Bayesian compressive sensing using wavelet based Markov random fields

作者：

Highlights：

•

• We proposed wavelet-based MRF model in BCS, named WMRF.

• WMRF uses both the tree structure and neighborhood relation of wavelet coefficients.

• VBEM inference procedure is used to derive posterior distributions.

• SESOP algorithm is employed to estimate model parameters.

摘要

•We proposed wavelet-based MRF model in BCS, named WMRF.•WMRF uses both the tree structure and neighborhood relation of wavelet coefficients.•VBEM inference procedure is used to derive posterior distributions.•SESOP algorithm is employed to estimate model parameters.

论文关键词：Bayesian compressive sensing,Markov random fields,Wavelet-tree structure,Variational Bayesian expectation maximization,Sequential subspace optimization (SESOP)

论文评审过程：Received 14 February 2017, Revised 1 June 2017, Accepted 19 June 2017, Available online 6 July 2017, Version of Record 19 July 2017.

论文官网地址：https://doi.org/10.1016/j.image.2017.06.004