Compressive Sensing. Examples in Image Compression. Lecture 4, July 30, Luiz Velho Eduardo A. B. da Silva Adriana Schulz

Size: px
Start display at page:

Download "Compressive Sensing. Examples in Image Compression. Lecture 4, July 30, Luiz Velho Eduardo A. B. da Silva Adriana Schulz"

Transcription

1 Compressive Sensing Examples in Image Compression Lecture 4, July, 09 Luiz Velho Eduardo A. B. da Silva Adriana Schulz

2 Today s Lecture Discuss applications of CS in image compression Evaluate CS efficiency Review important definitions and theorems

3 Software Matlab functions for CS imaging Optimization algorithms Data Demos aschulz/cs/course.html

4 Experimental Setup 4 images of size (a) Phantom (b) Lena (c) Camera man (d) Text

5 Frequency Distributions (a) Phantom (b) Lena (c) Camera man (d) Text

6 Acquisition Images are stored as a matrix of pixels

7 Noiselets RIP orthogonal self-adjoint Φ =

8 Sparsity 1 0 2

9 Reconstruction Acquisition: y = Φ Ω x

10 Reconstruction Acquisition: y = Φ Ω x

11 Reconstruction Acquisition: y = Φ Ω x Reconstruction: y = Θ Ω s s = min s s l1 s. t. x = Ψ s { y = Θs y Θs l2 ɛ

12 Reconstruction Acquisition: y = Φ Ω x Reconstruction: y = Θ Ω s s = min s s l1 s. t. x = Ψ s { y = Θs y Θs l2 ɛ

13 Basic CS Images are sparse in the Ψ domain No noise is added Ordinary images are only approximately sparse : (a)

14 Basic CS Images are sparse in the Ψ domain No noise is added Ordinary images are only approximately sparse : (a) (b) Figure: Different visualizations of Lena s DCT.

15 Forcing Sparsity DCT Example: (a) Original Image (b) Original DCT (c) Original DCT (d) Result Image (e) Result DCT (f) Result DCT

16 First Experiment x Ψ s force sparsity Ψ s S x S Acquisition: y = Φ Ω x S Reconstruction of x via l 1 minimization Compare x and x S using (Peak Signal to Noise Ratio)

17 First Result Measurements( 3 ) Figure: Basic CS experiment: k-sparse Lena. Theorem Reconstruction is exact if M S µ 2 (Φ, Ψ) log N

18 Varying S Measurements( 3 ) 3.5k Sparse 6k Sparse k Sparse 14k Sparse

19 DCT Linear Compression

20 DCT Linear Compression

21 DCT Linear Compression We measure M N coefficients on the upper-left corner

22 DCT Linear Compression We measure M N coefficients on the upper-left corner

23 DCT Linear Compression Reconstruction: x = Ψ s

24 DCT Linear Compression

25 DCT Linear Compression Problem: we don t measure all the significant coefficients and measure some zero coefficients!

26 CS: Linear + Noiselets Measurements Reconstruction: s = min s s l1 s. t. Φ Ω Ψ s = y x = Ψ s

27 Results: CS Linear Compression [DCT l N] [DCT l L] [DCT l LN] Measurements( 3 ) Measurements( 3 ) (a) 3.5K-Sparse (b) 6K-Sparse Measurements( 3 ) Measurements( 3 ) (c) K-Sparse (d) 14K-Sparse

28 Coherence Before Boundary: best to use linear compression After Boundary: best to use CS Measurements( 3 ) DCT l 1 N DCT l 1 LN

29 Different Images What happens when we consider the other images? Measurements( 3 ) Measurements( 3 ) (e) Linear Compression (f) Compressive Sensing (a) Phantom (b) Lena (c) Camera man (d) Text

30 Different Images What happens when we consider the other images? 3 (f) Compressive Sensing (a) Phantom 2 0 Measurements( ) (e) Linear Compression 0 3 Measurements( ) 1 (b) Lena (c) Camera man 0 1 (d) Text 0 2

31 Different Images What happens when we consider the other images? phantom lena cameraman text Measurements( 3) (a) Linear Compression (a) Phantom 2 0 (b) Compressive Sensing 0 0 Measurements( 3) 1 (b) Lena (c) Camera man 0 1 (d) Text 0 2

32 Different Sparsity Domains What happens when we consider different sparsity domains? nz = 00 (a) DCT nz = 00 (b) Block DCT nz = 00 (c) Wavelets

33 Different Sparsity Domains What happens when we consider different sparsity domains? nz = 00 (a) DCT nz = 00 (b) Block DCT nz = 00 (c) Wavelets Measurements( 3 ) DCT l 1 N B_DCT l 1 N DWT l 1 N

34 But this is not for real... All this happens because we FORCE sparsity! Theorem Reconstruction is exact if M S µ 2 (Φ, Ψ) log N Important Parameters: µ - Noiselets are highly incoherent with all considered domains S - We are tempering with it! - Lets stop and see what happens...

35 Sparsity Errors When Φ are Noiselet measurements the RIP holds and we can use: Theorem Let s S be the best S-sparse approximation of s. If the RIP holds, the solution s to s = min s s l1 subject to Θ Ω s = y obeys s s l2 1 S s s S l1

36 Let s verify this theorem empirically

37 Let s verify this theorem empirically Measurements( 3 ) k Sparse 6k Sparse k Sparse 14k Sparse Measurements( 3 ) Test image Lena Measurements Sparsity is forced Sparsity is not forced M = k S = 3.5k = 28.8 = 26.6 M = 25k S = 6k =.7 = 27.8 M = 35k S = k = 33.0 =.2 M = k S = 14k = 34.9 = 31.5

38 Let s verify this theorem empirically Measurements( 3 ) k Sparse 6k Sparse k Sparse 14k Sparse Measurements( 3 ) Test image Lena Measurements Sparsity is forced Sparsity is not forced M = k S = 3.5k = 28.8 = 26.6 M = 25k S = 6k =.7 = 27.8 M = 35k S = k = 33.0 =.2 M = k S = 14k = 34.9 = 31.5

39 Different Sparsity Domains [DCT l L] 2 [B_DCT l L] 2 [DCT l N] 1 [B_DCT l N] 1 [DWT l N] 1 [SVD l N] 1 [TV l N] 1 Measurements( 3 ) Measurements( 3 ) (b) Phantom (c) Lena Measurements( 3 ) Measurements( 3 ) (d) Camera man (e) Text

40 Different Sparsity Domains [DCT l L] 2 [B_DCT l L] 2 [DCT l N] 1 [B_DCT l N] 1 [DWT l N] 1 [SVD l N] 1 [TV l N] 1 Measurements( 3 ) Measurements( 3 ) (b) Phantom (c) Lena Measurements( 3 ) Measurements( 3 ) (d) Camera man (e) Text

41 Different Sparsity Domains [DCT l L] 2 [B_DCT l L] 2 [DCT l N] 1 [B_DCT l N] 1 [DWT l N] 1 [SVD l N] 1 [TV l N] 1 Measurements( 3 ) Measurements( 3 ) (b) Phantom (c) Lena Measurements( 3 ) Measurements( 3 ) (d) Camera man (e) Text

42 Different Sparsity Domains Total Variation (TV) min x x TV subject to y = Φ Ω x l 1 -norm of the (appropriately discretized) gradient.

43 Different Sparsity Domains [DCT l L] 2 [B_DCT l L] 2 [DCT l N] 1 [B_DCT l N] 1 [DWT l N] 1 [SVD l N] 1 [TV l N] 1 Measurements( 3 ) Measurements( 3 ) (b) Phantom (c) Lena Measurements( 3 ) Measurements( 3 ) (d) Camera man (e) Text

44 Different Sparsity Domains Singular Value Decomposition (SVD) Calculated for each image! Upper bounds - insights into performance limitations.

45 Different Sparsity Domains [DCT l L] 2 [B_DCT l L] 2 [DCT l N] 1 [B_DCT l N] 1 [DWT l N] 1 [SVD l N] 1 [TV l N] 1 Measurements( 3 ) Measurements( 3 ) (b) Phantom (c) Lena Measurements( 3 ) Measurements( 3 ) (d) Camera man (e) Text

46 Measurement Errors Measurements: y = Φ Ω x + n n is a random variable with normal distribution Theorem If the RIP holds and n l2 ɛ, the solution s to s = min s s l1 s.t. Θ Ω s y l2 ɛ obeys s s l2 1 S s s S l1 + ɛ

47 Gaussian Errors Measurements( 3 ) var = 0.1 var = 0.5 var = 1 var = 2 var = 3 var = 4 var = 5 var =

48 Compression Number of Measurements Number of Bits To compress we have to quantize the acquired data!

49 Quantization Each measurement may assume K values.

50 Quantization Larger quatization steps: less bits Smaller quatization steps: less errors

51 Rate Rate = number of bits per pixel Original Image: N pixels 8 bits/pixel Acquired Measurements: M N coefficients H y bits/coefficients Rate = M N H y bits/pixel If Rate < 8 data is being compressed!

52 The Rate Distortion Criteria

53 The Rate Distortion Criteria = which is the best compression scheme?

54 Rate Distortion 90 Measurements x 3 step = 0.01 step = 0.02 step = 0.05 step = 0.1 step = 0.2 step = 0.5 step = 1 step = 2 step = 3 step = 4 step = 5 step = step = step = step =

55 Rate Distortion Rate step = 0.01 step = 0.02 step = 0.05 step = 0.1 step = 0.2 step = 0.5 step = 1 step = 2 step = 3 step = 4 step = 5 step = step = step = step =

56 Rate Distortion Rate step = 0.01 step = 0.02 step = 0.05 step = 0.1 step = 0.2 step = 0.5 step = 1 step = 2 step = 3 step = 4 step = 5 step = step = step = step =

57 Rate Distortion

58 Rate Distortion Rate step = 0.01 step = 0.02 step = 0.05 step = 0.1 step = 0.2 step = 0.5 step = 1 step = 2 step = 3 step = 4 step = 5 step = step = step = step =

59 Both Sparsity and Measurement Errors A real example: real image and quantized measurements. Recovery Strategies: Ψ = DCT Ψ = block DCT Ψ = DWT TV- minimization Ψ = SVD - of each image!

60 Both Sparsity and Measurement Errors Jpeg00 DCT l N B_DCT l N DWT l N TV N SVD l N Rate Rate (b) Phantom (c) Lena Rate Rate (d) Camera man (e) Text

61 Sparsity Quantization Errors x x l2 ɛ q + 1 x S x l1 S, }{{} ɛ s The reconstruction error in CS is of the order of the maximum of the quantization (ɛ q ) and sparsity errors (ɛ s ).

62 (a) DCT (b) DWT (c) TV (d) SVD Figure: Rate of image Lena

63 (a) DCT (b) DWT (c) TV (d) SVD Figure: Measurements of image Lena

64 Lets wrap it up! CS Applications Image acquisition and compression Interesting, but Limitations... Tomorow Applications in graphics and vision One Pixel Camera Dual Photography

JPEG Image Compression by Using DCT

JPEG Image Compression by Using DCT International Journal of Computer Sciences and Engineering Open Access Research Paper Volume-4, Issue-4 E-ISSN: 2347-2693 JPEG Image Compression by Using DCT Sarika P. Bagal 1* and Vishal B. Raskar 2 1*

More information

Part II Redundant Dictionaries and Pursuit Algorithms

Part II Redundant Dictionaries and Pursuit Algorithms Aisenstadt Chair Course CRM September 2009 Part II Redundant Dictionaries and Pursuit Algorithms Stéphane Mallat Centre de Mathématiques Appliquées Ecole Polytechnique Sparsity in Redundant Dictionaries

More information

Sparse recovery and compressed sensing in inverse problems

Sparse recovery and compressed sensing in inverse problems Gerd Teschke (7. Juni 2010) 1/68 Sparse recovery and compressed sensing in inverse problems Gerd Teschke (joint work with Evelyn Herrholz) Institute for Computational Mathematics in Science and Technology

More information

Sublinear Algorithms for Big Data. Part 4: Random Topics

Sublinear Algorithms for Big Data. Part 4: Random Topics Sublinear Algorithms for Big Data Part 4: Random Topics Qin Zhang 1-1 2-1 Topic 1: Compressive sensing Compressive sensing The model (Candes-Romberg-Tao 04; Donoho 04) Applicaitons Medical imaging reconstruction

More information

A Negative Result Concerning Explicit Matrices With The Restricted Isometry Property

A Negative Result Concerning Explicit Matrices With The Restricted Isometry Property A Negative Result Concerning Explicit Matrices With The Restricted Isometry Property Venkat Chandar March 1, 2008 Abstract In this note, we prove that matrices whose entries are all 0 or 1 cannot achieve

More information

Video-Conferencing System

Video-Conferencing System Video-Conferencing System Evan Broder and C. Christoher Post Introductory Digital Systems Laboratory November 2, 2007 Abstract The goal of this project is to create a video/audio conferencing system. Video

More information

Image Compression through DCT and Huffman Coding Technique

Image Compression through DCT and Huffman Coding Technique International Journal of Current Engineering and Technology E-ISSN 2277 4106, P-ISSN 2347 5161 2015 INPRESSCO, All Rights Reserved Available at http://inpressco.com/category/ijcet Research Article Rahul

More information

CHAPTER 7 CONCLUSION AND FUTURE WORK

CHAPTER 7 CONCLUSION AND FUTURE WORK 158 CHAPTER 7 CONCLUSION AND FUTURE WORK The aim of this thesis was to present robust watermarking techniques for medical image. Section 7.1, consolidates the contributions made by the researcher and Section

More information

Image Authentication Scheme using Digital Signature and Digital Watermarking

Image Authentication Scheme using Digital Signature and Digital Watermarking www..org 59 Image Authentication Scheme using Digital Signature and Digital Watermarking Seyed Mohammad Mousavi Industrial Management Institute, Tehran, Iran Abstract Usual digital signature schemes for

More information

JPEG compression of monochrome 2D-barcode images using DCT coefficient distributions

JPEG compression of monochrome 2D-barcode images using DCT coefficient distributions Edith Cowan University Research Online ECU Publications Pre. JPEG compression of monochrome D-barcode images using DCT coefficient distributions Keng Teong Tan Hong Kong Baptist University Douglas Chai

More information

Introduction to image coding

Introduction to image coding Introduction to image coding Image coding aims at reducing amount of data required for image representation, storage or transmission. This is achieved by removing redundant data from an image, i.e. by

More information

Computational Optical Imaging - Optique Numerique. -- Deconvolution --

Computational Optical Imaging - Optique Numerique. -- Deconvolution -- Computational Optical Imaging - Optique Numerique -- Deconvolution -- Winter 2014 Ivo Ihrke Deconvolution Ivo Ihrke Outline Deconvolution Theory example 1D deconvolution Fourier method Algebraic method

More information

Publicações Matemáticas

Publicações Matemáticas Compressive Sensing Publicações Matemáticas Compressive Sensing Adriana Schulz IMPA Eduardo A. B. da Silva UFRJ Luiz Velho IMPA impa 27 o Colóquio Brasileiro de Matemática Copyright 2009 by Adriana Schulz,

More information

Compressive Sensing for Light Fields Ingomar Wesp Institute of Computer Graphics Johannes Kepler University Linz

Compressive Sensing for Light Fields Ingomar Wesp Institute of Computer Graphics Johannes Kepler University Linz Compressive Sensing for Light Fields Ingomar Wesp Institute of Computer Graphics Johannes Kepler University Linz 16th March 2011 Contents 1 Introduction 2 2 Related Work 3 2.1 Light Transport..................................

More information

Security Based Data Transfer and Privacy Storage through Watermark Detection

Security Based Data Transfer and Privacy Storage through Watermark Detection Security Based Data Transfer and Privacy Storage through Watermark Detection Gowtham.T 1 Pradeep Kumar.G 2 1PG Scholar, Applied Electronics, Nandha Engineering College, Anna University, Erode, India. 2Assistant

More information

Sparsity-promoting recovery from simultaneous data: a compressive sensing approach

Sparsity-promoting recovery from simultaneous data: a compressive sensing approach SEG 2011 San Antonio Sparsity-promoting recovery from simultaneous data: a compressive sensing approach Haneet Wason*, Tim T. Y. Lin, and Felix J. Herrmann September 19, 2011 SLIM University of British

More information

A Secure File Transfer based on Discrete Wavelet Transformation and Audio Watermarking Techniques

A Secure File Transfer based on Discrete Wavelet Transformation and Audio Watermarking Techniques A Secure File Transfer based on Discrete Wavelet Transformation and Audio Watermarking Techniques Vineela Behara,Y Ramesh Department of Computer Science and Engineering Aditya institute of Technology and

More information

Subspace Pursuit for Compressive Sensing: Closing the Gap Between Performance and Complexity

Subspace Pursuit for Compressive Sensing: Closing the Gap Between Performance and Complexity Subspace Pursuit for Compressive Sensing: Closing the Gap Between Performance and Complexity Wei Dai and Olgica Milenkovic Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign

More information

Sachin Dhawan Deptt. of ECE, UIET, Kurukshetra University, Kurukshetra, Haryana, India

Sachin Dhawan Deptt. of ECE, UIET, Kurukshetra University, Kurukshetra, Haryana, India Abstract Image compression is now essential for applications such as transmission and storage in data bases. In this paper we review and discuss about the image compression, need of compression, its principles,

More information

CHAPTER 2 LITERATURE REVIEW

CHAPTER 2 LITERATURE REVIEW 11 CHAPTER 2 LITERATURE REVIEW 2.1 INTRODUCTION Image compression is mainly used to reduce storage space, transmission time and bandwidth requirements. In the subsequent sections of this chapter, general

More information

CS 5614: (Big) Data Management Systems. B. Aditya Prakash Lecture #18: Dimensionality Reduc7on

CS 5614: (Big) Data Management Systems. B. Aditya Prakash Lecture #18: Dimensionality Reduc7on CS 5614: (Big) Data Management Systems B. Aditya Prakash Lecture #18: Dimensionality Reduc7on Dimensionality Reduc=on Assump=on: Data lies on or near a low d- dimensional subspace Axes of this subspace

More information

WATERMARKING FOR IMAGE AUTHENTICATION

WATERMARKING FOR IMAGE AUTHENTICATION WATERMARKING FOR IMAGE AUTHENTICATION Min Wu Bede Liu Department of Electrical Engineering Princeton University, Princeton, NJ 08544, USA Fax: +1-609-258-3745 {minwu, liu}@ee.princeton.edu ABSTRACT A data

More information

Study and Implementation of Video Compression Standards (H.264/AVC and Dirac)

Study and Implementation of Video Compression Standards (H.264/AVC and Dirac) Project Proposal Study and Implementation of Video Compression Standards (H.264/AVC and Dirac) Sumedha Phatak-1000731131- sumedha.phatak@mavs.uta.edu Objective: A study, implementation and comparison of

More information

Redundant Wavelet Transform Based Image Super Resolution

Redundant Wavelet Transform Based Image Super Resolution Redundant Wavelet Transform Based Image Super Resolution Arti Sharma, Prof. Preety D Swami Department of Electronics &Telecommunication Samrat Ashok Technological Institute Vidisha Department of Electronics

More information

Multi-factor Authentication in Banking Sector

Multi-factor Authentication in Banking Sector Multi-factor Authentication in Banking Sector Tushar Bhivgade, Mithilesh Bhusari, Ajay Kuthe, Bhavna Jiddewar,Prof. Pooja Dubey Department of Computer Science & Engineering, Rajiv Gandhi College of Engineering

More information

Disjoint sparsity for signal separation and applications to hybrid imaging inverse problems

Disjoint sparsity for signal separation and applications to hybrid imaging inverse problems Disjoint sparsity for signal separation and applications to hybrid imaging inverse problems Giovanni S Alberti (joint with H Ammari) DMA, École Normale Supérieure, Paris June 16, 2015 Giovanni S Alberti

More information

Numerical Methods For Image Restoration

Numerical Methods For Image Restoration Numerical Methods For Image Restoration CIRAM Alessandro Lanza University of Bologna, Italy Faculty of Engineering CIRAM Outline 1. Image Restoration as an inverse problem 2. Image degradation models:

More information

Survey of the Mathematics of Big Data

Survey of the Mathematics of Big Data Survey of the Mathematics of Big Data Philippe B. Laval KSU September 12, 2014 Philippe B. Laval (KSU) Math & Big Data September 12, 2014 1 / 23 Introduction We survey some mathematical techniques used

More information

Introduction to Medical Image Compression Using Wavelet Transform

Introduction to Medical Image Compression Using Wavelet Transform National Taiwan University Graduate Institute of Communication Engineering Time Frequency Analysis and Wavelet Transform Term Paper Introduction to Medical Image Compression Using Wavelet Transform 李 自

More information

The Image Deblurring Problem

The Image Deblurring Problem page 1 Chapter 1 The Image Deblurring Problem You cannot depend on your eyes when your imagination is out of focus. Mark Twain When we use a camera, we want the recorded image to be a faithful representation

More information

Conceptual Framework Strategies for Image Compression: A Review

Conceptual Framework Strategies for Image Compression: A Review International Journal of Computer Sciences and Engineering Open Access Review Paper Volume-4, Special Issue-1 E-ISSN: 2347-2693 Conceptual Framework Strategies for Image Compression: A Review Sumanta Lal

More information

Text Analytics (Text Mining)

Text Analytics (Text Mining) CSE 6242 / CX 4242 Apr 3, 2014 Text Analytics (Text Mining) LSI (uses SVD), Visualization Duen Horng (Polo) Chau Georgia Tech Some lectures are partly based on materials by Professors Guy Lebanon, Jeffrey

More information

NMR Measurement of T1-T2 Spectra with Partial Measurements using Compressive Sensing

NMR Measurement of T1-T2 Spectra with Partial Measurements using Compressive Sensing NMR Measurement of T1-T2 Spectra with Partial Measurements using Compressive Sensing Alex Cloninger Norbert Wiener Center Department of Mathematics University of Maryland, College Park http://www.norbertwiener.umd.edu

More information

Research Article A Method of Data Recovery Based on Compressive Sensing in Wireless Structural Health Monitoring

Research Article A Method of Data Recovery Based on Compressive Sensing in Wireless Structural Health Monitoring Mathematical Problems in Engineering Volume 214, Article ID 46478, 9 pages http://dx.doi.org/1.11/214/46478 Research Article A Method of Data Recovery Based on Compressive Sensing in Wireless Structural

More information

An Improved Reconstruction methods of Compressive Sensing Data Recovery in Wireless Sensor Networks

An Improved Reconstruction methods of Compressive Sensing Data Recovery in Wireless Sensor Networks Vol.8, No.1 (14), pp.1-8 http://dx.doi.org/1.1457/ijsia.14.8.1.1 An Improved Reconstruction methods of Compressive Sensing Data Recovery in Wireless Sensor Networks Sai Ji 1,, Liping Huang, Jin Wang, Jian

More information

Sampling Theorem Notes. Recall: That a time sampled signal is like taking a snap shot or picture of signal periodically.

Sampling Theorem Notes. Recall: That a time sampled signal is like taking a snap shot or picture of signal periodically. Sampling Theorem We will show that a band limited signal can be reconstructed exactly from its discrete time samples. Recall: That a time sampled signal is like taking a snap shot or picture of signal

More information

Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches

Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches PhD Thesis by Payam Birjandi Director: Prof. Mihai Datcu Problematic

More information

DYNAMIC DOMAIN CLASSIFICATION FOR FRACTAL IMAGE COMPRESSION

DYNAMIC DOMAIN CLASSIFICATION FOR FRACTAL IMAGE COMPRESSION DYNAMIC DOMAIN CLASSIFICATION FOR FRACTAL IMAGE COMPRESSION K. Revathy 1 & M. Jayamohan 2 Department of Computer Science, University of Kerala, Thiruvananthapuram, Kerala, India 1 revathysrp@gmail.com

More information

Comparison of different image compression formats. ECE 533 Project Report Paula Aguilera

Comparison of different image compression formats. ECE 533 Project Report Paula Aguilera Comparison of different image compression formats ECE 533 Project Report Paula Aguilera Introduction: Images are very important documents nowadays; to work with them in some applications they need to be

More information

Lecture 5 Least-squares

Lecture 5 Least-squares EE263 Autumn 2007-08 Stephen Boyd Lecture 5 Least-squares least-squares (approximate) solution of overdetermined equations projection and orthogonality principle least-squares estimation BLUE property

More information

So which is the best?

So which is the best? Manifold Learning Techniques: So which is the best? Todd Wittman Math 8600: Geometric Data Analysis Instructor: Gilad Lerman Spring 2005 Note: This presentation does not contain information on LTSA, which

More information

Compressed Sensing with Non-Gaussian Noise and Partial Support Information

Compressed Sensing with Non-Gaussian Noise and Partial Support Information IEEE SIGNAL PROCESSING LETTERS, VOL. 22, NO. 10, OCTOBER 2015 1703 Compressed Sensing with Non-Gaussian Noise Partial Support Information Ahmad Abou Saleh, Fady Alajaji, Wai-Yip Chan Abstract We study

More information

A VARIANT OF THE EMD METHOD FOR MULTI-SCALE DATA

A VARIANT OF THE EMD METHOD FOR MULTI-SCALE DATA Advances in Adaptive Data Analysis Vol. 1, No. 4 (2009) 483 516 c World Scientific Publishing Company A VARIANT OF THE EMD METHOD FOR MULTI-SCALE DATA THOMAS Y. HOU and MIKE P. YAN Applied and Computational

More information

Non-negative Matrix Factorization (NMF) in Semi-supervised Learning Reducing Dimension and Maintaining Meaning

Non-negative Matrix Factorization (NMF) in Semi-supervised Learning Reducing Dimension and Maintaining Meaning Non-negative Matrix Factorization (NMF) in Semi-supervised Learning Reducing Dimension and Maintaining Meaning SAMSI 10 May 2013 Outline Introduction to NMF Applications Motivations NMF as a middle step

More information

Comparing Multiresolution SVD with Other Methods for Image Compression

Comparing Multiresolution SVD with Other Methods for Image Compression Comparing Multiresolution SVD with Other Methods for Image Compression Ryuichi Ashino Akira Morimoto Michihiro Nagase Rémi Vaillancourt CRM-2987 December 2003 This research was partially supported by the

More information

Low Contrast Image Enhancement Based On Undecimated Wavelet Transform with SSR

Low Contrast Image Enhancement Based On Undecimated Wavelet Transform with SSR International Journal of Computer Sciences and Engineering Open Access Research Paper Volume-4, Issue-02 E-ISSN: 2347-2693 Low Contrast Image Enhancement Based On Undecimated Wavelet Transform with SSR

More information

Admin stuff. 4 Image Pyramids. Spatial Domain. Projects. Fourier domain 2/26/2008. Fourier as a change of basis

Admin stuff. 4 Image Pyramids. Spatial Domain. Projects. Fourier domain 2/26/2008. Fourier as a change of basis Admin stuff 4 Image Pyramids Change of office hours on Wed 4 th April Mon 3 st March 9.3.3pm (right after class) Change of time/date t of last class Currently Mon 5 th May What about Thursday 8 th May?

More information

3. Interpolation. Closing the Gaps of Discretization... Beyond Polynomials

3. Interpolation. Closing the Gaps of Discretization... Beyond Polynomials 3. Interpolation Closing the Gaps of Discretization... Beyond Polynomials Closing the Gaps of Discretization... Beyond Polynomials, December 19, 2012 1 3.3. Polynomial Splines Idea of Polynomial Splines

More information

CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen

CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen LECTURE 3: DATA TRANSFORMATION AND DIMENSIONALITY REDUCTION Chapter 3: Data Preprocessing Data Preprocessing: An Overview Data Quality Major

More information

Stable Signal Recovery from Incomplete and Inaccurate Measurements

Stable Signal Recovery from Incomplete and Inaccurate Measurements Stable Signal Recovery from Incomplete and Inaccurate Measurements EMMANUEL J. CANDÈS California Institute of Technology JUSTIN K. ROMBERG California Institute of Technology AND TERENCE TAO University

More information

Reversible Data Hiding for Security Applications

Reversible Data Hiding for Security Applications Reversible Data Hiding for Security Applications Baig Firdous Sulthana, M.Tech Student (DECS), Gudlavalleru engineering college, Gudlavalleru, Krishna (District), Andhra Pradesh (state), PIN-521356 S.

More information

When is missing data recoverable?

When is missing data recoverable? When is missing data recoverable? Yin Zhang CAAM Technical Report TR06-15 Department of Computational and Applied Mathematics Rice University, Houston, TX 77005 October, 2006 Abstract Suppose a non-random

More information

Combating Anti-forensics of Jpeg Compression

Combating Anti-forensics of Jpeg Compression IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 6, No 3, November 212 ISSN (Online): 1694-814 www.ijcsi.org 454 Combating Anti-forensics of Jpeg Compression Zhenxing Qian 1, Xinpeng

More information

APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY

APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY JIAN-FENG CAI, ZUOWEI SHEN, AND GUI-BO YE Abstract. Recovering missing data from its partial samples is a fundamental problem in mathematics and it has

More information

APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY

APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY APPROXIMATION OF FRAME BASED MISSING DATA RECOVERY JIAN-FENG CAI, ZUOWEI SHEN, AND GUI-BO YE Abstract. Recovering missing data from its partial samples is a fundamental problem in mathematics and it has

More information

Statistical Modeling of Huffman Tables Coding

Statistical Modeling of Huffman Tables Coding Statistical Modeling of Huffman Tables Coding S. Battiato 1, C. Bosco 1, A. Bruna 2, G. Di Blasi 1, G.Gallo 1 1 D.M.I. University of Catania - Viale A. Doria 6, 95125, Catania, Italy {battiato, bosco,

More information

AUTHORIZED WATERMARKING AND ENCRYPTION SYSTEM BASED ON WAVELET TRANSFORM FOR TELERADIOLOGY SECURITY ISSUES

AUTHORIZED WATERMARKING AND ENCRYPTION SYSTEM BASED ON WAVELET TRANSFORM FOR TELERADIOLOGY SECURITY ISSUES AUTHORIZED WATERMARKING AND ENCRYPTION SYSTEM BASED ON WAVELET TRANSFORM FOR TELERADIOLOGY SECURITY ISSUES S.NANDHINI PG SCHOLAR NandhaEngg. College Erode, Tamilnadu, India. Dr.S.KAVITHA M.E.,Ph.d PROFESSOR

More information

ROI Based Medical Image Watermarking with Zero Distortion and Enhanced Security

ROI Based Medical Image Watermarking with Zero Distortion and Enhanced Security I.J. Modern Education and Computer Science, 2014, 10, 40-48 Published Online October 2014 in MECS (http://www.mecs-press.org/) DOI: 10.5815/ijmecs.2014.10.06 ROI Based Medical Image Watermarking with Zero

More information

10/27/14. Streaming & Sampling. Tackling the Challenges of Big Data Big Data Analytics. Piotr Indyk

10/27/14. Streaming & Sampling. Tackling the Challenges of Big Data Big Data Analytics. Piotr Indyk Introduction Streaming & Sampling Two approaches to dealing with massive data using limited resources Sampling: pick a random sample of the data, and perform the computation on the sample 8 2 1 9 1 9 2

More information

A Lightweight Rao-Cauchy Detector for Additive Watermarking in the DWT-Domain

A Lightweight Rao-Cauchy Detector for Additive Watermarking in the DWT-Domain A Lightweight Rao-Cauchy Detector for Additive Watermarking in the DWT-Domain Roland Kwitt, Peter Meerwald, Andreas Uhl Dept. of Computer Sciences, University of Salzburg, Austria E-Mail: {rkwitt, pmeerw,

More information

An Efficient Architecture for Image Compression and Lightweight Encryption using Parameterized DWT

An Efficient Architecture for Image Compression and Lightweight Encryption using Parameterized DWT An Efficient Architecture for Image Compression and Lightweight Encryption using Parameterized DWT Babu M., Mukuntharaj C., Saranya S. Abstract Discrete Wavelet Transform (DWT) based architecture serves

More information

2K Processor AJ-HDP2000

2K Processor AJ-HDP2000 Jan.31, 2007 2K Processor AJ-HDP2000 Technical Overview Version 2.0 January 31, 2007 Professional AV Systems Business Unit Panasonic AVC Networks Company Panasonic Broadcast & Television Systems Company

More information

Probabilistic Latent Semantic Analysis (plsa)

Probabilistic Latent Semantic Analysis (plsa) Probabilistic Latent Semantic Analysis (plsa) SS 2008 Bayesian Networks Multimedia Computing, Universität Augsburg Rainer.Lienhart@informatik.uni-augsburg.de www.multimedia-computing.{de,org} References

More information

HSI BASED COLOUR IMAGE EQUALIZATION USING ITERATIVE n th ROOT AND n th POWER

HSI BASED COLOUR IMAGE EQUALIZATION USING ITERATIVE n th ROOT AND n th POWER HSI BASED COLOUR IMAGE EQUALIZATION USING ITERATIVE n th ROOT AND n th POWER Gholamreza Anbarjafari icv Group, IMS Lab, Institute of Technology, University of Tartu, Tartu 50411, Estonia sjafari@ut.ee

More information

The p-norm generalization of the LMS algorithm for adaptive filtering

The p-norm generalization of the LMS algorithm for adaptive filtering The p-norm generalization of the LMS algorithm for adaptive filtering Jyrki Kivinen University of Helsinki Manfred Warmuth University of California, Santa Cruz Babak Hassibi California Institute of Technology

More information

EECS 556 Image Processing W 09. Interpolation. Interpolation techniques B splines

EECS 556 Image Processing W 09. Interpolation. Interpolation techniques B splines EECS 556 Image Processing W 09 Interpolation Interpolation techniques B splines What is image processing? Image processing is the application of 2D signal processing methods to images Image representation

More information

STUDY OF MUTUAL INFORMATION IN PERCEPTUAL CODING WITH APPLICATION FOR LOW BIT-RATE COMPRESSION

STUDY OF MUTUAL INFORMATION IN PERCEPTUAL CODING WITH APPLICATION FOR LOW BIT-RATE COMPRESSION STUDY OF MUTUAL INFORMATION IN PERCEPTUAL CODING WITH APPLICATION FOR LOW BIT-RATE COMPRESSION Adiel Ben-Shalom, Michael Werman School of Computer Science Hebrew University Jerusalem, Israel. {chopin,werman}@cs.huji.ac.il

More information

Wavelet analysis. Wavelet requirements. Example signals. Stationary signal 2 Hz + 10 Hz + 20Hz. Zero mean, oscillatory (wave) Fast decay (let)

Wavelet analysis. Wavelet requirements. Example signals. Stationary signal 2 Hz + 10 Hz + 20Hz. Zero mean, oscillatory (wave) Fast decay (let) Wavelet analysis In the case of Fourier series, the orthonormal basis is generated by integral dilation of a single function e jx Every 2π-periodic square-integrable function is generated by a superposition

More information

Inference from sub-nyquist Samples

Inference from sub-nyquist Samples Inference from sub-nyquist Samples Alireza Razavi, Mikko Valkama Department of Electronics and Communications Engineering/TUT Characteristics of Big Data (4 V s) Volume: Traditional computing methods are

More information

Volume 2, Issue 12, December 2014 International Journal of Advance Research in Computer Science and Management Studies

Volume 2, Issue 12, December 2014 International Journal of Advance Research in Computer Science and Management Studies Volume 2, Issue 12, December 2014 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online at: www.ijarcsms.com

More information

Lecture Topic: Low-Rank Approximations

Lecture Topic: Low-Rank Approximations Lecture Topic: Low-Rank Approximations Low-Rank Approximations We have seen principal component analysis. The extraction of the first principle eigenvalue could be seen as an approximation of the original

More information

Learning, Sparsity and Big Data

Learning, Sparsity and Big Data Learning, Sparsity and Big Data M. Magdon-Ismail (Joint Work) January 22, 2014. Out-of-Sample is What Counts NO YES A pattern exists We don t know it We have data to learn it Tested on new cases? Teaching

More information

Dynamic Binary Location based Multi-watermark Embedding Algorithm in DWT

Dynamic Binary Location based Multi-watermark Embedding Algorithm in DWT Dynamic Binary Location based Multi-watermark Embedding Algorithm in DWT Ammar Jameel Hussein, Seda Yuksel, and Ersin Elbasi Abstract In order to achieve a good imperceptibility and robustness, using 4-level

More information

A Robust and Lossless Information Embedding in Image Based on DCT and Scrambling Algorithms

A Robust and Lossless Information Embedding in Image Based on DCT and Scrambling Algorithms A Robust and Lossless Information Embedding in Image Based on DCT and Scrambling Algorithms Dr. Mohammad V. Malakooti Faculty and Head of Department of Computer Engineering, Islamic Azad University, UAE

More information

System Identification for Acoustic Comms.:

System Identification for Acoustic Comms.: System Identification for Acoustic Comms.: New Insights and Approaches for Tracking Sparse and Rapidly Fluctuating Channels Weichang Li and James Preisig Woods Hole Oceanographic Institution The demodulation

More information

A multi-scale approach to InSAR time series analysis

A multi-scale approach to InSAR time series analysis A multi-scale approach to InSAR time series analysis M. Simons, E. Hetland, P. Muse, Y. N. Lin & C. DiCaprio U Interferogram stack time A geophysical perspective on deformation tomography Examples: Long

More information

Stable Signal Recovery from Incomplete and Inaccurate Measurements

Stable Signal Recovery from Incomplete and Inaccurate Measurements Stable Signal Recovery from Incomplete and Inaccurate Measurements Emmanuel Candes, Justin Romberg, and Terence Tao Applied and Computational Mathematics, Caltech, Pasadena, CA 91125 Department of Mathematics,

More information

CHAPTER VII CONCLUSIONS

CHAPTER VII CONCLUSIONS CHAPTER VII CONCLUSIONS To do successful research, you don t need to know everything, you just need to know of one thing that isn t known. -Arthur Schawlow In this chapter, we provide the summery of the

More information

FFT Algorithms. Chapter 6. Contents 6.1

FFT Algorithms. Chapter 6. Contents 6.1 Chapter 6 FFT Algorithms Contents Efficient computation of the DFT............................................ 6.2 Applications of FFT................................................... 6.6 Computing DFT

More information

EASTERN ARIZONA COLLEGE Differential Equations

EASTERN ARIZONA COLLEGE Differential Equations EASTERN ARIZONA COLLEGE Differential Equations Course Design 2015-2016 Course Information Division Mathematics Course Number MAT 260 (SUN# MAT 2262) Title Differential Equations Credits 3 Developed by

More information

BUILDING A PREDICTIVE MODEL AN EXAMPLE OF A PRODUCT RECOMMENDATION ENGINE

BUILDING A PREDICTIVE MODEL AN EXAMPLE OF A PRODUCT RECOMMENDATION ENGINE BUILDING A PREDICTIVE MODEL AN EXAMPLE OF A PRODUCT RECOMMENDATION ENGINE Alex Lin Senior Architect Intelligent Mining alin@intelligentmining.com Outline Predictive modeling methodology k-nearest Neighbor

More information

Morphological Diversity and Sparsity for Multichannel Data Restoration

Morphological Diversity and Sparsity for Multichannel Data Restoration Morphological Diversity and Sparsity for Multichannel Data Restoration J.Bobin 1, Y.Moudden 1, J.Fadili and J-L.Starck 1 1 jerome.bobin@cea.fr, ymoudden@cea.fr, jstarck@cea.fr - CEA-DAPNIA/SEDI, Service

More information

Region of Interest Access with Three-Dimensional SBHP Algorithm CIPR Technical Report TR-2006-1

Region of Interest Access with Three-Dimensional SBHP Algorithm CIPR Technical Report TR-2006-1 Region of Interest Access with Three-Dimensional SBHP Algorithm CIPR Technical Report TR-2006-1 Ying Liu and William A. Pearlman January 2006 Center for Image Processing Research Rensselaer Polytechnic

More information

Singular Spectrum Analysis with Rssa

Singular Spectrum Analysis with Rssa Singular Spectrum Analysis with Rssa Maurizio Sanarico Chief Data Scientist SDG Consulting Milano R June 4, 2014 Motivation Discover structural components in complex time series Working hypothesis: a signal

More information

Collaborative Filtering. Radek Pelánek

Collaborative Filtering. Radek Pelánek Collaborative Filtering Radek Pelánek 2015 Collaborative Filtering assumption: users with similar taste in past will have similar taste in future requires only matrix of ratings applicable in many domains

More information

Robot Perception Continued

Robot Perception Continued Robot Perception Continued 1 Visual Perception Visual Odometry Reconstruction Recognition CS 685 11 Range Sensing strategies Active range sensors Ultrasound Laser range sensor Slides adopted from Siegwart

More information

MEDICAL IMAGE COMPRESSION USING HYBRID CODER WITH FUZZY EDGE DETECTION

MEDICAL IMAGE COMPRESSION USING HYBRID CODER WITH FUZZY EDGE DETECTION MEDICAL IMAGE COMPRESSION USING HYBRID CODER WITH FUZZY EDGE DETECTION K. Vidhya 1 and S. Shenbagadevi Department of Electrical & Communication Engineering, College of Engineering, Anna University, Chennai,

More information

ANALYSIS OF THE EFFECTIVENESS IN IMAGE COMPRESSION FOR CLOUD STORAGE FOR VARIOUS IMAGE FORMATS

ANALYSIS OF THE EFFECTIVENESS IN IMAGE COMPRESSION FOR CLOUD STORAGE FOR VARIOUS IMAGE FORMATS ANALYSIS OF THE EFFECTIVENESS IN IMAGE COMPRESSION FOR CLOUD STORAGE FOR VARIOUS IMAGE FORMATS Dasaradha Ramaiah K. 1 and T. Venugopal 2 1 IT Department, BVRIT, Hyderabad, India 2 CSE Department, JNTUH,

More information

Watermarking Techniques for Protecting Intellectual Properties in a Digital Environment

Watermarking Techniques for Protecting Intellectual Properties in a Digital Environment Watermarking Techniques for Protecting Intellectual Properties in a Digital Environment Isinkaye F. O*. and Aroge T. K. Department of Computer Science and Information Technology University of Science and

More information

Mesh Discretization Error and Criteria for Accuracy of Finite Element Solutions

Mesh Discretization Error and Criteria for Accuracy of Finite Element Solutions Mesh Discretization Error and Criteria for Accuracy of Finite Element Solutions Chandresh Shah Cummins, Inc. Abstract Any finite element analysis performed by an engineer is subject to several types of

More information

A Novel Method to Improve Resolution of Satellite Images Using DWT and Interpolation

A Novel Method to Improve Resolution of Satellite Images Using DWT and Interpolation A Novel Method to Improve Resolution of Satellite Images Using DWT and Interpolation S.VENKATA RAMANA ¹, S. NARAYANA REDDY ² M.Tech student, Department of ECE, SVU college of Engineering, Tirupati, 517502,

More information

MIMO CHANNEL CAPACITY

MIMO CHANNEL CAPACITY MIMO CHANNEL CAPACITY Ochi Laboratory Nguyen Dang Khoa (D1) 1 Contents Introduction Review of information theory Fixed MIMO channel Fading MIMO channel Summary and Conclusions 2 1. Introduction The use

More information

Computational Foundations of Cognitive Science

Computational Foundations of Cognitive Science Computational Foundations of Cognitive Science Lecture 15: Convolutions and Kernels Frank Keller School of Informatics University of Edinburgh keller@inf.ed.ac.uk February 23, 2010 Frank Keller Computational

More information

Several Views of Support Vector Machines

Several Views of Support Vector Machines Several Views of Support Vector Machines Ryan M. Rifkin Honda Research Institute USA, Inc. Human Intention Understanding Group 2007 Tikhonov Regularization We are considering algorithms of the form min

More information

Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections

Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections Maximilian Hung, Bohyun B. Kim, Xiling Zhang August 17, 2013 Abstract While current systems already provide

More information

Big Data Interpolation: An Effcient Sampling Alternative for Sensor Data Aggregation

Big Data Interpolation: An Effcient Sampling Alternative for Sensor Data Aggregation Big Data Interpolation: An Effcient Sampling Alternative for Sensor Data Aggregation Hadassa Daltrophe, Shlomi Dolev, Zvi Lotker Ben-Gurion University Outline Introduction Motivation Problem definition

More information

Conventional approaches to sampling signals or images follow Shannon s celebrated. An Introduction To Compressive Sampling

Conventional approaches to sampling signals or images follow Shannon s celebrated. An Introduction To Compressive Sampling DIGITAL VISION An Introduction To Compressive Sampling [A sensing/sampling paradigm that goes against the common knowledge in data acquisition] [ Emmanuel J. Candès and Michael B. Wakin ] Conventional

More information

Assessment. Presenter: Yupu Zhang, Guoliang Jin, Tuo Wang Computer Vision 2008 Fall

Assessment. Presenter: Yupu Zhang, Guoliang Jin, Tuo Wang Computer Vision 2008 Fall Automatic Photo Quality Assessment Presenter: Yupu Zhang, Guoliang Jin, Tuo Wang Computer Vision 2008 Fall Estimating i the photorealism of images: Distinguishing i i paintings from photographs h Florin

More information

Mathematical finance and linear programming (optimization)

Mathematical finance and linear programming (optimization) Mathematical finance and linear programming (optimization) Geir Dahl September 15, 2009 1 Introduction The purpose of this short note is to explain how linear programming (LP) (=linear optimization) may

More information