Bayesian method for image recovery from block compressive sensing

We consider the problem of recovering an image using block compressed sensing (BCS). Traditional BCS algorithms recovers each image block independently and utilizes post-processing methods for removing the blocking artifacts. In contrast, we propose an image recovery method free of post-processing, where we utilize a lapped transform (LT) for the sparse representation of the image in order to reduce the blocking artifacts. Specifically, we derive an iterative image reconstruction method, where a small number of adjacent measurement blocks are jointly processed for recovering an image block. For this purpose, we propose a novel sparse Bayesian learning (SBL) algorithm.

[1]  Michael E. Tipping.  Bayesian Inference: An Introduction to Principles and Practice in Machine Learning , 2003, Advanced Lectures on Machine Learning.

[2]  Lawrence Carin,et al.  Bayesian Compressive Sensing , 2008, IEEE Transactions on Signal Processing.

[3]  Marian Codreanu,et al.  Streaming signal recovery using sparse Bayesian learning , 2014, 2014 48th Asilomar Conference on Signals, Systems and Computers.

[4]  Lu Gan.  Block Compressed Sensing of Natural Images , 2007, 2007 15th International Conference on Digital Signal Processing.

[5]  Henrique S. Malvar.  Biorthogonal and nonuniform lapped transforms for transform coding with reduced blocking and ringing artifacts , 1998, IEEE Trans. Signal Process..

[6]  田中 聡久,et al.  Lapped transforms and their applications in image processing , 2002 .

[7]  Michael E. Tipping,et al.  Fast Marginal Likelihood Maximisation for Sparse Bayesian Models , 2003 .

[8]  James E. Fowler,et al.  Block Compressed Sensing of Images Using Directional Transforms , 2010, 2010 Data Compression Conference.

[9]  Marian Codreanu,et al.  A Bayesian Approach for Online Recovery of Streaming Signals From Compressive Measurements , 2017, IEEE Transactions on Signal Processing.

[10]  Michael E. Tipping,et al.  Analysis of Sparse Bayesian Learning , 2001, NIPS.

[11]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[12]  David L Donoho,et al.  Compressed sensing , 2006, IEEE Transactions on Information Theory.

[13]  Justin K. Romberg,et al.  Sparse Recovery of Streaming Signals Using $\ell_1$-Homotopy , 2013, IEEE Transactions on Signal Processing.

[14]  Feng Wu,et al.  Directional Lapped Transforms for Image Coding , 2008, DCC.

[15]  James E. Fowler,et al.  Block-Based Compressed Sensing of Images and Video , 2012, Found. Trends Signal Process..

[16]  Justin Romberg,et al.  Practical Signal Recovery from Random Projections , 2005 .

[17]  Henrique S. Malvar,et al.  Signal processing with lapped transforms , 1992 .

[18]  Ricardo L. de Queiroz,et al.  LAPPED TRANSFORMS FOR IMAGE COMPRESSION , 1999 .

[19]  H. T. Kung,et al.  Partitioned compressive sensing with neighbor-weighted decoding , 2011, 2011 - MILCOM 2011 Military Communications Conference.

[20]  Stéphane Mallat,et al.  A Wavelet Tour of Signal Processing, 2nd Edition , 1999 .

[21]  George Eastman House,et al.  Sparse Bayesian Learning and the Relevance Vector Machine , 2001 .

[22]  Shogo Muramatsu,et al.  Directional Lapped Orthogonal Transform: Theory and Design , 2012, IEEE Transactions on Image Processing.

[23]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.