Compressive Sampling: A Brief Overview

Slides:



Advertisements
Similar presentations
An Introduction to Compressed Sensing Student : Shenghan TSAI Advisor : Hsuan-Jung Su and Pin-Hsun Lin Date : May 02,
Advertisements

Compressive Sensing IT530, Lecture Notes.
Joint work with Irad Yavneh
Pixel Recovery via Minimization in the Wavelet Domain Ivan W. Selesnick, Richard Van Slyke, and Onur G. Guleryuz *: Polytechnic University, Brooklyn, NY.
Structured Sparse Principal Component Analysis Reading Group Presenter: Peng Zhang Cognitive Radio Institute Friday, October 01, 2010 Authors: Rodolphe.
Learning Measurement Matrices for Redundant Dictionaries Richard Baraniuk Rice University Chinmay Hegde MIT Aswin Sankaranarayanan CMU.
Online Performance Guarantees for Sparse Recovery Raja Giryes ICASSP 2011 Volkan Cevher.
+ Compressed Sensing. + Mobashir Mohammad Aditya Kulkarni Tobias Bertelsen Malay Singh Hirak Sarkar Nirandika Wanigasekara Yamilet Serrano Llerena Parvathy.
Submodular Dictionary Selection for Sparse Representation Volkan Cevher Laboratory for Information and Inference Systems - LIONS.
Multi-Task Compressive Sensing with Dirichlet Process Priors Yuting Qi 1, Dehong Liu 1, David Dunson 2, and Lawrence Carin 1 1 Department of Electrical.
An Introduction and Survey of Applications
Beyond Nyquist: Compressed Sensing of Analog Signals
Contents 1. Introduction 2. UWB Signal processing 3. Compressed Sensing Theory 3.1 Sparse representation of signals 3.2 AIC (analog to information converter)
Exact or stable image\signal reconstruction from incomplete information Project guide: Dr. Pradeep Sen UNM (Abq) Submitted by: Nitesh Agarwal IIT Roorkee.
Learning sparse representations to restore, classify, and sense images and videos Guillermo Sapiro University of Minnesota Supported by NSF, NGA, NIH,
An Introduction to Sparse Coding, Sparse Sensing, and Optimization Speaker: Wei-Lun Chao Date: Nov. 23, 2011 DISP Lab, Graduate Institute of Communication.
Compressed sensing Carlos Becker, Guillaume Lemaître & Peter Rennert
ECE Department Rice University dsp.rice.edu/cs Measurements and Bits: Compressed Sensing meets Information Theory Shriram Sarvotham Dror Baron Richard.
Bayesian Robust Principal Component Analysis Presenter: Raghu Ranganathan ECE / CMR Tennessee Technological University January 21, 2011 Reading Group (Xinghao.
“Random Projections on Smooth Manifolds” -A short summary
Volkan Cevher, Marco F. Duarte, and Richard G. Baraniuk European Signal Processing Conference 2008.
Sparse and Overcomplete Data Representation
Richard Baraniuk Rice University dsp.rice.edu/cs Compressive Signal Processing.
Compressed Sensing for Networked Information Processing Reza Malek-Madani, 311/ Computational Analysis Don Wagner, 311/ Resource Optimization Tristan Nguyen,
Image Denoising via Learned Dictionaries and Sparse Representations
Compressive Signal Processing
Random Convolution in Compressive Sampling Michael Fleyer.
Introduction to Compressive Sensing
Rice University dsp.rice.edu/cs Distributed Compressive Sensing A Framework for Integrated Sensing and Processing for Signal Ensembles Marco Duarte Shriram.
Compressive Sensing Lecture notes by Richard G. Baraniuk
Compressive sensing: Theory, Algorithms and Applications
Compressed Sensing Compressive Sampling
An ALPS’ view of Sparse Recovery Volkan Cevher Laboratory for Information and Inference Systems - LIONS
Compressive Sensing IT530, Lecture Notes.
Game Theory Meets Compressed Sensing
Richard Baraniuk Chinmay Hegde Marco Duarte Mark Davenport Rice University Michael Wakin University of Michigan Compressive Learning and Inference.
Compressed Sensing Based UWB System Peng Zhang Wireless Networking System Lab WiNSys.
Compressive Sensing Based on Local Regional Data in Wireless Sensor Networks Hao Yang, Liusheng Huang, Hongli Xu, Wei Yang 2012 IEEE Wireless Communications.
Cs: compressed sensing
Introduction to Compressive Sensing
Introduction to Compressed Sensing and its applications
Learning With Structured Sparsity
Source Localization on a budget Volkan Cevher Rice University Petros RichAnna Martin Lance.
The Secrecy of Compressed Sensing Measurements Yaron Rachlin & Dror Baron TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.:
Model-Based Compressive Sensing Presenter: Jason David Bonior ECE / CMR Tennessee Technological University November 5, 2010 Reading Group (Richard G. Baraniuk,
Compressive Sensing for Multimedia Communications in Wireless Sensor Networks By: Wael BarakatRabih Saliba EE381K-14 MDDSP Literary Survey Presentation.
Shriram Sarvotham Dror Baron Richard Baraniuk ECE Department Rice University dsp.rice.edu/cs Sudocodes Fast measurement and reconstruction of sparse signals.
EE381K-14 MDDSP Literary Survey Presentation March 4th, 2008
ALISSA M. STAFFORD MENTOR: ALEX CLONINGER DIRECTED READING PROJECT MAY 3, 2013 Compressive Sensing & Applications.
An Introduction to Compressive Sensing Speaker: Ying-Jou Chen Advisor: Jian-Jiun Ding.
Learning to Sense Sparse Signals: Simultaneous Sensing Matrix and Sparsifying Dictionary Optimization Julio Martin Duarte-Carvajalino, and Guillermo Sapiro.
The Scaling Law of SNR-Monitoring in Dynamic Wireless Networks Soung Chang Liew Hongyi YaoXiaohang Li.
SuperResolution (SR): “Classical” SR (model-based) Linear interpolation (with post-processing) Edge-directed interpolation (simple idea) Example-based.
An Introduction to Compressive Sensing
Compressive Sensing Techniques for Video Acquisition EE5359 Multimedia Processing December 8,2009 Madhu P. Krishnan.
Jianchao Yang, John Wright, Thomas Huang, Yi Ma CVPR 2008 Image Super-Resolution as Sparse Representation of Raw Image Patches.
From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images Alfred M. Bruckstein (Technion), David L. Donoho (Stanford), Michael.
Compressive Coded Aperture Video Reconstruction
Computing and Compressive Sensing in Wireless Sensor Networks
Peng Zhang Cognitive Radio Institute
Compressive Sensing Imaging
Towards Understanding the Invertibility of Convolutional Neural Networks Anna C. Gilbert1, Yi Zhang1, Kibok Lee1, Yuting Zhang1, Honglak Lee1,2 1University.
Bounds for Optimal Compressed Sensing Matrices
Sudocodes Fast measurement and reconstruction of sparse signals
Introduction to Compressive Sensing Aswin Sankaranarayanan
Aishwarya sreenivasan 15 December 2006.
INFONET Seminar Application Group
CIS 700: “algorithms for Big Data”
Sudocodes Fast measurement and reconstruction of sparse signals
Outline Sparse Reconstruction RIP Condition
Presentation transcript:

Compressive Sampling: A Brief Overview Ravi Garg With slides contributed by W.H.Chuang and Dr. Avinash L. Varna

Sampling Theorem Sampling: record a signal in the form of samples Nyquist Sampling Theorem: Signal can be perfectly reconstructed from samples (i.e., free from aliasing) if sampling rate ≥ 2 × signal bandwidth B Samples are “measurements” of the signal  serve as constraints that guide the reconstruction of remaining signal

Sample-then-Compress Paradigm Signal of interest is often compressible / sparse in a proper basis only small portion has large / non-zero values If non-zero values spread wide, sampling rate has to be high, per Sampling Theorem In Fourier basis Conventional data acquisition – sample at or above Nyquist rate compress to meet desired data rate May lose information

Sample-then-Compress Paradigm often costly and wasteful! Why even capture unnecessary data? Romberg, “Compressed Sensing: A Tutorial”, IEEE Statistical Signal Processing Workshop, August 2007

Signal Sampling by Linear Measurement Linear measurements: inner product between signal and sampling basis functions E.g..: Pixels Sinusoids Romberg, “Compressed Sensing: A Tutorial”, IEEE Statistical Signal Processing Workshop, August 2007

Signal Sampling by Linear Measurement Assume: f is sparse under proper basis (sparsity basis) Overall linear measurements: linear combinations of columns in Φ corresponding to non-zero entries in f Φ is known as measurement basis Signal recovery requires special properties of Φ

What Makes a Good Sampling Basis – Incoherence Signal is local, measurements are global Each measurement picks up a little info. about each component “Triangulate” signal components from measurements Sparse signal Incoherent measurements Romberg, “Compressed Sensing: A Tutorial”, IEEE Statistical Signal Processing Workshop, August 2007

Signal Reconstruction by L-0 / L-1 Minimization Given the sparsity of signal and the incoherence between signal and sampling basis… Perfect signal reconstruction by L-0 minimization: Believed to be NP hard: requires exhaustive enumeration of possible locations of the nonzero entries Alternative: Signal reconstruction by L-1 minimization: Surprisingly, this can lead to perfect reconstruction under certain conditions!

Example Length 256 signal with 16 non-zero Fourier coefficients Sparse signal in Fourier domain Dense in time domain Length 256 signal with 16 non-zero Fourier coefficients Given only 80 samples From: http://www.l1-magic.com

Reconstruction Perfect signal reconstruction Recovered signal in Fourier domain Recovered signal in time domain Perfect signal reconstruction

Original Phantom Image L-1 norm minimization of gradient Image Reconstruction Original Phantom Image Fourier Sampling Mask Min Energy Solution L-1 norm minimization of gradient From Notes with the l-1magic source package

General Problem Statement Suppose we are given M linear measurements of x Is it possible to recover x ? How large should M be? Image from: Richard Baraniuk, Compressive Sensing

Restricted Isometry Property If the K locations of non-zero entries are known, then M ≥ K is sufficient, if the following property holds: Restricted Isometry Property (RIP): for any vector v sharing the same K locations and some s sufficiently small δK Θ= Φ Ψ “preserves” the lengths of these sparse vectors RIP ensures that measurements and sparse vectors have good correspondence

Restricted Isometry Property In general, locations of non-zero entries are unknown A sufficient condition for signal recovery: for arbitrary 3K–sparse vectors RIP also ensures “stable” signal recovery: good recovery accuracy in presence of Non-zero small entries Measurement errors

Random Measurement Matrices In general, sparsifying basis Ψ may not be known Φ is non-adaptive, i.e., deterministic Construction of deterministic sampling matrix is difficult Suppose Φ is an M x N matrix with i.i.d. Gaussian entries with M > C K log(N/K) << N Φ I = Φ satisfies RIP with high probability Φ is incoherent with the delta basis Further, Θ = Φ Ψ is also i.i.d. Gaussian for any orthonormal Ψ  Φ is incoherent with every Ψ with high probability Random matrices with i.i.d. ±1 entries also have RIP

Signal Reconstruction: L-2 vs L-0 vs L-1 Minimum L-2 norm solution Closed form solution exists; Almost always never finds sparsest solution Solution usually has lot of ringing Minimum L-0 norm solution Requires exhaustive enumeration of possible locations of the nonzero entries NP hard Minimum L-1 norm solution Can be reformulated as a linear program “L-1 trick”

Signal Reconstruction Methods Convex optimization with efficient algorithms Basis pursuit by linear programming LASSO Danzig selector etc Non-global optimization solutions are also available e.g.: Orthogonal Matching Pursuit

Summary Given an N-dimensional vector x which is S-sparse in some basis We obtain K random measurements of x of the form with φi a vector with i.i.d Gaussian / ±1 entries If we have sufficient measurements (<< N), then x can be almost always perfectly reconstructed by solving

Single Pixel Camera Capture Random Projections by setting the Digital Micromirror Device (DMD) Implements a ±1 random matrix generated using a seed Some sort of inherent “security” provided by seed Image reconstruction after obtaining sufficient number of measurements Michael Wakin, Jason Laska, Marco Duarte, Dror Baron, Shriram Sarvotham, Dharmpal Takhar, Kevin Kelly, and Richard Baraniuk, “An architecture for compressive imaging”. ICIP 2006

Advantages of CS camera Single Low cost photodetector Can be used in wavelength ranges where difficult / expensive to build CCD / CMOS arrays Scalable progressive reconstruction Image quality can be progressively refined with more measurements Suited to distributed sensing applications (such as sensor networks) where resources are severely restricted at sensor Has been extended to the case of video

Images from http://www.dsp.rice.edu/cs/cscamera Experimental Setup Images from http://www.dsp.rice.edu/cs/cscamera

Experimental Results 1600 meas. (10%) 3300 meas. (20%)

Experimental Results Original Object (4096 pixels) 4096 Pixels 800 Measurements (20%) 4096 Pixels 1600 Measurements (40%) Original Object 4096 Pixels 800 Measurements (20%) 4096 Pixels 1600 Measurements (40%)

Error Correction Let x denote a message to be transmitted (N – vector) Choose A as an M x N (M > N) matrix with i.i.d. Gaussian entries Transmit c = A x over the channel Suppose y is the received vector which has errors in some unknown locations y = c + e where e is an unknown sparse vector Let F be such that FA = 0  y’ = F y = FA x + Fe = Fe e can be recovered by

Image Recovery Main signal recovery problems can be approached by harnessing inherent signal sparsity Assumption: image x can be sparsely represented by a “over-complete dictionary” D Fourier Wavelet Data-generated basis? Signal recovery can be cast as

Image Denoising using Learned Dictionary Two different types of dictionaries Recovery results (origin – noisy – recovered) Over-complete DCT dictionary Trained Patch Dictionary

Compressive Sampling… Has significant implications on data acquisition process Allows us to exploit the underlying structure of the signal Mainly sparsity in some basis High potential for cases where resources are scarce Medical imaging Distributed sensing in sensor networks Ultra wideband communications …. Also has applications in Error-free communication Image processing …

References Websites: Tutorials: Research Papers http://www.dsp.rice.edu/cs/ http://www.l1-magic.org/ Tutorials: Candes, “Compressive Sampling” , Proc. Intl. Congress of Mathematics, 2006 Baraniuk, “Compressive Sensing”, IEEE Signal Processing Magazine, July 2007 Candès and Wakin, “An Introduction to Compressive Sampling”. IEEE Signal Processing Magazine, March 2008. Romberg, “Compressed Sensing: A Tutorial”, IEEE Statistical Signal Processing Workshop, August 2007 Research Papers Candès, Romberg and Tao, “Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information”, IEEE Trans. Inform. Theory, vol. 52 (2006), 489–509 Wakin, et al., “An architecture for compressive imaging”. ICIP 2006 Candès and Tao, “Decoding by linear programming”, IEEE Trans. on Information Theory, 51(12), pp. 4203 - 4215, Dec. 2005 Elad and Aharon, "Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries," IEEE Trans. On Image Processing, Dec. 2006