从稀疏到结构化稀疏:贝叶斯方法

被引:26
作者
孙洪 [1 ]
张智林 [2 ]
余磊 [1 ,3 ]
机构
[1] 武汉大学电子信息学院
[2] DeptElectrical and Computer Engineering,University of California,CA-,USA
[3] VisAGeS U Research Unit,INRIA, Rennes,France
关键词
压缩感知; 稀疏理论; 结构化稀疏分解算法; 贝叶斯压缩感知;
D O I
暂无
中图分类号
TN911.7 [信号处理];
学科分类号
0711 ; 080401 ; 080402 ;
摘要
稀疏分解算法是稀疏表达理论和压缩感知理论中的核心问题,也是当前信号处理领域的一个热门话题。近年来,研究人员发现除了稀疏以外,如果引入稀疏系数之间的相关性先验信息,可以大大提高稀疏分解算法的精度,这种方法称为"结构化稀疏分解算法"。本文归纳和总结了从稀疏到结构化稀疏的信号模型,并且介绍了两种不同的贝叶斯稀疏(或者结构化稀疏)算法,以及从稀疏到结构化稀疏贝叶斯稀疏分解算法的扩展。同时,本文还介绍了结构化稀疏分解算法在医学信号处理和语音信号处理中的应用。
引用
收藏
页码:759 / 773
页数:15
相关论文
共 9 条
[1]  
Implementation of Compressed Sensing in Telecardiology Sensor Networks[J] . International Journal of Telemedicine and Applications . 2010
[2]   A short note on compressed sensing with partially known signal support [J].
Jacques, Laurent .
SIGNAL PROCESSING, 2010, 90 (12) :3308-3312
[3]  
Enhancing Sparsity by Reweighted ? 1 Minimization[J] . Emmanuel J. Candès,Michael B. Wakin,Stephen P. Boyd.Journal of Fourier Analysis and Applications . 2008 (5)
[4]  
The restricted isometry property and its implications for compressed sensing[J] . Emmanuel J. Candès.Comptes rendus - Mathématique . 2008 (9)
[5]  
Wireless sensor networks for personal health monitoring: Issues and an implementation[J] . Computer Communications . 2006 (13)
[6]   Simultaneous cartoon and texture image inpainting using morphological component analysis (MCA) [J].
Elad, M ;
Starck, JL ;
Querre, P ;
Donoho, DL .
APPLIED AND COMPUTATIONAL HARMONIC ANALYSIS, 2005, 19 (03) :340-358
[7]  
Recovery of Clustered Sparse Signals from Compressive Measurements .2 V. Cevher,P. Indyk,C. Hegde,et al. Sampling Theory and Applications . 2009
[8]  
Sparse signal recovery using Markovrandom fields .2 V. Cevher,M. Duarte,C. Hegde,R. Baraniuk. Proc. Neural Information Processing Systems (NIPS) . 2008
[9]  
Sparse Representation for SignalClassification .2 Huang K,Aviyent S. NIPS 2007 .