Sparse Bayesian Learning: A Diversified Scheme

This set of Matlab (Vision R2021b) functions contain the core code to reproduce the results of the above paper.

• DivSBL.m ---> the main function for Diversified Block Sparse Bayesian Learning.

For convenience, 'DivSBL.m' is directly provided in each demo folders.

# Please note that the value of the `PRUNE_GAMMA` parameter in 'DivSBL.m' function needs to be manually adjusted according to the data.

# We have already set appropriate sizes for the data style in the demo. If you need to change the data, you may need to adjust its values accordingly.

Set path

The following 5 sub-files in DivSBL folder need to be manually read by the user into the MATLAB path, for use in comparative experiments in the demo folders.

* BCS_fast_rvm.m        ---> The function file for SBL(RVM) algorithm from [1],

* CVX                   ---> Matlab software for disciplined convex programming from [2],

* PCSBL.m               ---> The function file for PC-SBL algorithm from [3],

* spgl1-2.1             ---> A solver for large-scale sparse reconstruction from [4],

* StructOMP Released 2  ---> The function file for StructOMP algorithm from [5].

Folder 1: demo_Synthetic signal

* DEMO_Synthetic_Signal.m   ---> generate a compressed sensing demo for synthetic signal using 7 different algorithms.
* DivSBL.m                  ---> The main function for Diversified Block Sparse Bayesian Learning (DivSBL) algorithm.
* BSBL.m                    ---> The function file for BSBL algorithm from [6].

Folder 2: demo_Audio

* DEMO_audio.m    ---> generate a compressed sensing demo for AudioSet from [7].
(Some audio in WAV format for testing purposes are also provided here.）

Folder 3: demo_Image

* DEMO_image_public_parrots.m   ---> generate a compressed sensing demo for Parrot image.
* DWTM.mat                      ---> A decrete wavelet transform matrix provided by [3].
* CS_test_images                ---> Some classic test images.

For bug reports, please contact me at email: yanhaozhang@buaa.edu.cn.

Authors: Yanhao Zhang, Zhihan Zhu and Yong Xia.

Beihang University, Jan, 27, 2024.

References

[1] Ji, Shihao, Ya Xue, and Lawrence Carin. "Bayesian compressive sensing." IEEE Transactions on signal processing 56.6 (2008): 2346-2356.

[2] Grant, Michael, and Stephen Boyd. "CVX: Matlab software for disciplined convex programming, version 2.1." (2014).

[3] Fang, Jun, et al. "Pattern-coupled sparse Bayesian learning for recovery of block-sparse signals." IEEE Transactions on Signal Processing 63.2 (2014): 360-372.

[4] Van Den Berg, Ewout, and Michael P. Friedlander. "SPGL1: A solver for large-scale sparse reconstruction." (2007): 135.

[5] Huang, Junzhou, Tong Zhang, and Dimitris Metaxas. "Learning with structured sparsity." Proceedings of the 26th Annual International Conference on Machine Learning. 2009.

[6] Zhang, Zhilin, and Bhaskar D. Rao. "Extension of SBL algorithms for the recovery of block sparse signals with intra-block correlation." IEEE Transactions on Signal Processing 61.8 (2013): 2009-2015.

[7] Gemmeke, Jort F., et al. "Audio set: An ontology and human-labeled dataset for audio events." 2017 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2017.