ROHIT PHILIP [email protected] (520) 208 - 4732 1240 N

Total Page:16

File Type:pdf, Size:1020Kb

ROHIT PHILIP Rphilip@Email.Arizona.Edu (520) 208 - 4732 1240 N ROHIT PHILIP [email protected] (520) 208 - 4732 www.ece.arizona.edu/~rphilip 1240 N. 7th Ave, Apt. 211, Tucson, AZ – 85705 www.linkedin.com/pub/rohit-philip/a1/41/586 PROFESSIONAL SUMMARY: • Computer Vision Ph.D. professional with over 8 years of experience as a graduate student, and 5 years of industry experience. • Strong theoretical background in Signal and Image Processing, Image Analysis, Machine Learning, Medical Imaging, Mathematics, and Statistics. • Exceptional analytical, programming, debugging, and technical writing skills. • Excellent leadership, communication, and interpersonal skills. • Proficient in MATLAB, OpenCV, Python, IDL, R, C, C++ programming languages. • Proficient in Perl, CGI, Javascript, VB Script and bash/shell scripting. • Professionally certified in Java, and SQL. • Exceptional MS Excel (including API programming), MS Word, MS Powerpoint, and LaTeX skills. • Extremely quick learner, and problem solver. SKILLS: Image Processing/Computer Vision MATLAB, OpenCV, IDL, ImageJ. Programming Languages C, C++, R, Python, Perl, Java, Bash. Scripting Languages VB Script, Javascript. Statistics Software JMP, Prism. General purpose Software MS Excel, MS Word, MS Powerpoint, LaTeX. Other Tools SQL, Putty, CMake. EDUCATION: Ph. D. - Doctor of Philosophy. Aug. ’13 – Present The University of Arizona, Tucson, Arizona. GPA: 3.63/4.00 (3.88/4.00 Post M.S.) Major: Electrical and Computer Engineering. Minor: Mathematics. Key Interests: Signal Processing, Image and Video Processing, Image Analysis, Computer Vision, Machine Learning, Pattern Recognition, Statistics, Mathematics. M.S. - Master of Science. Aug. ’05 – Dec. ’08 The University of Arizona, Tucson, Arizona. GPA: 3.38/4.00 Major: Electrical and Computer Engineering. B.E. – Bachelor of Engineering Jul. ’01 – May ’05 Anna University, Chennai, India. GPA: 77.71/100.00 Major: Electronics and Communication Engineering. (Honors: First Class with Distinction) 1 PUBLICATIONS: Peer-Reviewed Journal Articles: DW Todd, Rohit C. Philip, JJ Rodriguez, et al. “A Fully Automated High-Throughput Zebrafish Behavioral Ototoxicity Assay” Accepted after peer review at Zebrafish. Rohit C. Philip, JJ Rodriguez, et al. “Automated High-Throughput Damage Scoring of Zebrafish Lateral Line Hair Cells After Ototoxin Exposure”. Under peer review at Zebrafish. Rohit C. Philip, JJ Rodriguez, et al. “Kallynodetection: Redefining classical detection measures such as Precision, Recall, F-Measure, and G-Measure”. Undergoing final edits. Peer-Reviewed Conference Proceedings: Rohit C. Philip, S Ram, X Gao, and JJ Rodriguez. “A Comparison of Tracking Algorithm Performance for Objects in Wide Area Imagery” 2014 IEEE Southwest Symp. On Image Analysis and Interp. (SSIAI), San Diego, CA, Apr 6-8, 2014, pp. 109-12. Rohit C. Philip, and JJ Rodriguez. “Seed Pruning Using a Multi-Resolution Approach for Automated Segmentation of Breast Cancer Tissue” 2008 IEEE Intl. Conf. on Image Processing (ICIP), San Diego, CA, Oct 12-15, 2008, pp. 1436-9. Other Publications: DW Todd, Rohit C. Philip, M Niihori, JJ Rodriguez, and A Jacob. “High-Throughput Behavioral Zebrafish Assay for Drug Development Targeting Hearing Loss”, (abstract), AOS 149th Annual Meeting, American Otological Society, Chicago, IL, May 20-21, 2016. Best Poster Award. Rohit C. Philip. “Seed Pruning for Multi-Resolution Segmentation of Vasculature in Immunohistochemical Images”, Master’s Thesis, Department of Electrical and Computer Engineering, University of Arizona, Tucson, AZ, 2008. 2 EXPERIENCE: University of Arizona, Tucson, AZ Aug. ’13 - Present Graduate Research Assistant As a member of the Signal and Image Laboratory (SaIL), at the University of Arizona, I worked as a Graduate Assistant with Dr. Jeffrey J. Rodriguez. During this period, I worked on numerous projects ranging from image segmentation, object detection, object tracking, web development, and hardware integration. Key Responsibilities: • Primarily responsible for detection of objects of interest (zebrafish), from video data. • Solely responsible for tracking these objects of interest from frame to frame. • Solely responsible for maintaining custom built system software to integrate a network of Raspberry Pi computer controlled infrared cameras. • Primary focus on research and development of image analysis solutions. • Developed a machine learning solution to automate a zebrafish damage scoring process, that was otherwise done manually. • Developed an object detection and tracking solution using state of the art object detection, and tracking algorithms to identify, and track zebrafish in video. • Assisted a colleague (and learnt a lot in the process) in developing a network of interconnected Raspberry Pi computers. • Developed new theory to solve an issue that arises with performance metrics such as precision, recall, and F-score when analyzing object detection algorithms that produced split positives and merged positives. • Primarily responsible for writing journal articles, conference papers, and posters. • Aided my advisor in writing an NIH grant proposal. • Developed a proof-of-principle solution for a patent application related to zebrafish detection and tracking. • Simulated pharmacokinetic release profiles for various pharmacological compounds. • Performed curve fitting to a library of equations, and linear regression to analyze pharmacokinetic release profiles of various drugs. • Simulated multiple dosing profiles. Environment: OpenCV, Python, MATLAB, Perl, Javascript, Bash, MS Excel, MS Word, MS Powerpoint, LaTeX, Ubuntu, Windows, Mac OS X. Takeda Pharmaceuticals (formerly Millennium), Cambridge, MA May ’14 - Aug. ’14 Biomedical Imaging Intern As a member of the Biomedical Imaging Group (BIG), I was responsible for developing custom software to analyze the efficacy of treatment in preclinical imaging studies, using PET scans of tumors implanted in mice, and treated with pharmacological compounds in different stages of development. 3 Key Responsibilities: • Solely responsible for developing a tumor classification algorithm using machine learning. • Segmentation of tumors from PET scans. • Extraction of salient features from the segmented tumors, including size, shape, and texture. • Training a support vector machine classifier to perform classification of tumors. • Analysis of data every week. • Archiving, and maintaining data. • Presentation of results. Environment: MATLAB, Windows, MS Excel, MS Word, MS Powerpoint. Webseer Technology Private Limited, Bangalore, India Jun. ’12 – Jul.’ 13 Research Engineer As a research engineer at a startup in India, I was primarily responsible for research and development of image analysis, signal processing solutions, some web development, and basic programming. Key Responsibilities: • Performed image segmentation on a wide variety of natural and medical images. • Developed an algorithm to analyze color content in strawberries. • Performed k-means clustering to separate strawberries from background. • Analyzed MRI images to determine volume. Environment: MATLAB, Windows. Agama Solutions Inc., Fremont, CA Jul. ’10 – Feb.’12 Systems QA Analyst As a systems quality assurance analyst, I worked as a contractor with Wells Fargo, on developing a regression test bed for an internal messaging application. We also worked on a mainframe database application, and a web service to post monetary transactions. Key Responsibilities: • Create test cases to cover 85% of the messages used in Wells Fargo’s internal messaging system. • Solely responsible for developing an automated system to create test cases in a different testing environment that cut programming time by 50%. • Worked on a CSC Hogan mainframe. Environment: Windows, QTP, QC, VB Script, Excel, Javascript, Perl. 4 Arizona Cancer Center, Tucson, AZ Jan. ’09 – Jun.’10 Assistant Research Engineer As a member of the Imaging Core group, at the Arizona Cancer Center, I was responsible for research and development of image processing software, primarily aimed at segmentation of tumors in MRI images. Key Responsibilities: • Development of image processing solutions to analyze volume of tumors implanted in the liver of a rat, using diffusion weighted magnetic resonance imaging (MRI). • Extract features from tumors resulting in the ability to quantify response to treatment. Environment: Windows, MATLAB, IDL. University of Arizona, Tucson, AZ Aug. ’05 – Dec. ’08 Graduate Research Assistant As a member of the Signal and Image Laboratory (SaIL), at the University of Arizona, I worked as a graduate assistant with Dr. Jeffrey J. Rodriguez, and the Imaging Core group with Dr. Robert J. Gillies during my Master’s degree. During this period, I worked on segmenting tumors from immunohistochemical images, registering them in three dimensions, and creating a 3D view of the tumors, and surrounding vasculature. Key Responsibilities: • Solely responsible for segmenting tumors, and vasculature from immunohistochemical images of breast cancer tissue. • Solely responsible for 3D registration of the tumors. • Developed a novel seed pruning algorithm to reduce the number of seed points from which to perform region growing to segment the tumors, and surrounding vasculature. Environment: Windows, MATLAB, IDL. University of Arizona, Tucson, AZ Jan. ’06 – May. ’06 Graduate Teaching Assistant Worked as a teaching assistant with the lunar and planetary sciences department at the University of Arizona, teaching a foundational course in natural sciences, including one guest lecture. REFERENCES: Dr. Jeffrey J. Rodriguez, Associate Professor, Dept. of Electrical and Computer Engineering, University of Arizona. Dr. Michael Mayersohn, Professor, College of Pharmacy, University of Arizona. 5 .
Recommended publications
  • Management of Large Sets of Image Data Capture, Databases, Image Processing, Storage, Visualization Karol Kozak
    Management of large sets of image data Capture, Databases, Image Processing, Storage, Visualization Karol Kozak Download free books at Karol Kozak Management of large sets of image data Capture, Databases, Image Processing, Storage, Visualization Download free eBooks at bookboon.com 2 Management of large sets of image data: Capture, Databases, Image Processing, Storage, Visualization 1st edition © 2014 Karol Kozak & bookboon.com ISBN 978-87-403-0726-9 Download free eBooks at bookboon.com 3 Management of large sets of image data Contents Contents 1 Digital image 6 2 History of digital imaging 10 3 Amount of produced images – is it danger? 18 4 Digital image and privacy 20 5 Digital cameras 27 5.1 Methods of image capture 31 6 Image formats 33 7 Image Metadata – data about data 39 8 Interactive visualization (IV) 44 9 Basic of image processing 49 Download free eBooks at bookboon.com 4 Click on the ad to read more Management of large sets of image data Contents 10 Image Processing software 62 11 Image management and image databases 79 12 Operating system (os) and images 97 13 Graphics processing unit (GPU) 100 14 Storage and archive 101 15 Images in different disciplines 109 15.1 Microscopy 109 360° 15.2 Medical imaging 114 15.3 Astronomical images 117 15.4 Industrial imaging 360° 118 thinking. 16 Selection of best digital images 120 References: thinking. 124 360° thinking . 360° thinking. Discover the truth at www.deloitte.ca/careers Discover the truth at www.deloitte.ca/careers © Deloitte & Touche LLP and affiliated entities. Discover the truth at www.deloitte.ca/careers © Deloitte & Touche LLP and affiliated entities.
    [Show full text]
  • Abschlussarbeit Im Fachbereich Elektrotechnik & Informatik an Der
    Bachelorthesis Adriana Bostandzhieva Design and Implementation of System for Managing Training Data for Artificial Intelligence Algorithms Fakultät Technik und Informatik Faculty of Engineering and Computer Science Department Informations- und Department of Information and Elektrotechnik Electrical Engineering Adriana Bostandzhieva Design and Implementation of System for Managing Training Data for Artificial Intelligence Algorithms Bachelorthesisbased on the study regulations for the Bachelor of Engineering degree programme Information Engineering at the Department of Information and Electrical Engineering of the Faculty of Engineering and Computer Science of the Hamburg University of Aplied Sciences Supervising examiner : Prof. Dr. -Ing. Lutz Leutelt Second Examiner : Prof. Dr. Klaus Jünemann Day of delivery 3. Juli 2019 Adriana Bostandzhieva Title of the Bachelorthesis Design and Implementation of System for Managing Training Data for Artificial Intelli- gence Algorithms Keywords AI, training data, database, labels, video Abstract This paper is part of a pilot project of the Hamburg University of Applied Sciences. The project aims to utilise object detection algorithms and visual data to analyse complex road scenes. The aim of this thesis is to determine the best tool to use to label data for training artificial intelligence algorithms, to specify what data should be saved and to determine what database is to be used to save the data. The validity of the findings is proved by building a small prototype to showcase integration between the labelling tool and the database. Adriana Bostandzhieva Titel der Arbeit Entwicklung und Aufbau eines System zur Verwaltung von Trainingsdaten für Algo- rithmen der künstlichen Intelligenz Stichworte Trainingsdaten, Datenbanke, Video, KI Kurzzusammenfassung Diese Arbeit ist Teil eines Pilotprojekts der Hochschule für Angewandte Wissenschaf- ten Hamburg.
    [Show full text]
  • A 3D Interactive Multi-Object Segmentation Tool Using Local Robust Statistics Driven Active Contours
    A 3D interactive multi-object segmentation tool using local robust statistics driven active contours The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Gao, Yi, Ron Kikinis, Sylvain Bouix, Martha Shenton, and Allen Tannenbaum. 2012. A 3D Interactive Multi-Object Segmentation Tool Using Local Robust Statistics Driven Active Contours. Medical Image Analysis 16, no. 6: 1216–1227. doi:10.1016/j.media.2012.06.002. Published Version doi:10.1016/j.media.2012.06.002 Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:28548930 Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA NIH Public Access Author Manuscript Med Image Anal. Author manuscript; available in PMC 2013 August 01. NIH-PA Author ManuscriptPublished NIH-PA Author Manuscript in final edited NIH-PA Author Manuscript form as: Med Image Anal. 2012 August ; 16(6): 1216–1227. doi:10.1016/j.media.2012.06.002. A 3D Interactive Multi-object Segmentation Tool using Local Robust Statistics Driven Active Contours Yi Gaoa,*, Ron Kikinisb, Sylvain Bouixa, Martha Shentona, and Allen Tannenbaumc aPsychiatry Neuroimaging Laboratory, Brigham & Women's Hospital, Harvard Medical School, Boston, MA 02115 bSurgical Planning Laboratory, Brigham & Women's Hospital, Harvard Medical School, Boston, MA 02115 cDepartments of Electrical and Computer Engineering and Biomedical Engineering, Boston University, Boston, MA 02115 Abstract Extracting anatomical and functional significant structures renders one of the important tasks for both the theoretical study of the medical image analysis, and the clinical and practical community.
    [Show full text]
  • Open Source Computer Vision-Based Layer-Wise 3D Printing Analysis
    Open Source Computer Vision-based Layer-wise 3D Printing Analysis Aliaksei L. Petsiuk1 and Joshua M. Pearce1,2,3 1Department of Electrical & Computer Engineering, Michigan Technological University, Houghton, MI 49931, USA 2Department of Material Science & Engineering, Michigan Technological University, Houghton, MI 49931, USA 3Department of Electronics and Nanoengineering, School of Electrical Engineering, Aalto University, Espoo, FI-00076, Finland [email protected], [email protected] Graphical Abstract Highlights • Developed a visual servoing platform using a monocular multistage image segmentation • Presented algorithm prevents critical failures during additive manufacturing • The developed system allows tracking printing errors on the interior and exterior Abstract The paper describes an open source computer vision-based hardware structure and software algorithm, which analyzes layer-wise the 3-D printing processes, tracks printing errors, and generates appropriate printer actions to improve reliability. This approach is built upon multiple- stage monocular image examination, which allows monitoring both the external shape of the printed object and internal structure of its layers. Starting with the side-view height validation, the developed program analyzes the virtual top view for outer shell contour correspondence using the multi-template matching and iterative closest point algorithms, as well as inner layer texture quality clustering the spatial-frequency filter responses with Gaussian mixture models and segmenting structural anomalies with the agglomerative hierarchical clustering algorithm. This allows evaluation of both global and local parameters of the printing modes. The experimentally- verified analysis time per layer is less than one minute, which can be considered a quasi-real-time process for large prints. The systems can work as an intelligent printing suspension tool designed to save time and material.
    [Show full text]
  • Downloaded from the Cellprofiler Site [31] to Provide a Starting Point for New Analyses
    Open Access Software2006CarpenteretVolume al. 7, Issue 10, Article R100 CellProfiler: image analysis software for identifying and quantifying comment cell phenotypes Anne E Carpenter*, Thouis R Jones*†, Michael R Lamprecht*, Colin Clarke*†, In Han Kang†, Ola Friman‡, David A Guertin*, Joo Han Chang*, Robert A Lindquist*, Jason Moffat*, Polina Golland† and David M Sabatini*§ reviews Addresses: *Whitehead Institute for Biomedical Research, Cambridge, MA 02142, USA. †Computer Sciences and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02142, USA. ‡Department of Radiology, Brigham and Women's Hospital, Boston, MA 02115, USA. §Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02142, USA. Correspondence: David M Sabatini. Email: [email protected] Published: 31 October 2006 Received: 15 September 2006 Accepted: 31 October 2006 reports Genome Biology 2006, 7:R100 (doi:10.1186/gb-2006-7-10-r100) The electronic version of this article is the complete one and can be found online at http://genomebiology.com/2006/7/10/R100 © 2006 Carpenter et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. deposited research Cell<p>CellProfiler, image analysis the software first free, open-source system for flexible and high-throughput cell image analysis is described.</p> Abstract Biologists can now prepare and image thousands of samples per day using automation, enabling chemical screens and functional genomics (for example, using RNA interference). Here we describe the first free, open-source system designed for flexible, high-throughput cell image analysis, research refereed CellProfiler.
    [Show full text]
  • Survey of Databases Used in Image Processing and Their Applications
    International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 1 ISSN 2229-5518 Survey of Databases Used in Image Processing and Their Applications Shubhpreet Kaur, Gagandeep Jindal Abstract- This paper gives review of Medical image database (MIDB) systems which have been developed in the past few years for research for medical fraternity and students. In this paper, I have surveyed all available medical image databases relevant for research and their use. Keywords: Image database, Medical Image Database System. —————————— —————————— 1. INTRODUCTION Measurement and recording techniques, such as electroencephalography, magnetoencephalography Medical imaging is the technique and process used to (MEG), Electrocardiography (EKG) and others, can create images of the human for clinical purposes be seen as forms of medical imaging. Image Analysis (medical procedures seeking to reveal, diagnose or is done to ensure database consistency and reliable examine disease) or medical science. As a discipline, image processing. it is part of biological imaging and incorporates radiology, nuclear medicine, investigative Open source software for medical image analysis radiological sciences, endoscopy, (medical) Several open source software packages are available thermography, medical photography and for performing analysis of medical images: microscopy. ImageJ 3D Slicer ITK Shubhpreet Kaur is currently pursuing masters degree OsiriX program in Computer Science and engineering in GemIdent Chandigarh Engineering College, Mohali, India. E-mail: MicroDicom [email protected] FreeSurfer Gagandeep Jindal is currently assistant processor in 1.1 Images used in Medical Research department Computer Science and Engineering in Here is the description of various modalities that are Chandigarh Engineering College, Mohali, India. E-mail: used for the purpose of research by medical and [email protected] engineering students as well as doctors.
    [Show full text]
  • About This Document Opencv and Matlab Mex Files Recommended Knowledge General Concepts Mex Files
    1 About this document ABOUT THIS DOCUMENT This document was created as part of a final project for a BSc degree at the Academic College of Tel- Aviv Yaffo, by Rachely Esman and Yoad Snapir, under the supervision of Tal Hassner. As part of the project, we used Intel's OpenCV library, calling its functions from MATLAB. We found the subject tricky and thus decided to share the experience of our work with others describing the common pitfalls. You can use this document freely according to the copyrights noted below but under your sole responsibility. Rachely and Yoad. OPENCV AND MATLAB MEX FILES This guide will provide a quick walk through on compiling C++ OpenCV modules as MEX runtime files of MATLAB under: MS Visual Studio 2005 MATLAB 7.5.0 Windows 32bit And will probably be clear enough for other platforms / versions. RECOMMENDED KNOWLEDGE Good understanding of C++ Compilation and linkage concepts. DLL general usage Visual Studio 2005 environment Basic MATLAB and MEX files knowledge GENERAL CONCEPTS MEX FILES MEX files are actually simple .DLL files compiled with extension .MEXW32 using ordinary compilation tools provided with VS2005. Those .DLL files have a fixed entry point "mexFunction" with a fixed signature. (List of IN/OUT parameters) Follow this link to get an overview of MEX files and information on using them: http://www.mathworks.com/support/tech-notes/1600/1605.html#intro MEX files are compiled (and linked) from within the MATLAB environment using the following syntax: mex 'srcfile1.cpp' 'srcfile2.cpp' 'objfile1.obj' … Before compiling, you need to setup the compilation options using the command: Copyright © 2008 R.
    [Show full text]
  • Imagej Basics (Version 1.38)
    ImageJ Basics (Version 1.38) ImageJ is a powerful image analysis program that was created at the National Institutes of Health. It is in the public domain, runs on a variety of operating systems and is updated frequently. You may download this program from the source (http://rsb.info.nih.gov/ij/) or copy the ImageJ folder from the C drive of your lab computer. The ImageJ website has instructions for use of the program and links to useful resources. Installing ImageJ on your PC (Windows operating system): Copy the ImageJ folder and transfer it to the C drive of your personal computer. Open the ImageJ folder in the C drive and copy the shortcut (microscope with arrow) to your computer’s desktop. Double click on this desktop shortcut to run ImageJ. See the ImageJ website for Macintosh instructions. ImageJ Window: The ImageJ window will appear on the desktop; do not enlarge this window. Note that this window has a Menu Bar, a Tool Bar and a Status Bar. Menu Bar → Tool Bar → Status Bar → Graphics are from the ImageJ website (http://rsb.info.nih.gov/ij/). Adjusting Memory Allocation: Use the Edit → Options → Memory command to adjust the default memory allocation. Setting the maximum memory value to more than about 75% of real RAM may result in poor performance due to virtual memory "thrashing". Opening an Image File: Select File → Open from the menu bar to open a stored image file. Tool Bar: The various buttons on the tool bar allow you measure, draw, label, fill, etc. A right- click or a double left-click may expand your options with some of the tool buttons.
    [Show full text]
  • Paleoanthropology Society Meeting Abstracts, St. Louis, Mo, 13-14 April 2010
    PALEOANTHROPOLOGY SOCIETY MEETING ABSTRACTS, ST. LOUIS, MO, 13-14 APRIL 2010 New Data on the Transition from the Gravettian to the Solutrean in Portuguese Estremadura Francisco Almeida , DIED DEPA, Igespar, IP, PORTUGAL Henrique Matias, Department of Geology, Faculdade de Ciências da Universidade de Lisboa, PORTUGAL Rui Carvalho, Department of Geology, Faculdade de Ciências da Universidade de Lisboa, PORTUGAL Telmo Pereira, FCHS - Departamento de História, Arqueologia e Património, Universidade do Algarve, PORTUGAL Adelaide Pinto, Crivarque. Lda., PORTUGAL From an anthropological perspective, the passage from the Gravettian to the Solutrean is one of the most interesting transition peri- ods in Old World Prehistory. Between 22 kyr BP and 21 kyr BP, during the beginning stages of the Last Glacial Maximum, Iberia and Southwest France witness a process of substitution of a Pan-European Technocomplex—the Gravettian—to one of the first examples of regionalism by Anatomically Modern Humans in the European continent—the Solutrean. While the question of the origins of the Solutrean is almost as old as its first definition, the process under which it substituted the Gravettian started to be readdressed, both in Portugal and in France, after the mid 1990’s. Two chronological models for the transition have been advanced, but until very recently the lack of new archaeological contexts of the period, and the fact that the many of the sequences have been drastically affected by post depositional disturbances during the Lascaux event, prevented their systematic evaluation. Between 2007 and 2009, and in the scope of mitigation projects, archaeological fieldwork has been carried in three open air sites—Terra do Manuel (Rio Maior), Portela 2 (Leiria), and Calvaria 2 (Porto de Mós) whose stratigraphic sequences date precisely to the beginning stages of the LGM.
    [Show full text]
  • Anastasia Tyurina [email protected]
    1 Anastasia Tyurina [email protected] Summary A specialist in applying or creating mathematical methods to solving problems of developing technologies. A rare expert in solving problems starting from the stage of a stated “word problem” to proof of concept and production software development. Such successful uses of an educational background in mathematics, intellectual courage, and tenacious character include: • developed a unique method of statistical analysis of spectral composition in 1D and 2D stochastic processes for quality control in ultra-precision mirror polishing • developed novel methods of detection, tracking and classification of small moving targets for aerial IR and EO sensors. Used SIFT, and SIRF features, and developed innovative feature-signatures of motion of interest. • developed image processing software for bioinformatics, point source (diffraction objects) detection semiconductor metrology, electron microscopy, failure analysis, diagnostics, system hardware support and pattern recognition • developed statistical software for surface metrology assessment, characterization and generation of statistically similar surfaces to assist development of new optical systems • documented, published and patented original results helping employers technical communications • supported sales with prototypes and presentations • worked well with people – colleagues, customers, researchers, scientists, engineers Tools MATLAB, Octave, OpenCV, ImageJ, Scion Image, Aphelion Image, Gimp, PhotoShop, C/C++, (Visual C environment), GNU development tools, UNIX (Solaris, SGI IRIX), Linux, Windows, MS DOS. Positions and Experience Second Star Algonumerixs – 2008-present, founder and CEO http://www.secondstaralgonumerix.com/ 1) Developed a method of statistical assessment, characterisation and generation of random surface metrology for sper precision X-ray mirror manufacturing in collaboration with Lawrence Berkeley National Laboratory of University of California Berkeley.
    [Show full text]
  • Medical Images Research Framework
    Medical Images Research Framework Sabrina Musatian Alexander Lomakin Angelina Chizhova Saint Petersburg State University Saint Petersburg State University Saint Petersburg State University Saint Petersburg, Russia Saint Petersburg, Russia Saint Petersburg, Russia Email: [email protected] Email: [email protected] Email: [email protected] Abstract—with a growing interest in medical research problems for the development of medical instruments and to show and the introduction of machine learning methods for solving successful applications of this library on some real medical those, a need in an environment for integrating modern solu- cases. tions and algorithms into medical applications developed. The main goal of our research is to create medical images research 2. Existing systems for medical image process- framework (MIRF) as a solution for the above problem. MIRF ing is a free open–source platform for the development of medical tools with image processing. We created it to fill in the gap be- There are many open–source packages and software tween innovative research with medical images and integrating systems for working with medical images. Some of them are it into real–world patients treatments workflow. Within a short specifically dedicated for these purposes, others are adapted time, a developer can create a rich medical tool, using MIRF's to be used for medical procedures. modular architecture and a set of included features. MIRF Many of them comprise a set of instruments, dedicated takes the responsibility of handling common functionality for to solving typical tasks, such as images pre–processing medical images processing. The only thing required from the and analysis of the results – ITK [1], visualization – developer is integrating his functionality into a module and VTK [2], real–time pre–processing of images and video – choosing which of the other MIRF's features are needed in the OpenCV [3].
    [Show full text]
  • Downloadable At
    remote sensing Article LabelRS: An Automated Toolbox to Make Deep Learning Samples from Remote Sensing Images Junjie Li, Lingkui Meng, Beibei Yang, Chongxin Tao, Linyi Li and Wen Zhang * School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China; [email protected] (J.L.); [email protected] (L.M.); [email protected] (B.Y.); [email protected] (C.T.); [email protected] (L.L.) * Correspondence: [email protected]; Tel.: +86-027-68770771 Abstract: Deep learning technology has achieved great success in the field of remote sensing pro- cessing. However, the lack of tools for making deep learning samples with remote sensing images is a problem, so researchers have to rely on a small amount of existing public data sets that may influence the learning effect. Therefore, we developed an add-in (LabelRS) based on ArcGIS to help researchers make their own deep learning samples in a simple way. In this work, we proposed a feature merging strategy that enables LabelRS to automatically adapt to both sparsely distributed and densely distributed scenarios. LabelRS solves the problem of size diversity of the targets in remote sensing images through sliding windows. We have designed and built in multiple band stretching, image resampling, and gray level transformation algorithms for LabelRS to deal with the high spectral remote sensing images. In addition, the attached geographic information helps to achieve seamless conversion between natural samples, and geographic samples. To evaluate the reliability of LabelRS, we used its three sub-tools to make semantic segmentation, object detection and image classification samples, respectively.
    [Show full text]