DOCSLIB.ORG

Fast Timing Techniques

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Fast Timing Techniques Fast timing techniques A. Rivetti INFN -Sezione di Torino [email protected] June 2, 2016 A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 1 / 32 Outline 1 Introduction 2 Time digitizers 3 Jitter and time walk 4 CFD and ARC timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 2 / 32 Other well known applications of high resolution timing: Mass analysis with ToF mass spectrometry Positrom Emission Tomography: ToF reduces image noise With a peak luminosity of ≈ 1035, the HL-LHC will produce 140 to 200 collisions per bunch crossing I Disentangling interesting events from background only with tracking and vertexing becomes challenging I The average collision distance in time is 100 ÷ 170 ps The case for fast timing Traditionally, high resolution timing detectors are used in HEP to indentify particles I Measure the time to fly between two points to obtain the velocity I Combine with momentum information to derive the mass 2 I Large systems: ALICE ToF: 160 m . Similar area for the CBM ToF wall at FAIR A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 3 / 32 With a peak luminosity of ≈ 1035, the HL-LHC will produce 140 to 200 collisions per bunch crossing I Disentangling interesting events from background only with tracking and vertexing becomes challenging I The average collision distance in time is 100 ÷ 170 ps The case for fast timing Traditionally, high resolution timing detectors are used in HEP to indentify particles I Measure the time to fly between two points to obtain the velocity I Combine with momentum information to derive the mass 2 I Large systems: ALICE ToF: 160 m . Similar area for the CBM ToF wall at FAIR Other well known applications of high resolution timing: Mass analysis with ToF mass spectrometry Positrom Emission Tomography: ToF reduces image noise A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 3 / 32 The case for fast timing Traditionally, high resolution timing detectors are used in HEP to indentify particles I Measure the time to fly between two points to obtain the velocity I Combine with momentum information to derive the mass 2 I Large systems: ALICE ToF: 160 m . Similar area for the CBM ToF wall at FAIR Other well known applications of high resolution timing: Mass analysis with ToF mass spectrometry Positrom Emission Tomography: ToF reduces image noise With a peak luminosity of ≈ 1035, the HL-LHC will produce 140 to 200 collisions per bunch crossing I Disentangling interesting events from background only with tracking and vertexing becomes challenging I The average collision distance in time is 100 ÷ 170 ps A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 3 / 32 Forward physics at HL-LHC I Collision survivors can be used to probe new physics pp ! pγγp sensitive to extra-dimensions Intact protons detected 250 m far from the collision point Need of 10 ps timing to suppress pile-up A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 4 / 32 A typical timing system The sensor signal is usually amplified and shaped A comparator generates a digital pulse The threshold crossing time is captured and digitized by a TDC TDC can be embedded on the front-end chip or external I Timing is derived from a single sample A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 5 / 32 Outline 1 Introduction 2 Time digitizers 3 Jitter and time walk 4 CFD and ARC timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 6 / 32 High resolution TDCs: ASICs (1) A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 7 / 32 High resolution TDCs: ASICs (2) A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 8 / 32 High resolution TDCs: ASICs (3) A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 9 / 32 Non-HEP use of TDCs TDC used to measure phase difference in ADPLL With scaling technologies speed of gates increases Work in the time domain also to measure voltages K. Otsuga et al, IEEE International SoC Conference, 2012 A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 10 / 32 High resolution TDCs: FPGA (1) A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 11 / 32 High resolution TDCs: FPGA (2) A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 12 / 32 Waveform sampling TDCs The sensor signal is usually amplified and shaped The full waveform is sampled and digitized at high speed In many systems, sampling and digitization are decoupled Timing is extracted with DSP algorithms from the digitized waveform samples I Timing is derived from multiple samples A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 13 / 32 Waveform samplers: some example ASIC Year Node Time res. Max sample/ch. Channels LABRADOR3 2005 250 nm 16 ps 260 8 BLAB 2009 250 nm < 5 ps 65536 1 DRS4 2014 250 nm ≈ 1 ps 1024 8 PSEC4 2014 130 nm ≈ 1 ps 256 6 SamPic 2014 180 nm ≈ 3ps 64 16 Typical small channel count per ASIC Resolution: same pulse split and sent to differnt channels and time difference measured A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 14 / 32 Timing: yesterday and today D. Breton et al., NIM A 629 (2011) 123-132 A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 15 / 32 Trends in timing systems Time digitizers with ps resolution are a commodity Detectors with time resolution of 100 ps or better already achieved in the sixties of last century Typical ToF systems have low channel density Electronics either discrete or based on front-end ASICs with few channels Improve time resolution well below 100 ps (target 10 ps) Extend timing to densely packed detector systems. Need of highly integrated ASICs for timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 16 / 32 Timing: the issues Several factors challenge the timing accuracy of a system: Random noise internal to the front-end electronics (can be traded with power) Random noise from external sources (e.g. clock distribution system) Signal integrity (substrate noise, PSSR, etc..) Pulse amplitude variations Pulse shape variations I Timing below 100 ps rms is not trivial I Research is now geared towards sub 10 ps system resolution A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 17 / 32 Outline 1 Introduction 2 Time digitizers 3 Jitter and time walk 4 CFD and ARC timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 18 / 32 Timing jitter: single sample σ = σv dV ≈ V ! σ = tr t dV dt tr t SNR dt Checks: tr = 1 ns, SNR = 10 ! σt = 100 ps tr = 40 ns, SNR = 500 ! σt = 80 ps t / 1 SNR / p 1 ! σ / p 1 r BW BW t BW I Match the front-end rise time with the sensor rise/collection time A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 19 / 32 Timing jitter: multiple sampling Sample the input signal beyond Nyquist Assume first-order system relationship tr 1 tr σt = p N = SNR N ts q 1 0:35 1 p 1 σt = SNR BW ·f = SNR s 3f−3dB fs SNR fs f−3db σt 10 1 Gs/s 150 MHz 150 ps 10 10 Gs/s 1.5 GHz 15 ps 100 1 Gs/s 150 MHz 15 ps 1000 10 Gs/s 1.5 GHz 0.15 ps Redundacy is advantageous only if noise in uncorrelated I Unfortunately, jitter is not the full story... A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 20 / 32 Common solutions Correct using the pulse amplitude Correction usually done off-line Time-over-threshold often used Track in real-time the pulse with the threshold I Constant Fraction Timing Time walk Pulses of same shape and different amplitude crosses the threshold at different times Even worse if also the shape changes This is a problem for accurate timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 21 / 32 I Constant Fraction Timing Time walk Pulses of same shape and different amplitude crosses the threshold at different times Even worse if also the shape changes This is a problem for accurate timing Common solutions Correct using the pulse amplitude Correction usually done off-line Time-over-threshold often used Track in real-time the pulse with the threshold A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 21 / 32 Time walk Pulses of same shape and different amplitude crosses the threshold at different times Even worse if also the shape changes This is a problem for accurate timing Common solutions Correct using the pulse amplitude Correction usually done off-line Time-over-threshold often used Track in real-time the pulse with the threshold I Constant Fraction Timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 21 / 32 Outline 1 Introduction 2 Time digitizers 3 Jitter and time walk 4 CFD and ARC timing A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 22 / 32 The input signal is both delayed and attenuated The delayed and attenuated signals are combined to yield a bipolar waveform The zero crossing of the bipolar waveform is used for timing CFD: the principle A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 23 / 32 The delayed and attenuated signals are combined to yield a bipolar waveform The zero crossing of the bipolar waveform is used for timing CFD: the principle The input signal is both delayed and attenuated A. Rivetti (INFN-Torino) Fast timing techniques June 2, 2016 23 / 32 The zero crossing of the bipolar waveform is used for timing CFD: the principle The input signal is both delayed and attenuated The delayed and attenuated signals are combined to yield a bipolar waveform A.
Load more

Recommended publications
  • Survey of Noise in Image and Efficient Technique for Noise Reduction
    International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2015): 78.96 | Impact Factor (2015): 6.391 Survey of Noise in Image and Efficient Technique for Noise Reduction Arti Singh1, Madhu2 1, 2Research Scholar, Computer Science, BBAU University, Lucknow , Uttar Pradesh, India Abstract: Removing noise from the image is big challenge for researcher because removal of noise in image causes the artifacts and image blurring. Noise occurred in image during the time of capturing and transmission of the image. There are many methods for noise removal from the images. Many algorithms and techniques are available for removing noise from image, but each method exist their own assumptions, merits and demerits. Noise reduction algorithms for remove noise is totally depends on what type of noise occur in the image. In this paper, focus on some important type of noise and noise removal techniques is done. Keywords: Impulse noise, Gaussian noise , Speckle noise ,Poisson noise, mean filter, median filter ,etc 1. Introduction or over heated faulty component can cause noise arise in image because of sharp and sudden changes of Digital image processing is using of algorithms for image signal. 퐼 푖, 푗 푥 < 푙 improve quality order of digital image .This error on 퐼푠푝 푖, 푗 = image is called noise in image which do not reflect real 퐼푚푖푛 + 푌 퐼푚푎푥 −퐼푚푖푛 푥 > 푙 intensities of actual scene. There are two problems found x,y∈[0,1] are two uniformly distributed random variables in image processing: Blurring and image noise. Noise image occurred by many reasons: When we capture image from camera (scratches are available in camera).
    [Show full text]
  • Performance Analysis of Spatial and Transform Filters for Efficient Image Noise Reduction
    Performance Analysis of Spatial and Transform Filters for Efficient Image Noise Reduction Santosh Paudel Ajay Kumar Shrestha Pradip Singh Maharjan Rameshwar Rijal Computer and Electronics Computer and Electronics Computer and Electronics Computer and Electronics Engineering Engineering Engineering Engineering KEC, TU KEC, TU KEC, TU KEC, TU Lalitpur, Nepal Lalitpur, Nepal Lalitpur, Nepal Lalitpur, Nepal [email protected] [email protected] [email protected] [email protected] Abstract—During the acquisition of an image from its systems such as laser, acoustics and SAR (Synthetic source, noise always becomes integral part of it. Various Aperture Radar) images [3]. algorithms have been used in past to denoise the images. Image denoising still has scope for improvement. Visual This paper introduces different types of noise to be information transmitted in the form of digital images has considered in an image and analyzed for various spatial become a considerable method of communication in the and transforms domain filters by considering the image modern age, but the image obtained after transmission is metrics such as mean square error (MSE), root mean often corrupted due to noise. In this paper, we review the squared error (RMSE), Peak Signal to Noise Ratio existing denoising algorithms such as filtering approach (PSNR) and universal quality index (UQI). and wavelets based approach, and then perform their comparative study with bilateral filters. We use different II. BACKGROUND noise models to describe additive and multiplicative noise The Wavelet Transform (WT) is a powerful tool of in an image. Based on the samples of degraded pixel signal and image processing, which has been neighborhoods as inputs, the output of an efficient filtering successfully used in many scientific fields such as signal approach has shown a better image denoising performance.
    [Show full text]
  • Nasa Tm- 77750 Nasa Technical Memorandum Nasa Tm-77750
    NASA TM- 77750 NASA TECHNICAL MEMORANDUM NASA TM-77750 NASA-TM-77750 19850004171 PATTERNS OF BEHAVIOD<TN LODGINGS EXPOSED TO TRAFFIC NOISE Jacques Lambert, Francois Simonnet Translation of "Comportements dans l'habitat soumis au bruit de circulation". Institut de Recherche des Transports, Arcueil, France, Rapport de Recherche I.R.T. No. 47, September, 1980, pp 1-145. 11.,. NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON D.C. 20546 NOVEMBER 1984 • • \ IT........ f.n.• PM. ,. II. 0 ... I. .M.,...·.c.......... NAS1\. TM.,-77750 .. , ...... S....... PATTERNS or' BEHAVIOR IN I. • .,.,. hie November, 1984 TF~FFIC LODGINGS EXPOSED TO NOISE • •. ,.,te-t". 0, ee. 70 A.......c.. I. ''''e''''. 0. H.. Jacques Lambert, Francois Simonnet 11...... "-'..... '. 1-------------------------1... e.......... 0......... t. ' .......... 0,.......'... N... et4 ........ MAS... .~C; 42 SCITRAN • lox S4S6 .' II. ,,,..,......., ...c.....4 r.__ • a _..,.... ClI·un. 'rraul.t1oll, 12. SU4t1~&:r"&;rD_==_ .. Sp.at MaiIliat~.t.io.....-----------..f VUD1qtOD. D~Ce ~0546 No Ate-f c... I'" ...........,.......Translatlon. .. of "COITInortements dans l'habltat. soumis au bruit de circulation".'" Institut de·Recherche des Transports, Arcuei1, France, Rapport de Recherche' I ~R •. ~.· No. 47, September, 1980, pp. 1-145. , .. M ......· Thresho1c values at which public services should intervene ~o attenuate the noise nuisance are defined. Observations were made in the field of daily life at ..­ home. Data was collecte<J. on the use of loe1gings, on effects of noise on health and sleep~ and on the incidence of running away from home. A correlation was made also with the equipment. and noise insulation of lodgings. The results s.how that abov.eGG dB in daytime, there are behavior patterns that are extreme so far as they modify in a considerable manner the way bf, life of-people, living in both collective housing Capartments) and in individual houses • • ~.
    [Show full text]
  • Noise Assessment Activities
    Noise assessment activities Interesting stories in Europe ETC/ACM Technical Paper 2015/6 April 2016 Gabriela Sousa Santos, Núria Blanes, Peter de Smet, Cristina Guerreiro, Colin Nugent The European Topic Centre on Air Pollution and Climate Change Mitigation (ETC/ACM) is a consortium of European institutes under contract of the European Environment Agency RIVM Aether CHMI CSIC EMISIA INERIS NILU ÖKO-Institut ÖKO-Recherche PBL UAB UBA-V VITO 4Sfera Front page picture: Composite that includes: photo of a street in Berlin redesigned with markings on the asphalt (from SSU, 2014); view of a noise barrier in Alverna (The Netherlands)(from http://www.eea.europa.eu/highlights/cutting-noise-with-quiet-asphalt), a page of the website http://rumeur.bruitparif.fr for informing the public about environmental noise in the region of Paris. Author affiliation: Gabriela Sousa Santos, Cristina Guerreiro, Norwegian Institute for Air Research, NILU, NO Núria Blanes, Universitat Autònoma de Barcelona, UAB, ES Peter de Smet, National Institute for Public Health and the Environment, RIVM, NL Colin Nugent, European Environment Agency, EEA, DK DISCLAIMER This ETC/ACM Technical Paper has not been subjected to European Environment Agency (EEA) member country review. It does not represent the formal views of the EEA. © ETC/ACM, 2016. ETC/ACM Technical Paper 2015/6 European Topic Centre on Air Pollution and Climate Change Mitigation PO Box 1 3720 BA Bilthoven The Netherlands Phone +31 30 2748562 Fax +31 30 2744433 Email [email protected] Website http://acm.eionet.europa.eu/ 2 ETC/ACM Technical Paper 2015/6 Contents 1 Introduction ...................................................................................................... 5 2 Noise Action Plans .........................................................................................
    [Show full text]
  • Analysis of Image Noise in Multispectral Color Acquisition
    ANALYSIS OF IMAGE NOISE IN MULTISPECTRAL COLOR ACQUISITION Peter D. Burns Submitted to the Center for Imaging Science in partial fulfillment of the requirements for Ph.D. degree at the Rochester Institute of Technology May 1997 The design of a system for multispectral image capture will be influenced by the imaging application, such as image archiving, vision research, illuminant modification or improved (trichromatic) color reproduction. A key aspect of the system performance is the effect of noise, or error, when acquiring multiple color image records and processing of the data. This research provides an analysis that allows the prediction of the image-noise characteristics of systems for the capture of multispectral images. The effects of both detector noise and image processing quantization on the color information are considered, as is the correlation between the errors in the component signals. The above multivariate error-propagation analysis is then applied to an actual prototype system. Sources of image noise in both digital camera and image processing are related to colorimetric errors. Recommendations for detector characteristics and image processing for future systems are then discussed. Indexing terms: color image capture, color image processing, image noise, error propagation, multispectral imaging. Electronic Distribution Edition 2001. ©Peter D. Burns 1997, 2001 All rights reserved. COPYRIGHT NOTICE P. D. Burns, ‘Analysis of Image Noise in Multispectral Color Acquisition’, Ph.D. Dissertation, Rochester Institute of Technology, 1997. Copyright © Peter D. Burns 1997, 2001 Published by the author All rights reserved. No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form, or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the copyright holder.
    [Show full text]
  • Local Noise Action Plans
    Practitioner Handbook for Local Noise Action Plans Recommendations from the SILENCE project SILENCE is an Integrated Project co-funded by the European Commission under the Sixth Framework Programme for R&D, Priority 6 Sustainable Development, Global Change and Ecosystems Guidance for readers Step 1: Getting started – responsibilities and competences • These pages give an overview on the steps of action planning and Objective To defi ne a leader with suffi cient capacities and competences to the noise abatement measures and are especially interesting for successfully setting up a local noise action plan. To involve all relevant stakeholders and make them contribute to the implementation of the plan clear competences with the leading department are needed. The END ... DECISION MAKERS and TRANSPORT PLANNERS. Content Requirements of the END and any other national or The current responsibilities for noise abatement within the local regional legislation regarding authorities will be considered and it will be assessed whether these noise abatement should be institutional settings are well fi tted for the complex task of noise considered from the very action planning. It might be advisable to attribute the leadership to beginning! another department or even to create a new organisation. The organisational settings for steering and carrying out the work to be done will be decided. The fi nancial situation will be clarifi ed. A work plan will be set up. If support from external experts is needed, it will be determined in this stage. To keep in mind For many departments, noise action planning will be an additional task. It is necessary to convince them of the benefi ts and the synergies with other policy fi elds and to include persons in the steering and working group that are willing and able to promote the issue within their departments.
    [Show full text]
  • Measurement of Noise and Resolution in PET 1071 Large ROI in a Single Static Image
    IOP PUBLISHING PHYSICS IN MEDICINE AND BIOLOGY Phys. Med. Biol. 55 (2010) 1069–1081 doi:10.1088/0031-9155/55/4/011 Simultaneous measurement of noise and spatial resolution in PET phantom images Martin A Lodge1, Arman Rahmim1 and Richard L Wahl1,2 1 Division of Nuclear Medicine, The Russell H. Morgan Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, USA 2 Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Johns Hopkins University School of Medicine, Baltimore, MD, USA E-mail: [email protected] Received 31 August 2009, in final form 21 December 2009 Published 28 January 2010 Online at stacks.iop.org/PMB/55/1069 Abstract As an aid to evaluating image reconstruction and correction algorithms in positron emission tomography, a phantom procedure has been developed that simultaneously measures image noise and spatial resolution. A commercially available 68Ge cylinder phantom (20 cm diameter) was positioned in the center of the field-of-view and two identical emission scans were sequentially performed. Image noise was measured by determining the difference between corresponding pixels in the two images and by calculating the standard deviation of these difference data. Spatial resolution was analyzed using a Fourier technique to measure the extent of the blurring at the edge of the phantom images. This paper addresses the noise aspects of the technique as the spatial resolution measurement has been described elsewhere. The noise measurement was validated by comparison with data obtained from multiple replicate images over a range of noise levels. In addition, we illustrate how simultaneous measurement of noise and resolution can be used to evaluate two different corrections for random coincidence events: delayed event subtraction and singles-based randoms correction.
    [Show full text]
  • Interaction of Image Noise, Spatial Resolution, and Low Contrast Fine
    Interaction of image noise, spatial resolution, and low contrast fine detail preservation in digital image processing Uwe Artmanna and Dietmar Wuellerb a,bImage Engineering, Augustinusstrasse 9d, 50226 Frechen, Germany; ABSTRACT We present a method to improve the validity of noise and resolution measurements on digital cameras. If non-linear adaptive noise reduction is part of the signal processing in the camera, the measurement results for image noise and spatial resolution can be good, while the image quality is low due to the loss of fine details and a watercolor like appearance of the image. To improve the correlation between objective measurement and subjective image quality we propose to supplement the standard test methods with an additional measurement of the texture preserving capabilities of the camera. The proposed method uses a test target showing white Gaussian noise. The camera under test reproduces this target and the image is analyzed. We propose to use the kurtosis of the derivative of the image as a metric for the texture preservation of the camera. Kurtosis is a statistical measure for the closeness of a distribution compared to the Gaussian distribution. It can be shown, that the distribution of digital values in the derivative of the image showing the chart becomes the more leptokurtic (increased kurtosis) the stronger the noise reduction has an impact on the image. Keywords: Noise, Noise Reduction, Texture, Resolution, Spatial Frequency, Kurtosis, MTF, SFR 1. INTRODUCTION ColorFoto is a German photography magazine with a focus on objective and complex tests on digital still camera systems. Since we started testing in 1997, the tests had to be adjusted from time to time to keep track with the development in the camera market, so the test results correlate with the subjective image quality, experienced by the user.
    [Show full text]
  • Arxiv:1701.01924V1 [Cs.CV] 8 Jan 2017 Charge Coupled Device (CCD) Inside the Camera
    ON CLASSIFICATION OF DISTORTED IMAGES WITH DEEP CONVOLUTIONAL NEURAL NETWORKS Yiren Zhou, Sibo Song, Ngai-Man Cheung Singapore University of Technology and Design ABSTRACT Some previous works have studied the effect of image distor- Image blur and image noise are common distortions during im- tion [10]. Focusing on DNN, Basu et al. [11] proposed a new model age acquisition. In this paper, we systematically study the effect of modified from deep belief nets to deal with noisy inputs. They re- image distortions on the deep neural network (DNN) image classi- ported good results on a noisy dataset called n-MNIST, which con- fiers. First, we examine the DNN classifier performance under four tains Gaussian noise, motion blur, and reduced contrast compared to types of distortions. Second, we propose two approaches to allevi- original MNIST dataset. Recently, Dodge and Karam [12] reported ate the effect of image distortion: re-training and fine-tuning with the degradation due to various image distortions in several DNN. noisy images. Our results suggest that, under certain conditions, Compared to these works, we perform a unified study to investigate fine-tuning with noisy images can alleviate much effect due to dis- effect of image distortion on (i) hand-written digit classification and torted inputs, and is more practical than re-training. (ii) natural image classification. Moreover, we examine using re- training and fine-tuning with noisy images to alleviate the effect. Index Terms— Image blur; image noise; deep convolutional In classification of “clean” images (i.e., without distortion), neural networks; re-training; fine-tuning some previous work has attempted to introduce noise to the train- 1.
    [Show full text]
  • 233 Noisy Image Classification Using Hybrid Deep Learning
    Journal of ICT, 18, No. 2 (April) 2018, pp: 233–269 How to cite this paper: Roy, S. S., Ahmed, M., & Akhand, M. A. H. (2018). Noisy image classification using hybrid deep learning methods. Journal of Information and Communication Technology, 17 (2), 233–269. NOISY IMAGE CLASSIFICATION USING HYBRID DEEP LEARNING METHODS 1Sudipta Singha Roy, 2Mahtab Ahmed & 2Muhammad Aminul Haque Akhand 1 Institute of Information and Communication Technology Khulna University of Engineering & Technology, Khulna, Bangladesh 2 Dept. of Computer Science and Engineering Khulna University of Engineering & Technology, Khulna, Bangladesh [email protected]; [email protected]; [email protected] ABSTRACT In real-world scenario, image classification models degrade in performance as the images are corrupted with noise, while these models are trained with preprocessed data. Although deep neural networks (DNNs) are found efficient for image classification due to their deep layer-wise design to emulate latent features from data, they suffer from the same noise issue. Noise in image is common phenomena in real life scenarios and a number of studies have been conducted in the previous couple of decades with the intention to overcome the effect of noise in the image data. The aim of this study was to investigate the DNN-based better noisy image classification system. At first, the autoencoder (AE)-based denoising techniques were considered to reconstruct native image from the input noisy image. Then, convolutional neural network (CNN) is employed to classify the reconstructed image; as CNN was a prominent DNN method with the ability to preserve better representation of the internal structure of the image data.
    [Show full text]
  • Vessel Motion Extraction from an Image Sequence
    Vessel Motion Extraction from an Image Sequence Esteban L. Castro-Feliciano Naval Architecture and Marine Engineering Dept. University of Michigan Ann Arbor, Michigan Email: [email protected] Abstract—The abstract goes here. vessel. Because the experiments recorded only the acceleration data, the velocity and displacement are estimated by offline I. INTRODUCTION processing. The use of motion data from a vehicle is common in all III. METHODOLOGY vehicle research fields. For marine applications, the motion of a vessel is recorded both in full and model scale tests. A. Vessel Motion DSP This motion information is used to measure the performance and safety of the vessel in a seaway. The method commonly One can see in the FFT of the signal in Fig. 4 that there used to record the vessel motion is to use accelerometers and is a lot of low frequency content, which causes non-realistic gyros to measure the rotational and translational accelerations drift in the velocity and displacement estimations (these lower and speeds. Then, data post-processing is used to estimate the frequencies are magnified when integrating). The method used vessel’s position [1]. However, if the actual displacements were in this project to estimate the velocity and displacement values not measured from a fixed frame of reference, it is very difficult is to filter and integrate the acceleration data in frequency – if not impossible – to validate the post-processing results. domain [4]. The filter used is a high-pass Sinc filter with a cut-off of 0.15Hz. This cut-off frequency appears to be a good With the wide availability of video cameras, it is an compromise between reducing drift and preserving the rigid- attractive idea to simply record the vessel tests, and extract body velocity and displacements especially at points where the motion from the sequence of images.
    [Show full text]
  • Medical Image Denoising Using Convolutional Denoising Autoencoders
    Medical image denoising using convolutional denoising autoencoders Lovedeep Gondara Department of Computer Science Simon Fraser University [email protected] Abstract—Image denoising is an important pre-processing step image denoising performance for their ability to exploit strong in medical image analysis. Different algorithms have been pro- spatial correlations. posed in past three decades with varying denoising performances. More recently, having outperformed all conventional methods, In this paper we present empirical evidence that stacked deep learning based models have shown a great promise. These denoising autoencoders built using convolutional layers work methods are however limited for requirement of large training well for small sample sizes, typical of medical image sample size and high computational costs. In this paper we show databases. Which is in contrary to the belief that for optimal that using small sample size, denoising autoencoders constructed performance, very large training datasets are needed for models using convolutional layers can be used for efficient denoising based on deep architectures. We also show that these methods of medical images. Heterogeneous images can be combined to can recover signal even when noise levels are very high, at the boost sample size for increased denoising performance. Simplest point where most other denoising methods would fail. of networks can reconstruct images with corruption levels so high that noise and signal are not differentiable to human eye. Rest of this paper is organized as following, next section Keywords—Image denoising, denoising autoencoder, convolu- discusses related work in image denoising using deep architec- tional autoencoder tures. Section III introduces autoencoders and their variants. Section IV explains our experimental set-up and details our empirical evaluation and section V presents our conclusions I.
    [Show full text]