like age in a focal plane array using Sailing (2014) materials to help take images of new linear pupil-plane phase ele- - “Rocking motion of an optical worlds, according to researchers at ments,” Ruane, Watnik, Swartz- wing: theory,” Artusio-Glimpse, NASA's Jet Propulsion Laboratory in lander, Applied Optics (2015) Schuster, Gomes, Swartzlander, Pasadena, California. - “Nonlinear response and stability Applied Optics (2014) Standard telescopes use solid mirrors of a 2D rolling semi-cylinder during - “Vortex-phase filtering tech- to image far-away objects. But the optical lift,” Schuster, Gomes, nique for extracting spatial large, complex mirrors needed for as- Artusio-Glimpse, Swartzlander, information from unresolved tronomy can be quite expensive and dif- Nonlinear Dynamics (2015) sources,” Ruane, Kanburapa, ficult to construct. Their size and weight Han, Swartzlander (2014) also add to the challenges of launching - “Image restoration from a se- a space telescope in the first place. quence of random masks,” Peng, - “Zernike amplitude pupil apod- Ruane, Artusio-Glimpse, Swartz- ization for vortex coronagraphy A concept called Orbiting Rainbows lander, SPIE Proceedings (2015) with obscured apertures,” Ru- seeks to address these issues. Re- ane, Swartzlander, arXiv (2014) searchers propose using clouds of - “Mirror swarm space telescope,” reflective glitter-like particles in place Peng, Swartzlander, IEEE Pro- - “Vortex-phase filtering tech- of mirrors to enable a telescope to

RIT Imaging Science 2014–2015 Annual Report trapping and manipulation have con- with maximum sensitivity to one spe- tributed to the Orbiting Rainbows con- cific frequency. 59 cept. The original idea for a telescope Orbiting Rainbows has not yet been based on a laser-trapped mirror was demonstrated in space. For a test in proposed in a 1979 paper by astrono- low-Earth orbit, the researchers would mer Antoine Labeyrie at the College de deploy a telescope with a small patch France in Paris. of particles, no larger than a bottle Now, the Orbiting Rainbows team is cap, to show that it can be trapped trying to identify ways to manipulate and shaped to reflect light. The next and maintain the shape of an orbiting step would be to make many of these Researchers made a mirror surface out of cloud of dust-like matter using laser patches and synthesize an aperture glitter to test the idea of using a cloud of pressure so it can function as an adap- with which to do imaging. reflective particles as a space telescope mir- tive surface with useful electromagnet- ror. They took images of two light sources The project represents a new appli- ic characteristics, for instance, in the cation of "granular matter," materials using this mirror in a laboratory at Roches- optical or radar bands. ter Institute of Technology. Credit: G. Swart- such as dust grains, powders and zlander/Rochester Institute of Technology. Because a cloud of glitter specks is not aerosols. Such materials are very light, a smooth surface, the image produced can be produced at low-cost and could view stars and exoplanets. The tech- from those specks in a telescope will be useful to the space exploration nology would enable high-resolution be noisier—with more speckled distor- community. In this particular project, imaging at a fraction of the cost. tion—than what a regular mirror would the "glitter" may be tiny granules of "It's a floating cloud that acts as a generate. That's why researchers are metallic-coated plastic, quartz or some mirror," said Marco Quadrelli from developing algorithms to take multiple other material. images and computationally remove Orbiting Rainbows is currently in JPL, the Orbiting Rainbows principal the speckle effect from the glitter. investigator. "There is no backing Phase II development through the structure, no steel around it, no hing- To test the idea, co-investigator Grover NASA Innovative Advanced Concepts es; just a cloud." Swartzlander, an associate professor at (NIAC) Program. It was one of five the Rochester Institute of Technology technology proposals chosen for con- In the proposed Orbiting Rainbows in New York, and his students spread tinued study in 2014. In the current system, the small cloud of glitter-like glitter on a concave lens in the labora- phase, Orbiting Rainbows researchers grains would be trapped and manipu- tory. His team used lasers to represent are conducting small-scale ground lated with multiple laser beams. The the light from a double star system. experiments to demonstrate how trapping happens because of pressure They pointed the speckled mirror at granular materials can be manipulated from the laser light -- specifically, the the simulated stars, then used a cam- using lasers and simulations of how momentum of photons translates into era to take pictures. With many expo- the imaging system would behave in two forces: one that pushes particles sures and lots of processing, an image orbit. away, and another that pushes the of the two "stars" emerged using the NIAC is a program of NASA's Space particles toward the axis of the light glitter mirror. beam. The pressure of the laser light Technology Mission Directorate, located coming from different directions "This is a major achievement," Qua- at the agency's headquarters in Wash- shapes the cloud and pushes the small drelli said. "This demonstrates a highly ington. JPL is managed by the California grains to align in the same direction. controlled experiment in which we Institute of Technology for NASA. In a space telescope, the tenuous were able to do imaging in the visible For a complete list of the selected cloud would be formed by millions of light spectrum." proposals and more information about grains, each possibly as small as frac- The technology could be used more NIAC, visit: tions of a millimeter in diameter. easily for radio-band signals. Because http://www.nasa.gov/niac Such a telescope would have a wide the wavelength is so much longer (about one centimeter, compared to For more information about the Space adjustable aperture, the space through Technology Mission Directorate, visit: which light passes during an optical nanometers in visible light), the mirror or photographic measurement; in grains don't have to be as precisely http://www.nasa.gov/spacetech controlled or aligned. This opens up fact, it might lead to possibly larger Glitter Mirror apertures than those of existing Earth science applications such as space telescopes. earthquake detection and remote sens- Researchers made a mirror surface ing of water and other phenomena. out of glitter to test the idea of using a It would also be much simpler to JPL's Darmindra Arumugam is investi- cloud of reflective particles as a space package, transport and deploy, than a gating possible mechanisms for remote telescope mirror. They took images of conventional space telescope. sensing with Orbiting Rainbows. two light sources using this mirror in "You deploy the cloud, trap it and The JPL optical design team, including a laboratory at Rochester Institute of shape it," Quadrelli said. Scott Basinger and Mayer Rud, has Technology. Credit: G. Swartzlander/ Rochester Institute of Technology Nature is full of structures that have been working on the adaptive optics light-scattering and focusing prop- techniques that would be needed by an erties, such as rainbows, optical Orbiting Rainbows telescope. So far, phenomena in clouds, or comet tails. the team has been exploring reflective, Observations of these phenomena, and refractive and diffractive versions of a recent laboratory successes in optical telescope based on Orbiting Rainbows, RESEARCH 60RESEARCH

A high-resolution TEM image of a maghemite (γ-Fe2O3) nanocrystal. The individual columns of atoms can be seen in the crystal and the spacing between columns is indicated. The inset is a Fast Fourier Transform of the image and represents the electron diffraction pattern of the nanocrystal. The spacing between spots allows the determination of crystal structure and confirms that the crystal is maghemite.

RIT Imaging Science 2014–2015 Annual Report RESEARCH 61 NANOIMAGING RESEARCH LABORATORY Laboratory Director’s Comments By Professor Rich Hailstone

The NanoImaging Laboratory has continued to grow in the diversity of materials imaged at the micro and nanoscale as part of collaborations with other RIT research groups and our industrial partner.

Phage Imaging. Phages (or more correctly bacteriophage) are biomaterials that have a specific affinity for bacteria. They inject their genetic material into the bacterium following infection and can cause the bacteria to synthesize phage nucleic acids and proteins. This is a joint project with Professor Julie Thomas, RESEARCHa phage researcher in the School of Life Sciences, to support her research, as well as the phage workshop she teaches in the Spring semester. Currently we are focused on determining the optimal pathway to preparing the phage for imaging in the transmission electron microscope (TEM). It is required to preserve the phage material in its natural state, but still be compatible with the vacuum of the electron microscope. Figure 1 shows both an overview of the phages, as well as a magnified view of one phage, showing its capsid and the tail used to penetrate the bacterium. Contrast is enhanced by staining the material with phosphotungstic acid. At this point the preparation is less than optimal, as the capsids are too dark.

Imaging Polymer Blends. This is a project with Professor Carlos Diaz-Acosta in Packaging Science. He is blending three biodegradable polymers in effort to make the packaging more “green,” but still possess the requisite physical properties for the intended packaging. The goal of the project is to connect the nano- and micro-structures seen in the electron microscope with macroscopic physical properties. Figure 2 shows TEM images of the polymer blend, indicating at least two phases—the darker structures and the more or less continuous background. Fig. 2. Left, TEM image at low magnification of the polymer blend. The darker “stringy” structures are from the lacey carbon support that the polymer cross section is resting on. Right, a higher magnification view of the polymer blend.

Fig. 1. TEM images of wild type phage. We have also looked at these blends Analysis of Pt Particles in SrTiO3. Pro- 62 in the scanning electron microscope fessor Michael Pierce (Physics) and (SEM) and an example is given in Fig. MS&E MS student Joshua Goodman 3. These cross-sections of the polymer have synthesized Pt nanoparticles on blend are prepared by a freeze-fracture a SrTiO3 substrate. After annealing technique. On the left is an indication in H2, the particles are about 100 nm that there are at least two phases in height and widths varying between present because of the brighter pseu- 10’s and 100’s of nanometers, as do-circular structures on the darker shown in Fig.5. The figure on the left background. On the right is a higher indicates most nanoparticles have magnification view in which the con- a large bright area and one or more trast has been enhanced. This image edges have a darker area. The latter may indicate there are three phases could indicate the nanoparticles have present. The larger white structures or a varying composition. The right-hand the very dark features where the larger image indicates that the nanoparti- white structures once were located cles have a height around 100 nm. comprise one phase. But the gray background may also indicate two phases— small darker features on a lighter continuous background. This possibility is being explored in future work.

Laser-Aided Manufacturing. One of Pro- fessor Jie Qiao’s research group’s in- Fig. 5. SEM images of Pt nanoparticles on terest is in using ultrafast lasers to en- SrTiO3 substrate. Left: Image looking per- hance manufacturing capabilities. The pendicular to nanoparticles. Right: Image CIS Microscopy Facility is being used with sample tilted 70 degrees to normal to to characterize laser-matter interac- reveal sides of the nanoparticles. tions. Using samples prepared by CIS PhD student Lauren Taylor, we imaged of the electron trajectory. From these the material in the SEM. An example is results it appears that the synthesis of shown in Fig. 4. Circularity and depth the Pt nanoparticles and/or the subse- of the laser-produced hole are import- quent anneal in H2 is also disturbing ant attributes in this test and the SEM the substrate. images are critical in quantifying the laser-matter interaction. Alkali-Activated Materials as Substi- tutes for Portland Cement. The world uses large volumes (4.18 billion metric tons in 2015) of Portland cement every year in the manufacture of concrete, Fig. 4. SEM images of a Ti alloy which has with huge effects on energy and green- been exposed to a high-power laser. Top: house gas production. In the latter case 500k pulses over 15 s. Bottom: 2000k it is estimated that 5% of total anthro- pulses over 15 s. pogenic emissions of CO2 come from cement production. Alkali-activated To check the composition of the materials are cementitious materials nanoparticles we used EDS in linescan that can be produced from industrial mode. EDS (energy dispersive spectros- by-products, with less impact on energy copy) uses the x-rays emitted by the and greenhouse gas emissions. sample when exposed to the high-energy beam electrons. The energy of these x-rays is characteristic of the element that emitted them, so this becomes a way to spatially determine the elemental composition of the nanoparticles. As shown in Fig. 6, the bright area of the particle is Pt, but the darker area has the composition similar to the substrate. The Sr signal does not go to zero when the beam is on the Pt phase of the Fig. 3. Top, SEM image at low magnification particle because the beam electrons of the polymer blend. Bottom, a higher mag- are energetic enough to penetrate Fig. 6. Linescan elemental analysis of a Pt nification view of the polymer blend with through the Pt and excite Sr x-rays in nanoparticle along the direction indicated contrast enhanced. the substrate. This ability of the beam by the yellow arrow. Superimposed on the electrons to penetrate the Pt phase was linescan arrow are the x-ray signals from Sr confirmed by Monte Carlo simulations (green) and Pt (blue).

RIT Imaging Science 2014–2015 Annual Report The purpose of this joint project with Professor Varela in Mechanical 63 Engineering and CIS MS student Najat Alharbi is to explore possible correla- tions of the nano- and micro-structure of the produced cements with the physical properties of the material, especially compressive strength. Currently, we are studying blast-fur- nace slag as a substitute. The slag material is predominately composed of the oxides of calcium and silicon, with minor constituents of the oxides of aluminum and magnesium. Fig.8. A comparison of pore distribution as measured from TEM images, left, and from the One of the possible factors affecting standard BET measurement, right. the physical properties of cement is the pore size and its distribution. sured with N2 adsorption-desorption obtained by omitting one of the raw Traditionally this property is mea- in an experiment called “BET.” As an material components. The total crack alternative, we directly measured length of the control measured over the diameter of the pores using TEM five randomly chosen fields of view images of the material such as those was more than twice that for the im- shown in Fig. 7. A comparison of proved version. pore size distribution measured by the two techniques is shown in Fig. 8 and suggests that the two techniques are measuring a similar property. The advantage of using TEM is that it images the pores directly, rather than indirectly as with the BET method. Another aspect of the cement material is the extent of cracking in the cured material. This property could be correlated with the material’s compressive strength. Figure 9 is a comparison of SEM images of the control material and an improved version with higher compressive strength

Fig. 10. Top, SEM image of a crack wall (right-hand side) and the adjoining surface (left-hand side). Numbers indicate regions where x-ray spectra were collected. Middle, EDS spectrum of the crack wall at the four different locations indicated in the SEM Fig.7. TEM images of slag-based cement. The image. Bottom, EDS spectra of surface wall lighter shapes are examples of voids or pores Fig.9. SEM images of the cement material at the four different locations indicated in in the material. Their diameter was measured after curing. Left, control, right, improved the SEM image. Au and Pd are from the sput- to produce the histogram in Fig. 8. version with higher compressive strength by ter-coated film applied to make the sample omitting one of the raw materials. electrically conductive. We are also interested in studying the Cerium Dioxide Nanoparticles, Grants and Contracts 2014-2015 elemental distribution in the cement, Japanese Pat. No. 5462627 64 Cerion Advanced Materials, $28k and in particular within the cracks. 4. A. G. DiFrancesco, R. K. Hail- Figure 10 shows an SEM image of a stone, K. J. Reed, and G. R. Prok Other Income 2014-2015 crack and the adjoining surface f one (2014). Cerium-Containing Microscopy Facility $30k of the improved versions. The num- Nanoparticles, U. S. Pat. No. bered circles indicate where x-ray 8,883,865 B2 spectra were collected, and these Facilities are shown in the lower part of the 5. R. Hailstone, K. Reed, T, Allston Microscopy: JEOL 6400V scanning figure for both the crack wall and the (2015) Oxidation Resistant, electron microscope with a LaB6 elec- surface. The most obvious difference Structured, Hexagonal Plate tron gun and energy dispersive X-ray is that of the O peak. It is much lower Nanoparticle Conductive Ink. RIT analysis for elemental detection. JEOL at the surface, as well as being more Invention Disclosure 2015-007 2010 transmission electron micro- variable between the different loca- 6. R. Hailstone, K. Reed, T, Allston scope with energy dispersive X-ray tions. At this time it is not clear what (2015) Oxidation Resistant, analysis for elemental detection. JEOL this might mean. Structured, Spherical Nanoparti- 100CX II transmission electron micro- cle Conductive Ink. RIT Invention scope. Reichert Jung ultramicrotome Publications, Patent Applications, Disclosure 2015-008 for cross-section preparation. Nikon Patents Issued, Invention Disclo- 9000 Film Scanner. sures, and Conference Presentations 7. R. Hailstone, K. Reed, T, Allston (2015) Oxidation Resistant, 2014-2015 Structured, Cubic Nanoparticle 1. A. G. DiFrancesco, R. K. Hail- Conductive Ink. RIT Invention stone, K. J. Reed, and G. R. Prok Disclosure 2015-009 (2014). Cerium-Containing 8. R. Hailstone, K. Reed, T, Allston, Nanoparticles, U. S. Pat. App. A. Cannella, S. Reed (2015). 14/537,161 Facile, Low Temperature Synthe- 2. A. G. DiFrancesco, R. K. Hail- sis of Metallic Nanoparticles. RIT stone, K. J. Reed, and G. R. Prok Invention Disclosure 2015-019. (2014). Cerium-Containing Nanoparticles, U. S. Pat. App. 14/537,993 3. A. G. DiFrancesco, R. K. Hail- stone, A. Langner, and K. J. Reed (2014). Method of Preparing

RIT Imaging Science 2014–2015 Annual Report 65 RESEARCH 66RESEARCH

Dr. Dube and undergraduate researcher Cailtin Kavanaugh discuss specific results of the neural network forecasts of space weather storms.

RIT Imaging Science 2014–2015 Annual Report RESEARCH 67 LABORATORY FOR SPACE WEATHER ALERT TECHNOLOGIES ASTRONOMICAL IMAGING

Laboratory Director’s Comments By Dr. Roger Dube

The Space Weather lab develops technologies to ensure sustainable colonies on the Moon, Mars and other interplanetary bodies by developing advanced early warning systems that forecast the arrival of deadly space weather storms at the colony location.

Severe space weather storms can also bring identify precursors to the storms so that ing of students in the local chapter of the down the power grid and disable or even we can get an early warning. Matthew American Indian Science and Engineer- destroy satellites orbiting around Earth due Murphy, a master’s student in Imaging ing Society (AISES), provided multiple to the high electromagnetic pulse that often Science, is completing his thesis using opportunities to attract members of vari- RESEARCHaccompanies storms of this nature. In order this tool. His work has validated the ous underrepresented minority groups to to assure the survivability of a human colony ability of the AI to correctly model the the project. In particular, during the past on the moon or Mars, there remains a critical functional behavior of these storms as a 3 years, the project has included: need to develop a predictive capability that function of certain parameters. His work - 3 Native Americans will monitor various precursor events on the has duplicated published results of space sun and develop a high-confidence predic- weather research with high accuracy. - 3 African Americans tion that a coronal mass ejection will erupt Our next phase will be the extension of this - 5 women at a future time. Moreover, such a system tool to identifying other parameters that should include forecasting awareness of can act as reliable indicators for storms. Press Coverage planetary and spacecraft positions so that Since its inception, the project has enjoyed warnings may be issued to appropriate Staff human operations in multiple places within a good deal of press coverage, ranging from the solar system. The amount of advanced One full time faculty member, Dr. Roger front-page local newspaper coverage to warning is clearly important, since the first Dube of Imaging Science, anchors the Space Coalition and Mars Daily websites. wave of radiation (X-rays) and very high-en- lab. With earlier seed funding by NASA, Democrat and Chronicle—front page: ergy particles (> 100 mev protons) will the Space Weather lab works closely with http://rocnow.com/article/lo- strike the Earth (for example) eight minutes NASA on advanced predictive algorithms cal-news/20109130331 and Mars 12 minutes after eruption. The ex- that monitor solar data for signatures that tent to which a reliable predictor with high can reliably predict a subsequent eruption. Azo Sensors: http://www.azosensors.com/ Details.asp?NewsID=130 confidence can be developed that provides Research in the lab has been and con- in excess of hours of advanced notice before tinues to be conducted by: Mars Daily.com: (we didn’t know there was a CME eruption will determine how much such a thing!) http://www.marsdaily.com/ • Matthew Murphy, MS candidate to preparation time Earth and a colony on Mars reports/Early_Warning_System_Would_Pre- graduate in the summer of 2015 have before being struck by the first wave of dict_Space_Storms_on_Mars_999.html radiation. This early warning should provide Summer REU students have included: any affected populations to take protective PlanetStewards.net: http://planetstewards. actions, thereby assuring survivability of • Alicia Lazore net/a919727-early-warning-system-would- predict-space.cfm their effort. • Alexis LaBoy The Space Coalition: http://spacecoalition. Today we have about a 30-minute warning • Ivana Molina before such a high speed storm strikes the com/blog/nasa/danger-mars-colonists- earth and our infrastructure. This is insuffi- • Stephen Chow need-for-predicting-space-storms cient time to prepare spacecraft orienta- • Andrew Ferris PR Web: http://www.prweb.com/releases/ tions, protect astronauts in interplanetary • Caitlin Kavanaugh RIT_Roger-Dube/space_storms_Mars/pr- travel, and, ultimately, to protect colonists web3808924.htm on the moon or Mars. Our work is directed Outreach towards extending this warning time. Softpedia News: http://news.softpedia.com/ Dr. Dube’s role as a Native American fac- news/How-to-Forecast-Space-Storms-on- Our current research employs an artificial ulty member, combined with his mentor- Mars-138729.shtml intelligence (AI) system that we hope will RESEARCH 68RESEARCH

Collages before imaging by infrared thermography at Thermal Wave Imaging, Ferndale MI, May 2014

RIT Imaging Science 2014–2015 Annual Report RESEARCH 69 HISTORICAL MANUSCRIPT IMAGING Laboratory Director’s Comments By Dr. Roger Easton

In the past year, the Laboratory for Historical Manuscript Imaging was engaged in a wide variety of research efforts involving both new manuscripts and new technologies.

The success achieved in these tasks has led to invitations to image more new items in the coming year. Dr. Roger Easton collaborates with faculty and staff at other institutions, including Dr. Gregory Heyworth of the University of Mississippi, Michael Phelps of the Early Manuscripts Electronic Library, and RESEARCHDr. Daniel Stoekl Ben Ezra of the École Pratique des Hautes Études. The students participating in this work include David Kelbe, Liz Bondi, and Kevin Sacca. In May 2014, Dr. Easton traveled with Dr. Heyworth to Thermal Wave Imaging in Ferndale MI (near Detroit) to use infrared thermography techniques to image a pair of collages evidently made by Juan Gris. The goal was to find indications of hidden writing. In the summer of 2014, Dr. Easton and Liz Bondi traveled to Vercelli, Italy to Dr. Gregory Heyworth and Dr. Roger Easton participate with the team that imaged the Codex Vercellensis (also known as Co- with imaging system in the Vatican Library, dex a) at the Museo del Tesoro del Duomo; the team also included Dr. Heyworth, August 2014 Michael Phelps, Dr. Keith Knox (then of the USAF Research Lab on Maui), Dr. Ira Rabin of the Technische Universität Berlin, Kenneth Boydston of Megavision, Inc., and several undergraduate honors students from the University of Mis- sissippi. This significantly damaged long manuscript (more than 600 leaves) is believed to be the earliest manuscript of the Old Latin Gospels, dating from the late fourth century. Immediately following, Dr. Easton traveled to Rome along with Dr. Knox, Ken Boydston, and Michael Phelps, where they met with David Kelbe for imaging and processing of a palimpsest in the collection of the Vatican Library. In August, Dr. Easton participated in imaging of a world map by Henricus Martellus Gemanus in the collection of the Beinecke Rare Book and Manuscript Library of Yale University. This project, funded by the National Endowment for the Humanities and led by map scholar Chet Van Duzer, included Dr. Heyworth, Leaf from the Codex Vercellensis in place for Ken Boydston, and Michael Phelps. This large map (approximately 2 m wide imaging at the Museo del Tesoro del Duomo and 1.25 m tall) of the world was made c. 1491 and likely was consulted by in Vercelli, Italy, July 2014. Christopher Columbus before his travels to the new world. The map was painted in tempera over a base of paper made backed by a large canvas. The writings are in several different colors on several different colors of background, which is a considerably different situation compared to that for manuscripts, which are usually in one or two colors of ink over bare parchment. The writings on the map have so faded over time that very few texts remain readable to the unaided eye. The images were processed to reveal much text that had not been read for hundreds of years. These writings confirmed that much of the text on the 1507 printed world map by Martin Waldseemüller (which was imaged in 2007) had been copied from this map or from its sources. This project received a considerable amount of press attention, including a one-page article in the June 2015 issue of Smithsonian Magazine. In September 2014, Dr. Easton trav- will be traveling there in July-August imaging capability to preserve the eled to Paris with Dr. Heyworth and of 2015 to demonstrate capabilities writings on palm leaf manuscripts in 70 Michael Phelps to present an imaging prior to proposing full projects India. This proposal was accepted and workshop at the École Pratique des Throughout the year, the team worked Dr. Easton now serves on the advisory Hautes Études, which is associated on processing of images of the board of the new organization. with the Sorbonne. This week-long palimpsests in the collection at St. workshop was attended by approxi- Catherine’s Monastery in Sinai and of Presentations: mately 20 participants from several the David Livingstone diaries in the 27 June 2014: Talk at Dartmouth institutions in France and led to the collection of the David Livingstone College proposal for a project to image the Center in Blantyre and the National 12, 14 August 2014: Two talks at the damaged manuscripts from the Cathe- Library of Scotland in Edinburgh. dral at Chartres. Beinecke Rare Book and Manuscript Dr. Easton traveled to Udupi, India Library Together with Michael Phelps and in May 2015 with Dr. P.R. Mukund Dr. Heyworth, Dr. Easton traveled to 15-19 September 2014: workshop to of the RIT Department of Electrical EPHE in Paris Tbilisi in March 2015 to propose pilot Engineering and Dr. Knox to propose a projects to image manuscripts at the new Center for the Preservation of Manu- 18 October 2014: talk at Brick City Georgia National Center for Manuscripts scripts at the Shri Madhwa Vadiraja Weekend and the Georgia State Archives. These Institute Of Technology & Management, proposals were accepted and a team which is intended to establish the

The c.1491 map of the world by Henricus Martellus Germanus being prepared for imaging at the Beinecke Rare Book and Man- uscript Library at Yale University, August 2014. Project Director Chet Van Duzer in the foreground.

Ken Boydston (Megavision) and Dr. Gregory Heyworth (University of Mississippi) adjust the frame holding the Martellus Map before imaging (August 2014)

The Shri Madhwa Vadiraja Institute Of Technology & Management in Udupi India after presentation of proposal for the Center for the Preservation of Manuscripts

RIT Imaging Science 2014–2015 Annual Report Imaging Science Undergraduate's of Imaging Science and Photonics pretty desperate to start working on AMA Reaches Reddit Front Page from SPIE. Press Release it. I bugged my advisor for a very long time before I started working with 71 [Edit]: I should mention that the rea- him. And while I know there are mil- son for this AMA is due to a number Fourth-year lions of people who know more about of people requesting it after I posted student Kevin history and maps than I do, I think on this front-page TIL thread: http:// Sacca's "Ask Me it’s so interesting to learn first-hand www.reddit.com/r/todayilearned/com- Anything" (AMA) what the educated people in the 15th ments/39xhr8/til_of_a_monk_who_had_ about multispec- century thought about the world. taken_an_old_book_written/ tral imaging of Scattered all around the map are historical documents proved highly What was the (persumabely) oldest these legends and cartouches which popular on social network reddit document you ever recovered? contain text that very briefly describe Jun. 16, 2015 I know my colleagues have worked on the land there. They can be pretty fun- Imaging Science senior Kevin Sacca the Dead Sea Scrolls, which I’m pretty ny to us now, but it’s really interest- is spending his summer working with sure are as old as they come. But per- ing to see how our knowledge of the Dr. Roger Easton Jr. capturing and sonally, the Archimedes Palimpsest is world progressed to what it is now. processing multispectral images of the oldest document I’ve worked on. Digital information doesnt fade but historic documents. In response to I’m new to this whole world and gave formats/filesystems change so rapidly public curiosity about the nature of both The Dead Sea Scrolls and Archime- that you can quickly find yourself with- his work, Sacca set up an AMA ("Ask des Palimpsest a Google. Your job is the out any data that is readable. https:// Me Anything") interview session on en.wikipedia.org/wiki/Digital_dark_age social networking website reddit. The coolest ever and so is this AMA. Can you session proved to be wildly popular, ELI5 the findings on both of those? On top of storing our images in mul- racking up over 350 comments and Wow thank you! While I’m not sure of tiple file formats, uncompressed and even making it to the front page of the significance of the findings for the compressed, we store copies of them reddit. Here's a summary of the inter- Dead Sea Scrolls or the Archimedes in multiple places! I don’t think that view, taken from amahighlights.com: (in terms of history, culture, or the we are concerned about not being translations), I do know that so far, able to read them. I am a scientist who utilizes multispec- there have been many successful re- tral imaging to recover and preserve Are there any dangers involved during the coveries on both documents. I am not processing of ancient manuscripts? Could information from old documents. AMA! literate in the languages of the text I a misstep lead to a portion of a text being try to reveal, so unfortunately I can’t My short bio: My name is Kevin Sacca lost, and has this ever happened? and I am an senior undergraduate read as I go, I pass along my data to student researcher at the Rochester the experts who can. Yes there is significant risk in remov- Institute of Technology (R.I.T) work- Getting successful recoveries is the ing some of the documents from their ing in the Chester F. Carlson Center for best outcome because: casing and exposing them to intense Imaging Science. I work with Dr. Roger light. However, my group is very con- 1) The information is now preserved scientious about how we expose the Easton Jr. and his multidisciplinary digitally and won’t fade over time, team of scientists and scholars who target. Previous imaging teams used so it can be studied for an indefinite broad-spectrum light which exposed share a common passion for pre- period of time. serving the information and cultural the targets to way more light than they heritage that is inherent in antique 2) We are able to reveal text that should have, and that resulted in heat manuscripts, paintings, palimpsest, hasn’t been looked at in over 2000 transfer and the loss of ink/pigment. scrolls, etc. years… We hope that the text is sig- We use LED light sources which emit a Personally, I have been working on nificant in terms of it’s content, and very narrow bandpass, which dramati- if it is, it can literally change what we cally reduces the light energy incident the Archimedes Palimpsest, the Martel- know about our world’s history. upon the target to the point where lus Map of 1491, and various unknown there is almost no risk. documents from the St. Catherine’s So you just focus in on the actual Monastery in Mount Sinai. My job is process of uncovering the info, not the I can’t imagine a job where I could to use the raw multispectral imagery, analysis? get fired for turning the lights on too perform statistical image process- Yeah, in other words, I really just try bright… What’s the worst thing you’ll ing routines, and generate imagery to provide useful images to those admit to screwing up? showing much more clearly the text, who can read and make sense of the Well, I’ve honestly been quite good diagrams, or figures that have been information. almost lost due to fading or “palimp- about not screwing things up. The sesting”. Was there ever a document you were worst I’ve done so far is forget which hard drive (of like 80 4TB hard drives working on that completely infatuated My Proof: It’s hard to show my proof that surround me) I stored my data because I can’t disclose 99.9% of my your interest outside of the technical on. So, I had to start over, which is work, and an image of me holding a details of uncovering hidden text? Like, fine, but all that data took some time card with my reddit username won’t a piece of history that really interested to process again. you in all ways? help much… So here’s a press release Wow. Move over Hillary Duff. I’ll be of a scholarship I was awarded this Yes, absolutely. The Martellus Map re- stealing the spotlight today. Thank year for my contributions to the field ally grabbed my attention, and I was you so much! To have so many people that was lost on the Martellus in the Why can’t you disclose your work? 72 ask me questions about the stuff I 500 years since it was painted, and Well… funding for our research is typi- spend so much time doing is actually also considering the fact that the cally from private donors, museums, really nice! documents that many people want and, in the case of remote sensing, imaged are much, much older than Can you walk us through the general the government or government 500 years old, we need to act as soon practices of finding the overwritten text? contractors. While maybe it’s okay to as we can to try to preserve as much show you guys stuff, I wouldn’t take a Well, we capture high-resolution, nar- as possible. chance disclosing any of my work that row-band multispectral imagery using In my opinion, I think not. We need could result in some legal shitstorm tunable LED light sources and color a larger community working on this that could affect my career or the filter wheels in combination with our type of research. The imaging sci- careers of my friends who I work with. multispectral camera. ence community is surprisingly small Has there ever been anything surpris- We then take our massive data prod- considering its applications in almost ucts into ENVI, a software package every other field in the world. While ing you’ve come across while imaging from Exelis Inc., which is very useful that kind of job security should make documents? for image processing. Usually used me happy, it actually worries me that Yes, the craziest thing is when you for Remote Sensing purposes, ENVI we aren’t putting enough resourc- find an entire layer of text below text also contains statistical processing es into preserving our history and you can see clearly. It makes you routines that are invaluable for our culture when it’s sitting in museums wonder all sorts of things about the research task too! doing nothing but deteriorating. limitations of our human visual sys- Within ENVI, we simply use com- Has Google been of any help with delicate tem, and that there is much more than meets the eye. binations of statistical processing works? I know their main effort has been routines, like Principal Component more of a feed-into-a-machine-and-go, but How was the original writing in these old Analysis, Independent Component they could do wonders with their ability to documents erased? Was it painted over Analysis, and spectral math to gen- charge forward on a project and make the or some such? erate imagery whose contents are visually enhanced. information so widely accessible. Many times the ink was washed off I’m not aware of the efforts Google is with water or chemicals, other times Can you talk a little more about these making in this field, but I’m sure they the ink was scraped off. Then, the instruments? Makes and models. Just have their chip in the dip (Is that a say- cleaned paper is reused and written/ curious to see the equipment. ing? I don’t know. Maybe I just made painted over. Otherwise, the text is I’m not positive of the model (maybe something up.) as they so often do. faint due to natural aging and fading of pigments. the EV?), but check out MegaVision! That being said, Google translate has MegaVision — Cultural Heritage been super helpful to me working in How accurate was National Treasure I Division the lab, to help me figure out if the and II, and why were they very accurate? We most often use a MegaVision words I recover mean anything! Not accurate. They had a very whimsi- multispectral camera for our imaging, I’m fascinated with Martellus Map. Do cal interpretation of document imag- mainly because we have a colleague you have the source for a high resolution ing techniques. Great movies though. who works for them, but also because their equipment is top! version of it? I tried to google it and the Would you consider National Treasure biggest image have only 2800px wide the Jurassic Park or Indiana Jones of I actually worked on a competitor’s and I can’t read anything. your career? Also on a related and more multispectral & hyperspectral image A huge reason why you can’t read any- serious note, do you consider yourself processing tools for remote sensing thing is because most of the text has an archaeologist? You are in essence about a decade ago. I’d like to second a faded! Chet van Duzer is leading the recovering ancient artifacts for the study request for more info on the hardware Martellus project with Yale, so please of cultures past, even if your methods of and frequencies used since I only really check out his article(s) on the map, excavating and far from traditional. worked with the software end. and consider buying his new book. I’d love to haphazardly travel across The wavelengths of light we capture at I have a copy of the book, which the world uncovering the dark secrets are 365 to 940 nanometers in steps of includes sections of the map after of history, so I’d totally try National ~50nm. Then we perform fluorescent ul- processing, and you would be amazed Treasure out. I’m afraid of dinosaurs, traviolet imaging, which uses ultraviolet at how much text you can see clearly loved the new movie, but I’ll have to light, but not at specific wavelengths. afterwords. say no to dinosaurs. My boss is Indi- In your estimation, is the image What has been your most interesting find ana Jones, so I can’t. science research community working so far? I don’t consider myself an archaeolo- at a sufficient rate to preserve our Cartouches are treasure troves of gist, I don’t actually uncover the docu- great libraries before the books are ments. I am proud to be part of a team inevitably lost? How much longer do hidden text. On the Martellus map, I’ve found text in small sections of the map who leads an effort into preserving the Martellus Map of 1491 and other cultural heritage! But I think the more artifacts like it stand to survive in where there was thought to be no text (or no recoverable text). Written in scholarly colleagues of mine are more their present cozy, climate-controlled suited to the title of archaeologist. environments? these cartouches are some pretty interesting passages by Martellus himself. What file format do you store the images Well, considering the amount of text

RIT Imaging Science 2014–2015 Annual Report in? Something standard (TIF, PDF) or into imaging them, but I know that absolutely love to image old treasure something proprietary? if we had all the time and money, my maps! That would be SO COOL. Very 73 We store the images in multiple team would image every document in National Treasure-like. the world. file formats, tiffs because they are If you could inspect a document that lossless, jpegs because they can be I work in these materials daily (Dead Sea no longer is know to exists, or can’t be convenient for viewing, PDFs for use Scrolls, Codex Ephraemi Syri Rescriptus, proved to have existed but is rumored to in documents and articles, and finally, etc.). I presumed St. Catherines had have, what would it be? .img files for working in ENVI. been thoroughly raked through by now. I think the Lost Library of Alexandria So, when you are imaging a document, Is there anything of historical religious exists, and I hope we are able to are you capturing reflective photons, significance (say before 4th century) that uncover some text that proves it’s ex- transmissive, or both? Are you looking for you expect will soon see the light of day? istence, or best-case scenario, where florescence in any of the target materials? Yes, we have been planning more it is! You never know! All of the above. We capture trans- imaging this summer at St. Cather- In terms of your hours and free time, missive, reflective, and fluorescence ines. There are very old copies of the what is a general work day for you? from both sides of the parchment. We Bible (presumably), and other mate- have images of both sides because rials that may contain undertext of a Well during the academic year, I work sometimes, the ink can stain through completely different nature. mainly on weekends. This summer, the layers and not fade as much as the however, I work full-time on process- What inspired you to want to study imaging data. surface layer. ing science? How did you end up where My typical work day starts with What guides/books/websites would you you are? checking my email, looking for re- reccomend where I can learn about the Long story, I changed my major a quests from my scholar colleagues for basics of image processing? I am about handful of times (even before arriving processing specific images that have to take a course on image processing and at R.I.T) from International Business, grabbed their attention. eager to get started! Physics, Film & Animation, and finally I have a long long list of documents Excellent! I recommend you check out Imaging Science. to process, and so I take the scholar’s the software ENVI if you can, as it is I knew I wanted to study math and requests as priority, slowly checking an environment where you can learn physics, but I wanted a specialized things off the list. about a lot of different image process- curriculum, so I looked into Imaging I usually spend around four hours on ing routines. Science because I had the connection one image, as that’s about the time it In terms of books… I’d check out Gon- with the Film & Animation program. takes for me to finish all of my pro- zalez and Woods book, Digital Image I attended a presentation given by cessing routines in my arsenal. Then, Processing. Any edition will do. I see a PhD student who worked on the I take the best results and create a their books everywhere in my building palimpsest research project, and it psuedocolor rendering of them and at school. was so interesting that I decided to send them off to the scholars for inter- I do a lot of image processing in switch. And I switched my first week pretation. Many times they are able to Python, so if you’re new to program- of freshman year and have loved it read entire words that they could not ming, I recommend CodeAcademy.com since. see or read before! to learn the basics of programming, So you switched from business into phys- In terms of free time, there’s honestly and then I’d check out any documen- ics and eventually imaging science? You so much work to be done that there tation for image manipulation in py- give me hope that wall street is yet not is never really any free time. Howev- thon using scipy or spectral libraries! able to steal away all of the best minds. er, while the computer is chugging Is it possible to check work that didn’t through the processing, I have time Yeah, I’ve thought a lot about what it to look into new statistical processing use this technology to see if we’ve meant to switch my major so drasti- methods, and try them out on samples missed anything? I realize that certain cally, and the career paths that open to see if they work or not! All in the pieces are almost untouchable but I up (and close) by doing so, and to be effort to improve the results! think it’s amazing knowing that we could summarized simply: be sitting on new information and never I have plans to go abroad to Germany I ended up here because I’m studying this summer to image documents from know it. something I genuinely think is inter- the bombing in Dresden, so that will Yes it is entirely possible, and highly esting, something that I can do for a be about a week or so. career that can help others, the peo- recommended that we do so soon! The Links to any papers that explain the re- longer we wait, the more time that ple around me every day are amazing, and I’m happy where I am in life. I sults you’ve obtained? Any data you’ve aging and fading has to destroy the worked with that we can play with? evidence! think as long as those remain true, I can accomplish what I want to. My first co-authorship paper will be fin- There is a huge call to museums and If you could inspect any document that ished and hopefully presented at the IEEE universities to get their important conference in Rome this year, so I don’t exists in the world today what would it be? artifacts imaged using multispectral/ have a paper explaining my personal re- hyperspectral techniques! Fun question! Okay, well I don’t know sults. But please check out Roger Easton’s Some pieces are much better candi- about specific documents, but I’m website which I have linked in the original dates to put the time and resources an aspiring treasure hunter, and I’d post. He has plenty of stuff to look at. I can’t give out any of my data right imaging of tissue for medical purposes. Were some scientific discoveries in 74 now, because I don’t want to some- Well for starters, the best book for those documents, of which we didn’t how get into any trouble. But! I en- any kind of spectral imaging work is thought people had it back then? Or courage you to Google for RIT Archi- Schott’s book, Remote Sensing: An were there even discoveries we didn’t medes data or RIT Martellus data or Imaging Chain Approach. Because make, but learned about trough the anything of the like, because I know honestly, we are always doing remote documents? there is some posted online that you sensing, but on a smaller scale. can download yourself! Because if you Well I know as a kid we were taught find and download it yourself, then For a multispectral medical imaging that people in the 14th and 15th cen- that’s A-okay! focused book, check out Dr. M. Ali tury thought the world was flat, and Roula’s book. I know I’ve seen it that Columbus was going to sail off My uncle has a journal from a Civil War around before, and it might hit the the edge of the Earth. soldier that he rescued in the 1960s nail on the head for you. from a refuse pile (a small local mu- This is a myth, and proven over and Also find this book: Introduction to over again. Educated people back seum that was clearing out stuff they Medical Imaging by Nadine Barrie then knew the Earth was round, and didn’t find important) that he wishes to Smith. My classmates used this book evidence from the Martellus backs it preserve. What’s something I could tell for a medical imaging physics class up. him to do to preserve it? At lease until and said it was amazing. the point he passes it onto his daughter I’m a big fan of old documents myself. Do you think we’ll ever get consumer/ (a curator of old documents). It’s still Whats the best way for someone to get commodity hyperspectral imagers? readable, if that tells you anything. into this field of research? Something akin to the FLIR One? Are This research was what made me I am not too knowledgeable on this there any hacks that allow cheap hyper- matter, but I do know this: Keep it in change my major into Imaging Sci- spectral imagery? What spectral ranges ence. After I switched, I bugged my a dust/dirt/water free container out are useful? Is polarization also a useful of the light. Light is a big player in advisor for over two years before he signal at all? fading the text. hired me and I got to start working on Yes! I think that is very possible! In this stuff. If you’re a fellow RIT student, you the near future too! This technology might know about the Carey Col- The best thing to do is actively is relatively new, but it’s advancing lection at the Wallace Library. They contact people in the field. Let them so fast, so I foresee this technology would be EXCELLENT people to go talk know you are interested, and study being rapidly available! to about just this. Please, please talk up on the subject so if an opportunity to them. Spectral imagery is achieved by presents itself, you can demonstrate simply collecting photons of a small that you are a useful person to have The team in which you work with –have bandgap of wavelengths with a helping! you lot ever encountered a document sensitive detector. So, if you had a Is there a way to tell if document is which unveiled information which broke light source that emitted at all in a “recycled” or not? new grounds and changed the percep- given wavelength, you could place a tion of what we know of the history of wavelength-specific filter over your If the document shows signs of un- that period? If so, what was it and how detector and, if the exposure is right, dertext when illuminated by infrared, did you (and the team) feel? you have yourself a spectral imager. ultraviolet fluorescent light, then the Multi and hyperspectral are terms document may have been recycled While I’ve been a part of the team, no, used to describe the spectral resolu- and is a good candidate for multispec- nothing ground-shattering. tion of your system. I don’t foresee tral imaging. However, in the recent past, the whole hyperspectral technology being read- How would a school or researcher go foundation for this type of research ily available just yet, but definitely about getting your team to preserve was, in all seriousness, accidentally multispectral. (multispectral: ~tens of some documents? stumbled upon. The main goal I be- bands — hyperspectral: ~hundreds of They should contact Dr. Roger Easton lieve was simply to enhance the text bands) that we could already see a little, but at [email protected] they ended up finding a palimpsest, Also curious about polarization detec- Another good resource is R.I.T’s Cary so an entirely new layer of text under tion. It is a relatively simple technique Collection They could offer insight in the existing text. which has seen many applications in the best way to preserve documents, biophysical fields and can easily be I know the efforts at the St. Cather- and they would easily be able to con- combined with hyperspectral imaging. ines Monastery have been quite fruit- tact our team for imaging. ful as well. There’s been talk about Yes! Polarization is an incredibly Do you currently have any technology or possibly discovering Shakespeare’s useful method to isolate the import- techniques that could see text that has 13th signature or more Archimedes ant signal. While I’m not aware of the been sharpied over? Like redacted infor- works, it’s all so amazing! measures we take to polarize our light mation from CIA reports? Perhaps the for document imaging, as an Imaging 29 pages about the Saudis and 9/11. What are some texts or papers you Science student, we study polariza- would recommend as “required read- tion a lot, and there is some incredi- Honestly, probably. I imagine that ing” for multispectral imaging? I’m ble research into polarization being back in the day, before knowledge of hoping to begin an engineering project done by a few graduate students I spectral reflectance was less under- at work that will involve multispectral know! stood in industry, sharpie markers

RIT Imaging Science 2014–2015 Annual Report didn’t absorb all wavelengths of light, so perhaps to some wavelengths of light, it would look like the sharpie 75 wasn’t even there. I’m sure it’s possible to recover at least some classified text like that. If you never hear from me again, you can be assured that it’s true. How are you able to keep from falling in love with every woman scientist you see in the lab? It’s been all over the news that girls are bad for science because everyone apparently falls in love and no work gets done. Easy. I’m the only person working in my lab this summer. I get my work done with (relatively) few distrac- tions! RESEARCH 76RESEARCH

RIT Imaging Science 2014–2015 Annual Report RESEARCH 77 LABORATORY FOR MULTIWAVELENGTH ASTROPHYSICS Laboratory Director's Comments By Dr. Joel Kastner

1. Summary A subset of the faculty, staff and students from the Center for Imaging Science (CIS) and Department of Physics within the College of Science at Rochester Institute of Technology participates in the Laboratory for Multiwavelength RESEARCHAstrophysics (LAMA). LAMA exists to foster the utilization and advancement of cutting-edge techniques in multiwavelength astrophysics by RIT faculty, research staff, and students, so as to improve human understanding of the origin and fate of the universe and its constituents. Calendar year 2014 marked LAMA’s “coming of age” as a designated research lab at RIT: CY 2014 constituted, by far, LAMA’s most prolific year thus far in terms of dissemination of research results. Spe- cifically, LAMA faculty, postdocs, and students were lead or co-authors ofover 30 refereed papers and nearly 50 conference presentations and other non-refereed publications—roughly double our CY 2013 publication totals. A significant fraction of these (82) publications resulted from projects in which LAMA personnel play leading roles within national and international teams of astrophysics researchers. In 2014, LAMA-affiliated PIs held or obtained 28 external grants, mostly from NASA and the NSF. LAMA PI grant expenditures totaled approximately $805K, yielding roughly $42.5K in overhead recovery return to the lab. These numbers were down by ~20% relative to CY 2013, perhaps reflecting the tightening NASA and NSF astrophysics budgets and the resulting increasingly competitive astrophysics funding environment. Expenditures by the research center account totaled approximately $20.4K. In 2014, LAMA continued its highly successful summer student research programs, again in association with (and in support of) the Center for Imaging Science (NSF-sponsored) Research Experience for Undergraduates program. LAMA also continued its investments in RIT’s astrophysics research infrastructure, and maintained its administrative and software support. Overall, LAMA continued to play a leading role in support of RIT's thriving community of astrophysics researchers and educators, and served to boost the profile and reputation of RIT astrophysics on national and international levels. 2. Mission Statement LAMA's Mission. The mission of LAMA is to foster the utilization and advancement of cutting-edge techniques in multiwavelength astrophysics by RIT faculty, research staff, and students, so as to improve human understanding of the origin and fate of the universe and its constituents. LAMA exists to support the following major astrophysics activities at RIT: • exploitation of existing and forthcoming national and international ground- and space-based astronomical observing facilities/missions; • exploitation and mining of the present and forthcoming generations of multiwavelength data archives; • LAMA contributed to RIT's investment & Physics AST Ph.D. program. As in 78 • development of scientific require- in the WIYN 0.9 m telescope consor- previous summers, our LAMA-sup- ments for future astronomical observ- tium via direct payment of member- ported student (Physics major Alyssa ing facilities/missions and future ship fees and student travel support. Hunter) was seamlessly integrated data archival and mining methods; • LAMA continued to use some of its dis- into the larger summer astrophysics student research community of • analysis and modeling of mul- cretionary funds for support of student travel to conferences, publication page grant-supported CIS and Physics tiwavelength astronomical and undergraduates and visiting REU stu- astrophysical data. charges, and general RIT astrophysics community-building activities. dents involved in summer astrophys- Goals & Objectives. Support of the four ics research, all with the support and major activities listed above drives 3. Personnel & Finances encouragement of our LAMA-spon- LAMA's primary goals and objectives. sored AST graduate students. Small Participating Faculty: Stefi Baum (CIS), Specific LAMA goals and objectives working groups, organized around Roger Dube (CIS), Joel Kastner (CIS; include: research themes and data analysis LAMA Director), Jacob Noel-Storr techniques, developed naturally. (1) obtain external funding sufficient to (CIS), Chris O'Dea (Physics), Michael Biweekly group meetings were held maintain a healthy cadre of student and Richmond (Physics), Andrew in which the LAMA fellows and REU postdoctoral scholars pursuing research Robinson (Physics) students (along with AST grad stu- in multiwavelength astrophysics; dents and LAMA faculty and postdocs) (2) widely disseminate the research Research Staff (Postdoctoral gave research status reports and results of LAMA-affiliated faculty, Researchers): Preeti Kharb, Rupal shared results with each other. These postdocs, and students; Mittal, Benjamin Sargent LAMA-inspired and LAMA-supported Graduate Students: Triana Almeyda, summer student projects led to sever- (3) promote a highly dynamic, inter- al student presentations at the 2014 active astrophysics research environ- Kevin Cooke, Marcus Freeman, Davide Lena Dave Principe, Kristina Punzi, RIT Summer Undergraduate Research ment at RIT and bolster national and Symposium. Some students also international astrophysics collabora- Valerie Rapson, Sravani Vaddi, Billy Vazquez (all AST Ph.D.) presented at the Jan. 2015 American tions involving RIT; Astronomical Society meeting, and (4) strategically invest in novel astro- Undergraduate Students: Cicely a few projects are developing into physics research initiatives and in new DiPaolo (Imaging Science), Alyssa papers for refereed journals. astrophysics research infrastructure in Hunter (Physics), Kaitlin Schmidt both the instrumentation and software (Physics), Lucas Shadler (Physics) RIT Astrophysics Community Building. domains, within and beyond RIT. Via hospitality support in 2014, LAMA Finances. A total of 28 external grants again facilitated informal interactions were maintained by LAMA faculty and between visiting RIT astrophysics col- Progress Toward Goals & Objectives. In research staff PIs during calendar 2014, LAMA's fourth year as a des- loquium speakers and RIT's community year 2014. The vast majority of this of AST and CIS graduate students and ignated research laboratory at RIT, funding came from NASA and the LAMA-affiliated faculty, postdocs and postdocs. These informal gatherings National Science Foundation. Expen- over lunch or dinner are very popular students made the following progress ditures in CY 2014 totaled approxi- toward these goals and objectives: with the students, as they serve as mately $805K, and overhead return opportunities to make connections • CY 2014 was, by far, LAMA’s most to LAMA in CY 2014 totaled $42,538. and ponder career choices. LAMA also prolific year thus far in terms of These numbers were down by ~20% provided pizza and drinks for the week- dissemination of research results. relative to CY 2013, with this de- ly RIT astrophysics lunch talk series, LAMA faculty, research staff, and crease likely reflecting tightening whose typical audience consists of 5-10 students were lead or co-authors NASA and NSF astrophysics budgets graduate students from the AST Ph.D. of over 80 refereed papers and and the resulting increasingly com- program and another half-dozen re- conference presentations, with a petitive astrophysics funding environ- searchers from all three RIT astrophys- significant fraction of these publi- ment. Expenses in CY 2014 totaled ics research labs (LAMA, CCRG, CfD). cations resulting from projects that $20,428. Details concerning LAMA have succeeded as a direct result of grants, grant expenditures, and lab ImagineRIT. As in the previous three LAMA’s leadership of large national account income and expenses for CY years, LAMA’s Valerie Rapson led the de- and international teams of astro- 2014 are available upon request. velopment and organization of the AST physics researchers. graduate program’s exhibit at the May 2014 edition of ImagineRIT. All of our • LAMA-affiliated PIs obtained or main- 4. Student Support and Community Building LAMA-sponsored AST graduate students tained 28 grants, with total CY 2014 played essential roles in this effort. expenditures of approximately $805K. Immersive Summer Undergraduate LAMA graduate students: outreach with- • In summer 2014, in association with Student Research Program. In summer in and beyond Rochester. In January (and in support of) the Center for 2014, LAMA continued its RIT sum- 2014, LAMA graduate student Valerie Imaging Science (NSF-sponsored) Re- mer undergraduate student research Rapson was inducted into the Ameri- search Experience for Undergraduates program, in association with (and in can Astronomical Society’s Astronomy program and the School of Astronomy support of) the Center for Imaging Ambassadors program. This is a select & Physics AST Ph.D. program, LAMA Science (NSF-sponsored) Research group of early career scientists who continued its highly successful sum- Experience for Undergraduates (REU) are actively involved in astronomy mer student research program. program and the School of Astronomy education and public outreach (EPO),

RIT Imaging Science 2014–2015 Annual Report and want to continue to improve the colleagues at the University of Tokyo large astronomical observatories on effectiveness of this public outreach. have recently upgraded the camera Earth and in space (e.g., ALMA, GAIA, Valerie attended professional devel- on a telescope at the Kiso Observato- GPI, SPHERE, JWST).” 79 opment workshops to improve her ry, in the mountains of central Japan. EPO skills. She continued to conduct Richmond helped to write some of a variety of outreach activities n the the software which analyzes imag- Rochester area. Most prominently, she es taken by this camera, which is served as President of the Astronomy currently scanning large sections of Section of the Rochester Academy of the sky in survey mode. In Septem- Sciences, the area’s foremost amateur ber, 2014, as the Kiso Observatory astronomy society. She also published celebrated its 40th anniversary, a paper in the Communicating Astron- Richmond travelled to Japan to par- omy with the Public Journal detailing ticipate in the event, which brought Fig. 2: Michael Richmond addressing the her experiences in astronomy educa- together scientists, administrators audience, in Japanese, on the occasion of tion for senior citizens (Fig. 1). and members of the Kiso community. the 40th anniversary of the University of As part of the after-dinner entertain- Tokyo’s Kiso Observatory. ment, Richmond stepped onto the stage to deliver an address in Japa- 5. Selected Research Highlights nese (Fig. 2), much to the amusement Michael Richmond of the audience. Stars that explode in brilliant fireworks • As CY 2014 ended, Joel Kastner was at the end of their lives are called hard at work as co-Chair of the Scientific supernovae. Astronomers find hundreds Organizing Committee of International each year, but almost all occur in the far Astronomical Union Symposium #314, reaches of the universe. Every few years, “Young Stars & Planets Near the Sun,” however, one of these giant explosions to be held in Atlanta, GA in May 2015. occurs in a relatively nearby galaxy, Quoting from the IAUS 314 website with the Milky Way's neighborhood. Fig. 1: In 2014, Valerie Rapson continued her (http://youngstars.gsu.edu/): “Research Michael Richmond uses the RIT Obser- programs of astronomy education for senior over the past two decades has led to vatory's small telescopes to study these citizens, such as this opportunity to view sun- the discovery of hundreds of young uncommon events. On July 27, 2013, he spots and limb darkening on the disk of the stars within 100 parsecs of the Sun. was observing an ordinary star when Sun using a small telescope and solar filter. Many of these stars have been classi- an e-mail announced the discovery of Kristina Punzi planned and participated fied as members of kinematic “moving supernova 2013ej in the galaxy M74; in various local outreach fairs, includ- groups”, whose ages range from an he quickly shifted his attention to the ing Super Science Saturday at Victor extremely youthful (by astrophysical new object. Between July and December, Intermediate School, Family Science standards) 8 million years, up to a still 2013, he measured the brightness of this Day at the University of Rochester, and very young 200 million years. Because star on almost every clear night through Celebrate Science for Brick City Week- these groups represent the closest several different filters. The resulting end. Additionally, she volunteered for young stars to the Earth, they constitute light curves (Fig. 3) show that this star public outreach for the Insight Lab and the best sample available to investigate had a large and massive envelope of hy- WISE and at local K-12 schools. the early evolution of low-¬mass to in- drogen when it exploded: it takes several termediate-mass stars. Their members Davide Lena and Sravani Vaddi partic- months for much of the energy generated represent the most convenient targets at the center of the star to fight its way ipated in the 2014 Science and Tech- with which to obtain direct informa- nology Entry Program (STEP). STEP out through the envelope. The results tion on the conditions and timescales are published in Journal of the American is an Imaging Science fair for middle associated with the evolution of school students and their parents. Association of Variable Star Observers, circumstellar disks, and the subsequent vol. 42 (2014). Davide and Sravani showed models of formation and early physical and dy- the core of a main sequence star and namical evolution of planetary systems. post-main sequence star, simulation This symposium of the International models of galaxy interactions, and Astronomical Union (IAU) will bring introduced the students to Star Walk together scientists working on nearby and other guides to the night sky. young stars from multiple research Outreach abroad. Astronomers around perspectives, and from all over the the world study the same sky, making world. The meeting will feature sessions international collaborations com- on the identification, ages, and origins mon. LAMA faculty and students are of nearby young stars, new constraints engaged in wide-ranging research these young stars put on theories of Fig. 3: Light curves of supernova 2013ej in the programs that usually involve interna- early stellar evolution, the dispersal of galaxy M74, obtained at the RIT Observatory. tional teams and interests. Examples: protoplanetary disks and the origins of Valerie Rapson debris disks, and the early evolution of • Michael Richmond has been work- planetary systems. Scientists will also In April of 2014, graduate student Val- ing with scientists in Japan since discuss the prospects for advances in erie Rapson, Professor Joel Kastner and his days as a post-doc on the Sloan the study of nearby young stars and their international team of astronomers Digital Sky Survey in the 1990s. His planets resulting from new and future won time on the new Gemini Planet Imager on Gemini South to image a and continued into 2014 with Cycle these stars’ protoplanetary disks. For planet-forming disk around a nearby 14 observations. Analysis of this some, the spectral absorption sig- 80 young star. Their observations tested new data with the previous observa- natures are consistent with gaseous the limits of the dimmest objects cations (found in Freeman et al. 2014) formaldehyde (H2CO), and, for others, pable of being observed with GPI, and revealed that more than half of the they are consistent with gaseous were spectacularly successful. An exam- surveyed nebulae had some sort of formic acid (HCOOH). Modeling of the ple image from GPI is exhibited in Fig. X-ray emission. We found that ~27% spectra of stars that show water vapor 4, and shows scattered light off dust hosted diffuse X-ray emission related emission features suggests these grains in the disk around the young star to hot bubbles within the nebulae, gases are present in the inner regions V4046 Sgr. This image, combined with while ~36% hosted point-source X-ray of their host disks, corresponding to previously published sub-mm data, emission related to the central stars. the habitable zones of stars at a very suggests that multiple planets may be Furthermore, we detected diffuse early age (a few million years). forming and carving out rings in the X-ray emission for the first time from Infrared spectroscopy also reveals the disk (Rapson et al., 2014). NGC 1501, NGC 3918, NGC 6153, and composition of the material around NGC 6369 (Fig. 5). dying stars. Spitzer-IRS spectra of red supergiant (RSG) and oxygen-rich asymptotic giant branch (AGB) stars in the nearby Large Magellanic Cloud and Small Magellanic Cloud galaxies and in our Galaxy show emission from amorphous dust grains of silicate composition that were produced by these stars. "Amorphous" means the dust grain lacks long-range crystalline order, though it is still a solid; also, "oxygen-rich" means the abundance of oxygen exceeds that of carbon in the star's photosphere. An analysis of spectra from the Spitzer Lega- Fig. 4: Near-infrared coronagraphic/pola- cy program "SAGE-Spectroscopy" rimetric image obtained with the Gemini (Principal Investigator: F. Kemper), Planet Imager, showing scattered light off Fig. 5: Chandra X-ray Observatory (CXO) vs. the Spitzer program "SMC-Spec" (PI: small dust grains in the planet-forming disk Hubble Space Telescope (HST) images of five G. Sloan), and other archival Spitzer- around V4046 Sgr. planetary nebulae newly imaged in X-rays by IRS programs shows differences in CXO as part of the Chandra Planetary Nebula the spectral emission features of the Valerie was also among 31 winners amorphous silicate dust from the (including Kevin Cooke; see below) Survey (ChanPlaNS). Images reproduced from Freeman et al. (2014). Oxygen-rich AGB stars versus the chosen from nearly 450 students who amorphous silicate dust from the RSG entered the Chambliss student poster Kristina Punzi stars. Radiative transfer modeling presentation competition at the suggests these spectral differences 223rd American Astronomical Soci- In May 2014, Kristina Punzi attended the National Radio Astronomy Obser- are due to differences in the proper- ety meeting in Washington, D.C., in ties of the dust grains themselves. January 2014. The Chambliss awards vatory's 14th Annual Imaging Synthe- sis Workshop in Socorro, NM. In June This, in turn, may suggest differences recognize exemplary research by in the circumstellar environments undergraduate and graduate students 2014, she attended and presented the results of her radio spectroscop- around RSG versus O-rich AGB stars in who present posters at the society which the stardust grains are forming. meetings. Rapson won an Honorable ic survey of the protoplanetary disk orbiting the young star LkCa 15 at the Mention for her poster on infrared Davide Lena spectroscopy of the V4046 Sgr proto- International Symposium on Molecular planetary disk. Spectroscopy in Champaign-Urbana, IL For decades it has been known that in the special mini-symposium enti- the nuclei of certain galaxies are char- Marcus Freeman tled: Astronomical Molecular Spectros- acterized by unusually large energy copy in the Age of ALMA. Kristina was outputs, the presence of radio-emit- As a part of The Chandra Planetary awarded an RIT Research and Creativi- ting jets, outflows and a number of Nebula Survey (ChanPlaNS), Marcus ty Grant to attend the conference. highly energetic phenomena varying Freeman, as well as Professor Joel Kast- on short time-scales (days). Such ner, continued to explore X-ray emission Ben Sargent galaxies are known as active galactic from several planetary nebulae within nuclei (AGNs). The key mechanism 1.5 kpc of the Sun. The survey was de- Spectra of a small sample of T Tauri stars—young stars with orbiting dust that keeps them active is believed signed to place constraints on the X-ray to be well understood: gravitational properties of planetary nebulae and and gas disks that may evolve into planetary systems—show emission potential energy is released in the their central stars, and give us insight form of radiation during the accre- into the evolution of these objects. in Spitzer Space Telescope Infrared Spectrograph (IRS) 5–7.5 micron tion of gas on a supermassive black ChanPlaNS began with Chandra Cycle wavelength spectra from water vapor hole (SMBH) residing at the galactic 12 observations and archival data, and absorption from other gases in center. Nevertheless it is not yet clear

RIT Imaging Science 2014–2015 Annual Report how the gas is funneled from the al. 2000). Top center: velocity dispersion 224th American Astronomical Society inner kiloparsec down to the sub-par- (km/s) derived from our observations. Top Meeting in Washington, D.C. and at 81 sec scale accretion disk that fuels the right: velocity residuals (km/s) derived by the 18th annual Cool Stars Conference SMBH. Is there a relation between the subtracting from the observed velocity map in Flagstaff, AZ. This research is being accretion rate of the SMBH and the a second velocity map representing gas prepared for publication and will be presence of outflows? rotating in circular orbits. Bottom: cartoon submitted in 2015. representing the model proposed to explain Davide Lena, Professor Andrew the observed features. Most of the gas Chris O’Dea and Stefi Baum Robinson, and their team have been experiences rotation, and perhaps inflow, in using the Integral Field Unit (IFU) of Galaxies grow as stars form amid the large-scale galactic disk (left). The AGN the Gemini Multi Object Spectrograph dense clouds of cold molecular gas. radiation cones intersect the interstellar (GMOS), mounted on the 8-m Gemini Their growth can stall if this gas is medium rotating in the disk producing the telescopes, to observe the ionized gas disrupted, expelled from the galaxy, lobes visible in the flux maps (central panel, in the nuclear regions of a sample of or prevented from forming in the to be compared with the flux map derived AGNs. We completed the analysis and first place. The kinetic and radiative by Ferruit et al.). Along with flux features modeling of the active galaxies NGC energy liberated as matter falls onto revealed from the HST observations, the line 1386 (Fig. 6) and NGC 1365, a Seyfert a galaxy’s central supermassive black profiles that we obtained show that gas in 2 and a strongly barred Seyfert 1 hole is, in principle, large enough the nucleus is outflowing along the axis of respectively. We found that the gas to inhibit star formation via some of the AGN radiation cone (green blobs in the kinematics in the inner kiloparsec these mechanisms, perhaps over a right panel). However, the velocity disper- of these galaxies can be explained significant fraction of cosmic time. sion map (top center) and the velocity resid- as a combination of rotation in the Were this indeed the case, black hole ual map (top right) reveal that the bar-like large-scale galactic disk, inflows and feedback could alleviate many prob- emission highlighted in the top left panel compact outflows along the axis of the lems in our understanding of galaxy has a kinematical counterpart which can be AGN "radiation cone". evolution, reconciling observations interpreted as rotation along the axis of the with a theory that would otherwise NGC 1386 was previously observed AGN radiation cone and/or outflow in the over-predict the size of galaxies and by Ferruit et al. 2000 with the Hubble equatorial plane of the torus (perpendicular the star formation history of the Space Telescope (HST). They revealed to the AGN radiation cone). Universe. The mechanical energy the presence of a faint bar-like emis- of jets powered by active galactic sion extending approximately 150 Kevin Cooke nuclei (AGN) can strongly couple pc east and west of the nucleus. We During the 223rd Meeting of the AAS to the multiphase gaseous environ- found strong evidence that this emis- in Washington, D.C. Kevin Cooke won ments through which they propagate. sion is approximately aligned with a Chambliss Astronomy Achievement However, we have yet to understand, the equatorial plane of the AGN torus, Award for his research poster on bright- beyond circumstantial evidence, how and we showed that it is associated est cluster galaxies, one of only 6 gold black hole feedback might govern the with kinematical features suggestive medals awarded to the 150 graduate fate of cold molecular gas and the of outflows and/or rotation. Classical student posters at the conference. formation of stars within it. AGN unification models do not predict Kevin’s poster described his research of such kinematic in the plane of the line emission indicative of star forma- O'Dea and Baum, working with former torus. Our observations might there- tion in massive elliptical galaxies. RIT graduate student Grant Tremblay fore be providing a new clue to the (now at Yale) have found that a super- physical process operating in AGNs. David Principe massive black hole can act much like a mechanical pump in a water foun- In 2014 David Principe successfully tain, driving a convective flow of mo- defended his Ph.D. dissertation, enti- lecular gas that drains into the black tled “Multiwavelength Observations hole accretion reservoir, is pushed of Young Stars and Their Circumstellar outward again in a jet-driven outflow, Disks”. Part of his dissertation was and falls once more back toward the focused on infrared and X-ray obser- galaxy center from which it came. This vations of the nearby star-forming molecular fountain has been discov- region L1630 where he character- ered in the central galaxy of the near- ized magnetic phenomena in ~50 by cool core cluster Abell 2597, using X-ray emitting young stars and their new observations from the Atacama variability over several years. This Large Millimeter/submillimeter Array work was published in the Astrophys- (ALMA). The data reveal giant molec- ical Journal Supplements (Principe ular clouds falling on ballistic trajec- et al. 2014). The remaining chapters tories towards the black hole in the of his thesis investigate variability innermost 500 parsecs of the galaxy. in three nearby, star-disk systems The black hole accretion reservoir, using simultaneous multiwavelength fueled by these infalling cold clouds, observations. Preliminary results sug- powers an AGN that drives a jet-driv- gest that variable obscuration of the en molecular outflow in the form of stellar photosphere from a disk warp a 10 kpc-scale, billion solar mass Figure 6. Top left: [OIII] flux map of the and/or clump viewed nearly edge- expanding molecular bubble. The mo- circum-nuclear region of NGC 1386 ob- on is responsible for the variability. lecular shell is permeated with young tained from HST observations (Ferruit et David presented this research at the stars, perhaps triggered in situ by the 6. Publications Astrophysical Journal 793, L26. jet. Buoyant X-ray cavities excavated 82 Refereed Journal Articles (10) Kastner, J. H., Hily-Blant, P., Rodri- by the propagating radio source may guez, D. R., Punzi, K., Forveille, T. (1) Morokuma, T., and 58 colleagues further uplift the molecular filaments, 2014. Unbiased Millimeter-wave 2014. Kiso Supernova Survey which are observed to fall inward Line Surveys of TW Hya and (KISS): Survey strategy. Publica- toward the center of the galaxy from V4046 Sgr: The Enhanced C2H tions of the Astronomical Society of which they came, presumably keeping and CN Abundances of Evolved Japan 66, 114. the fountain long-lived. The results Protoplanetary Disks. The Astro- are highly inconsistent with the “hot (2) Richmond, M. W. 2014. BVRI physical Journal 793, 55. mode” accretion scenario envisaged Photometry of SN 2013ej in M74. (11) Taylor, M., and 10 colleagues by many theorists, and show that Journal of the American Associ- 2014. The Core Collapse Su- both the mass and kinetic luminosity ation of Variable Star Observers pernova Rate from the SDSS-II budgets of black hole accretion and (JAAVSO) 42, 333. feedback can be dominated by the Supernova Survey. The Astro- cold molecular phase, suggesting (3) Bharadwaj, V., Reiprich, T. H., physical Journal 792, 135. Schellenberger, G., Eckmiller, H. J., that an AGN can indeed couple to the (12) Sargent, B. A., and 10 colleagues Mittal, R., Israel, H. 2014. Intra- most important catalyst for galaxy 2014. Emission fromWater Vapor cluster medium cooling, AGN feed- evolution. However, the results also and Absorption from Other Gases back, and brightest cluster galaxy demonstrate that kinetic AGN feed- at 5-7.5 _m in Spitzer-IRS Spectra properties of galaxy groups. Five back is not a switch that shuts off star of Protoplanetary Disks. The Astro- properties where groups difier formation, as the fountain consists physical Journal 792, 83. not only of molecular gas, but young from clusters. Astronomy and stars as well. Astrophysics 572, A46. (13) Paragi, Z., Frey, S., Kaaret, P., Cseh, D., Overzier, R., Kharb, P. 2014. (4) Lena, D., Robinson, A., Marconi, Probing the Active Massive Black Billy Vasquez A., Axon, D. J., Capetti, A., Merritt, Hole Candidate in the Center of D., Batcheldor, D. 2014. Recoil- Billy Vasquez has been analyzing the NGC 404 with VLBI. The Astrophys- ing Supermassive Black Holes: A initial results of observations ob- ical Journal 791, 2. tained by the Spitzer Space Telescope Search in the Nearby Universe. The of the Seyfert Type 1 Active Galactic Astrophysical Journal 795, 146. (14) Tremblay, G. R., and 12 colleagues 2014. A 30 kpc Chain of Nucleus (AGN) NGC 6418 over a time- (5) Kato, T., and 85 colleagues 2014. "Beads on a String" Star Forma- span of more than a year (Fig. 7). The Survey of period variations of tion between Two Merging Early Spitzer observational campaign and superhumps in SU UMa-type dwarf Type Galaxies in the Core of a ultimately the publication of this first novae. VI. The sixth year (2013- Stronglensing Galaxy Cluster. The paper are the results of the efforts of 2014). Publications of the Astro- Astrophysical Journal 790, L26. an international collaboration that nomical Society of Japan 66, 90. includes leading experts on AGN. (15) Betoule, M., and 62 colleagues (6) Russell, H. R., Fabian, A. C., Our team has determined the radial 2014. Improved cosmological McNamara, B. R., Edge, A. C., distance to the 3.6 and 4.5 micron constraints from a joint analysis Sanders, J. S., Nulsen, P. E. J., characteristic region of the molecu- of the SDSS-II and SNLS super- Baum, S. A., Donahue, M., O'Dea, lar dusty region, known as the torus nova samples. Astronomy and C. P. 2014. The bow shock, cold in the Unified Model of AGN, using Astrophysics 568, A22. reverberation lag time series and fronts and disintegrating cool core cross correlation analysis. The results in the merging galaxy group RX (16) Rapson, V. 2014. Reaching for the are consistent with current models of J0751.3+5012. Monthly Notices Stars in your Golden Years: The clumpy tori in AGN. of the Royal Astronomical Society Importance of Outreach for Senior 444, 629-641. Citizens. Communicating Astrono- my with the Public Journal 15, 18. (7) Rapson, V. A., Pipher, J. L., Guter- muth, R. A., Megeath, S. T., Allen, (17) Principe, D. A., Kastner, J. H., T. S., Myers, P. C., Allen, L. E. Grosso, N., Hamaguchi, K., Rich- 2014. A Spitzer View of the Giant mond, M., Teets, W. K., Wein- Molecular Cloud MON OB1 EAST/ traub, D. A. 2014. Star Formation NGC 2264. The Astrophysical Jour- in Orion's L1630 Cloud: An nal 794, 124. Infrared and Multi-epoch X-Ray Study. The Astrophysical Journal (8) Freeman, M., and 26 colleagues Supplement Series 213, 4. 2014. The Chandra Planetary Neb- ula Survey (CHANPLANS). II. X-Ray (18) Kharb, P., O'Dea, C. P., Baum, Emission from Compact Planetary S. A., Hardcastle, M. J., Dicken, Nebulae. The Astrophysical Jour- D., Croston, J. H., Mingo, B., nal 794, 99. Noel-Storr, J. 2014. Very Large Baseline Array observations of (9) Tanaka, M., and 57 colleagues Fig. 7: The 3.6 micron (near-infrared) and Mrk 6: probing the jet-lobe con- 2014. Discovery of Dramat- time-shifted B-band (visible-light) light curves of nection. Monthly Notices of the ic Optical Variability in SDSS NGC 6418, representing the results of the time Royal Astronomical Society 440, J1100+4421: A Peculiar Radio-loud series analysis conducted by Vasquez et al. 2976-2987. Narrow-line Seyfert 1 Galaxy?. The

RIT Imaging Science 2014–2015 Annual Report (19) Dicken, D., and 16 colleagues (29) Schnorr-Meuller, A., Storchi-Berg- IRAS16399-0937. Revista Mexi- 2014. Spitzer Mid-IR Spectros- mann, T., Nagar, N. M., Robinson, cana de Astronomia y Astrofisica 83 copy of Powerful 2Jy and 3CRR A., Lena, D., Ri_el, R. A., Couto, G. Conference Series 44, 103-103. Radio Galaxies. II. AGN Power S. 2014. Feeding and feedback in (6) Rodriguez, D. R., Zuckerman, Indicators and Unification. The the inner kiloparsec of the active B., Kastner, J. H., Bessell, M. S., Astrophysical Journal 788, 98. galaxy NGC 2110. Monthly No- Faherty, J. K.,Murphy, S. J., Vican, (20) Aleman, I., and 17 colleagues tices of the Royal Astronomical L. 2014. The GALEX Nearby Young- 2014. Herschel Planetary Nebula Society 437, 1708-1724. Star Survey. Revista Mexicana de Survey (HerPlaNS). First detec- (30) Hamer, S. L., and 17 colleagues Astronomia y Astrofisica Confer- tion of OH+ in planetary nebulae. 2014. Cold gas dynamics in ence Series 44, 53-53. Astronomy and Astrophysics Hydra-A: evidence for a rotating (7) Robinson, A. 2014. Do Supermas- 566, A79. disc. Monthly Notices of the sive Black Holes really reside at (21) Dopita, M. A., and 12 colleagues Royal Astronomical Society 437, the centers of their host galaxies? 2014. Probing the physics of 862-878. HST Proposal 13922. narrow-line regions of Seyfert (31) Sacco, G. G., Kastner, J. H., For- (8) Rapson, V., Kastner, J. 2014. galaxies. I. The case of NGC veille, T., Principe, D., Montez, R., Seeing Double: Multiwavelength 5427. Astronomy and Astrophys- Zuckerman, B., Hily-Blant, P. 2014. Studies of the Remarkable beta ics 566, A41. Molecules in the transition disk or- Pic Moving Group Member V4046 (22) Ueta, T., and 32 colleagues 2014. biting T Chamaeleontis. Astronomy Sgr. Thirty years of Beta Pic and The Herschel Planetary Nebula and Astrophysics 561, A42. Debris Disks Studies 56. Survey (Her- PlaNS). I. Data over- (32) Audard, M., and 12 colleagues (9) Kastner, J. 2014. Thefirst young view and analysis demonstration 2014. Episodic Accretion in moving groups. Thirty years of Beta with NGC 6781. Astronomy and Young Stars. Protostars and Plan- Pic and Debris Disks Studies 4. Astrophysics 565, A36. ets VI 387-410. (10) Baum, S. 2014. First Chandra Ob- (23) Kato, T., and 86 colleagues 2014. (33) Principe, D. A. 2014. Multiwave- servation of a remarkable 'String Survey of period variations of length Observations of Young of Pearls' between two Merging superhumps in SU UMa-type Stars and Their Circumstellar Elliptical Galaxies. Chandra Pro- dwarf novae. V. The fifth year Disks. Ph.D. Thesis. posal 4475. (2012-2013). Publications of the Astronomical Society of Japan Other (Non-refereed) Publications (11) Sargent, B., Srinivasan, S., Rieb- el, D., Meixner, M. 2014. Probing 66, 30. (1) Sargent, B., and 14 colleagues Dust Formation Around Evolved 2014. Spitzer Light Curves of Dusty (24) Matsuura, M., and 28 colleagues Stars with Near-Infrared Interfer- AGB Stars in the Small Magellanic 2014. Spitzer Space Telescope ometry. Resolving The Future Of Cloud. Spitzer Proposal 11163. spectra of post- AGB stars in the Astronomy With Long-Baseline Large Magellanic Cloud - poly- (2) Kastner, J. 2014. X-ray Investiga- Interferometry 487, 371. cyclic aromatic hydrocarbons at tions of Young Stars Near Earth. (12) Lena, D. 2014. Reduction of low metallicities. Monthly No- 15 Years of Science with Chandra, Integral Field Spectroscopic Data tices of the Royal Astronomical Talks from the Chandra Science from the Gemini Multi-Object Society 439, 1472- 1493. Symposium held 18-21 November, Spectrograph (a commented 2014 in Boston, MA., id.23 23. (25) Chen, C. H., Mittal, T., Kuchner, example). ArXiv e-prints arX- M., Forrest, W. J., Lisse, C. M., (3) Couto, G. S., Storchi-Bergmann, T., iv:1409.8264. Manoj, P., Sargent, B. A., Watson, Axon, D. J., Robinson, A., Kharb, P., (13) Tremblay, G., O'Dea, C., Labia- D. M. 2014. The Spitzer Infra- Ri_el, R. A. 2014. Kinematics and no, A., Baum, S., McDermid, R., red Spectrograph Debris Disk excitation of the nuclear spiral Combes, F., Garcia-Burillo, S., Catalog. I. Continuum Analysis of in the active galaxy Arp 102B. Davis, T. 2014. A Sleeping Giant Unresolved Targets. The Astro- Revista Mexicana de Astronomia y Awakened: Reignition of AGN physical Journal Supplement Astrofisica Conference Series 44, Activity, Reborn Star Formation, Series 211, 25. 191-192. and a Multiphase Outow in one (26) McNamara, B. R., and 19 col- (4) Brum, C., Ri_el, R. A., Storchi-Berg- of the Largest Radio Galaxies leagues 2014. A 1010 Solar Mass mann, T., Schnorr Muller, A., Known. NOAO Proposal 401. Flow of Molecular Gas in the Robinson, A. 2014. Two-dimen- (14) Argiro_, C., and 13 colleagues A1835 Brightest Cluster Galaxy. sional kinematics of the central 2014. V4046 Sgr: X-rays from The Astrophysical Journal 785, 44. region of NGC4501 from GMOS/ accretion shock. IAU Symposium Gemini integral field spectroscopy. (27) Russell, H. R., and 16 colleagues 302, 46-47. 2014. Massive Molecular Gas Revista Mexicana de Astronomia y Flows in the A1664 Brightest Astrofisica Conference Series 44, (15) Gregory, S. G., and 16 colleagues Cluster Galaxy. The Astrophysical 190-191. 2014. The magnetosphere of the close accreting PMS binary Journal 784, 78. (5) Sales, D. A., Robinson, A., Axon, D. V4046 Sgr AB. IAU Symposium J., Gallimore, J., Kharb, P., Curran, (28) Bogd_an, _ A., and 10 colleagues 302, 44-45. 2014. Young AGN Outburst Run- R. L., O'Dea, C., Baum, S., Mittal, ning over Older X-Ray Cavities. The R. 2014. The Multi-wavelength (16) Baum, S. 2014. Synergy Between Astrophysical Journal 782, L19. Study of OH Megamaser Galaxy the Jansky Very Large Array, LSST and the Next Generation 30 leagues 2014. A $10^f10g$ #441.30. 84 Meter Telescopes. Thirty Meter Solar Mass Flow of Molecu- lar (35) Cooke, K., O'Dea, C. P., Trem- Telescope Science Forum 57. Gas in the Abell 1835 Brightest blay, G. 2014. Investigation of (17) Sebring, T., O'Dea, C., Baum, S., Cluster Galaxy. ArXiv e-prints Extended Emission Line Regions Teran, J., Loewen, N., Stutzki, C., arXiv:1403.4249. in Intermediate Redshift BCGs. Egerman, R., Bonomi, G. 2014. (27) Lena, D., Robinson, A., Stor- American Astronomical Society Progress on the New York State chi-Bergmann, T., Ri_el, R., Meeting Abstracts #223 223, Observatory: a new 12-meter Schnor-Muller, A., Brum, C., #358.16. astronomical telescope. Society Couto, G. 2014. Mapping sub-kpc (36) Richmond, M. W. 2014. BVRI of Photo-Optical Instrumentation gas ows in a sample of nearby, Photometry of SN 2013ej in M74. Engineers (SPIE) Conference hard X-ray selected AGNs. NOAO American Astronomical Society Series 9145, 91453D. Proposal 276. Meeting Abstracts #223 223, (18) Bombardini, M., and 16 col- (28) Kharb, P., Singh, V., Gallimore, #354.25. leagues 2014. Supernova J. F., Ishwara-Chandra, C. H. (37) Freeman, M., Kastner, J. H., 2014by in UGC 9267 = Psn 2014. Probing Radio Emission in Montez, R., ChanPlaNS Team J14274900+1133400. Central Seyfert Galaxies on Parsec- and 2014. The Chandra Planetary Bureau Electronic Telegrams Kiloparsec-scales. ArXiv e-prints Nebula Survey (ChanPlaNS): 3927, 1. arXiv:1402.1577. Results from Cycle 14. American (19) Betoule, M., and 62 colleagues (29) Gregory, S. G., and 16 colleagues Astronomical Society Meeting 2014. VizieR Online Data Cata- 2014. The magnetosphere of Abstracts #223 223, #353.27. log: Joint analysis of the SDSS- the close accreting PMS binary (38) Rapson, V., Kastner, J. H., Sac- II and SNLS SNe Ia (Betoule+, V4046 Sgr. European Physical co, G., Sargent, B. A. 2014. A 2014). VizieR Online Data Cata- Journal Web of Conferences 64, Spitzer and Herschel Study of the log 356, 80022. 08009. (30) Sako, M., and 48 col- Protoplanetary Disk Around the (20) Punzi, K. M., Kastner, J. H., Hi- leagues 2014. The Data Release Young Nearby System V4046 Sgr. ly-Blant, P., Forveille, T., Sacco, G. of the Sloan Digital Sky Sur- American Astronomical Society G. 2014. The First Unbiased Radio vey- II Supernova Survey. ArXiv Meeting Abstracts #223 223, Emission Line Survey of the Pro- e-prints arXiv:1401.3317. #350.07. toplanetary Disk Orbiting Lkca 15. (31) Noel-Storr, J., Campanelli, M., (39) Principe, D., Huenemoerder, D., 69th International Symposium on Bochner, J., Warfield, T., Bis- Kastner, J. H., Bessell, M. S., Sac- Molecular Spectroscopy 2. chof, H., Zlochower, Y., Nor- co, G. 2014. Heating the Primor- (21) Richmond, M. W. 2014. MATCH: A dhaus, J., Watkins, G., NSF dial Soup: X-raying the Circum- program for matching star lists. CRPA AstroDance Team 2014. stellar Disk of T Cha. American Astrophysics Source Code Library AstroDance: Teaching Astrophys- Astronomical Society Meeting 1406.010. ics Through Dance?. American Abstracts #223 223, #345.10. Astronomical Society Meeting (22) Almeyda, T., Robinson, A., Abstracts #223 223, #452.02. (40) Smith, C.-T., Rapson, V., Sargent, Richmond, M. W., Vazquez, B. B. A., Kastner, J. H., Rayner, J. 2014. Modeling the Reverbera- (32) Noel-Storr, J., Baum, S. A., RIT 2014. How Cool Is That? An IRTF/ tion Response of the Dusty Torus Insight Lab SSR Team, Carlson SPEX Spectroscopic Study of Emission from AGN. American CenterImaging Science Faculty, the Close Binary T Tauri System Astronomical Society Meeting C. F. 2014. Stepping Stones to V4046 Sgr. American Astronom- Abstracts #224 224, #221.17. Research: Providing Pipelines ical Society Meeting Abstracts from Middle School through PhD. #223 223, #345.04. (23) Ueta, T., and 32 colleagues American Astronomical Society 2014. Herschel Planetary Nebula Meeting Abstracts #223 223, (41) Rodriguez, D., Zuckerman, B. M., Survey (HerPlaNS). Asymmetrical #445.05. Kastner, J. H., Vican, L., Bessell, Planetary Nebulae VI Conference M. S., Faherty, J. K., Murphy, S. 106. (33) Rapson, V., Almeyda, T., Free- 2014. The GALEX Nearby Young- man, M., Lena, D., Principe, D., Star Survey. American Astronom- (24) Freeman, M., Kastner, J. H., Mon- Punzi, K., Sargent, B. A., Vaddi, tez, R. 2014. Cycle 14 Results ical Society Meeting Abstracts S., Vazquez, B., Vorobiev, D. #223 223, #334.06. from the Chandra Planetary 2014. Imagine Astronomy at the Nebula Survey (ChanPlaNS). Rochester Institute of Technol- (42) Lena, D., Robinson, A., Seelig, T., Asymmetrical Planetary Nebulae ogy. American Astronomical Schnorr-Muller, A., Ri_el, R. A., VI Conference 23. Society Meeting Abstracts #223 Storchi-Bergmann, T., Couto, G. (25) Gnerucci, A., Marconi, A., Ca- 223, #444.03. 2014. Gas Flows in the Inner Ki- loparsec of NGC 1386. American petti, A., Axon, D. J., Robinson, (34) Kastner, J. H., Baum, N., Principe, A. 2014. VizieR Online Data Astronomical Society Meeting D., Rodriguez, D. 2014. Serendip- Abstracts #223 223, #251.07. Catalog: Circinus galaxy CRIRES itous Chandra X-ray Spectrosco- observations (Gnerucci+, 2013). py of GALEX Nearby Young-Star (43) Vazquez, B., and 23 colleagues VizieR Online Data Catalog 354, Survey (GALNYSS) Candidates. 2014. Reverberation Mapping 90139. American Astronomical Society of the Dusty Torus of AGN NGC (26) McNamara, B. R., and 19 col- Meeting Abstracts #223 223, 6418. American Astronomical Society Meeting Abstracts #223

RIT Imaging Science 2014–2015 Annual Report 223, #251.04. (44) Every, M., O'Dea, C. P., Baum, 85 S. A., Noel-Storr, J., Vaddi, S. 2014. UV Emission of AGN in the 2Jy Sample of Southern Radio Galaxies. American Astronomical Society Meeting Abstracts #223 223, #250.30. (45) Chari, V., Noel-Storr, J., Paradis, J., Keenan, J., Dioguardi, P., UGC FR-I Collaboration 2014. Mor- phological Classifications of the Nuclear Disks and Radio Jets for a Complete Sample of Nearby Radio-Loud Elliptical Galaxies. American Astronomical Society Meeting Abstracts #223 223, #250.22. (46) Vaddi, S., O'Dea, C. P., Baum, S. A., Jones, C., Forman, B., Whit- more, S., Ahmed, R. 2014. Feed- back in the Local Universe: The Relation Between Star Formation and AGN Activity in Typical Elliptical Galaxies. American Astronomical Society Meeting Abstracts #223 223, #250.20. (47) Myers, S. T., Baum, S. A., Chan- dler, C. J. 2014. The Karl G. Jan- sky Very Large Array Sky Survey (VLASS). American Astronomical Society Meeting Abstracts #223 223, #236.01. (48) Montez, R., Humphreys, R. M., Kastner, J. H., Turok, R. L. 2014. X-ray Constraints on Magnetic Activity and Star Formation As- sociated with the Red Supergiant VY CMa. American Astronomical Society Meeting Abstracts #223 223, #152.15. (49) Sargent, B. A., Speck, A., Volk, K., Kemper, C., Reach, W. T., Laga- dec, E., Bernard, J., McDonald, I., Meixner, M., Srinivasan, S. 2014. Spitzer-IRS Spectroscopic Studies of the Properties of Dust from Oxygen-Rich Asymptotic Giant Branch and Red Supergiant Stars. American Astronomical Society Meeting Abstracts #223 223, #113.05.

OUTREACH OUTREACH86

RIT Imaging Science 2014–2015 Annual Report OUTREACH 87

Comments By Bethany Choate

Outreach efforts for the Center for Imaging Science have historically been manifested in a variety of programs, demonstrations, activities, materials, events, and publications which serve three main audiences: prospective students; the community at large; and alumni.

2014–2015 marked a transition year as outreach organizer and facilitator Bethany Choate, 2006 Imaging Science BS graduate, was reduced to part-time in anticipation of her departure in June 2015. Prior to her departure, Bethany prepared as her Masters Capstone Project a thorough OUTREACHStrategic Plan for Undergraduate Recruitment specifically for CIS. This plan investigates trends and best practices, benchmarks similar academic programs, details CIS’s enrollment and outreach history, includes a full marketing plan and analysis, and provides goals and suggested activities in support of those goals. It is hoped that this strategic plan will guide, inform, and support future outreach endeavors in CIS. CIS Intern Program The Center for Imaging Science summer intern program continued for its 15th year in 2014, with some changes to the program. For one, the program was shortened to six weeks due to RIT’s conversion to a semester calendar system and the resulting earlier start to the academic year. Additionally, due to the rising costs of maintaining the program, all interns were unpaid for the first time in the program’s history. Despite the funding challenges we were able to offer the same high level of experience and extracurricular programming with 15 interns as listed in the chart below:

Intern High School CIS Research Lab Maryam Bahrani Hotchkiss School (CT) Perform Rory Bloechl Harley Insight Jacqueline Chan Webster Schroeder Visual Perception Abraham Glasser Pittsford Mendon Remote Sensing Lindsay Haefner W. Irondequoit Astronomical Imaging Alex Kautz Brighton Insight David Lewis Wheatland-Chili DIY Camera Lab Adam Maier McQuaid Biomedical Imaging and Modeling Emma Pratt Rush-Henrietta Insight Rachel Shadler W. Irondequoit DIY Camera Jimmy Shih Webster Schroeder Insight Wesley Smith Brighton Astronomical Imaging Elizabeth Thrasher Gates-Chili Visual Perception Victoria Thrasher Gates-Chili Perform Madeline Wolters Hilton Participants benefit from research and Optics Classes 88 DIY Camera experience, exposure to a college Greece Olympia students enrolled Interns conducted authentic research environment, experientially exploring in Project-Lead-The-Way and Optics in several CIS laboratories as outlined their academic interests, and learning classes toured Sydor Optics and then in the table above. The program also the responsibilities of a full time job. visited CIS as a field trip. Students included a field trip to Mees Observa- The intern program is a community were provided an informative and in- tory and a career exploration lun- builder, as schools and students from teractive presentation about Imaging cheon featuring a live Skype session across the area participate, and local Science, complete with a hands-on with CIS alumnus and NASA scientist companies and research centers are thermal camera demo. showcased through field trips. The Matthew Montanaro. CIS contin- IYOL Career Day at RMSC ued the tradition of weekly cookout program is also CIS’s most success- lunches organized and coordinated by ful recruitment initiative: ~50% of CIS participated in a panel discus- the interns, which also included team participants have enrolled at RIT for sion and career fair at the Rochester building games and activities. Interns STEM-related disciplines; and 15% Museum and Science Center (RMSC) were required to maintain blogs re- of all past interns have applied to focused on Optics, Photonics, and cording their memorable experiences Imaging Science, with 12% ultimately Imaging. The event was presented to and presented their research at a half enrolling as students in the Imaging K-12 students as part of the museum’s day symposium at the end of August. Science program. We are excited to International Year of Light festivities. For the second year, we recorded report that three of our 2014 interns these presentations for later viewing. have enrolled as undergraduate Im- Science Saturdays and Light Weekends Meet our 2014 interns, view their aging Science students for Fall 2015. at the RMSC presentations, and read their blogs at Lastly, our internships encourage the CIS presented an exhibit on thermal http://www.cis.rit.edu/2014interns/. further pursuit of STEM disciplines by imaging in a booth at the Rochester up-and-coming young talent. This CIS Intern Program has jump Museum and Science Center as part of started successful academic and pro- K-12 and Community Outreach their “Light Weekends” programming for the International Year of Light. fessional careers for 173 participants Events over fourteen years. The benefits of Several objects and materials were the CIS intern program are four-fold: CIS’s in-person programming has available for interaction with the to students; to the community; to CIS been greatly reduced as of 2014, camera including ice, hair dryers, hot and RIT; and to STEM (science, tech- though several select events continue and cool liquids, plastic bags, space nology, education, and math) educa- to be supported. blankets, glasses, and more. Printed tion in the United States as a whole. Greece Olympia Project-Lead-The-Way images were available as souve- nirs to take home. Additionally, CIS

RIT Imaging Science 2014–2015 Annual Report students involved with the RIT SPIE/ The newsletter is e-mailed quarterly OSA student chapter participated in to CIS’s alumni population. If you are Light Weekends by demonstrating a an alumnus and wish to be added to 89 tabletop Schlieren imaging system on our mailing list, contact Joe Pow at a separate Saturday. [email protected]. New Website Social Networking CIS completed overhauled our entire CIS continues to maintain its so- website, www.cis.rit.edu, in 2014. cial networking sites, which have We partnered with the Wallace Center proven themselves quite successful. on campus to produce a modern, Please connect with us on Facebook appealing, highly-functional website and Twitter via the username RITi- that serves the needs of our diverse magingsci. Here you can find news audiences. We are proud of the results and information about the Center; of this significant undertaking and opportunities for potential students, welcome your comments on the new current students, and alumni; links to design and functionality. interesting Imaging-related articles; CIS event announcements; as well as Imaging Connection Newsletter general announcements. We encour- We have continued producing the CIS age anyone involved with CIS to Imaging Connection current events become a fan, interact with the site, newsletter in 2014-2015 and were and spread the word to others. There excited to unveil a new look and feel is also a LinkedIn group called RIT with the Spring newsletter. While the Imaging Science and Photographic intended audience is Imaging/Color/ Science Alumni. Photo Science alumni, most topics are of general public interest. Content includes: news stories related to CIS; upcoming events; alumni updates; research spotlights; and more. Past issues can be viewed online at http:// www.cis.rit.edu/about/newsletter. 90

RIT Imaging Science 2014–2015 Annual Report Chester F. Carlson Center for Imaging Science Rochester Institute of Technology 54 Lomb Memorial Drive Rochester New York 14623 (585) 475–5944

http://www.cis.rit.edu/