×

Backtracking-based iterative regularization method for image compressive sensing recovery. (English) Zbl 1461.68243

Summary: This paper presents a variant of the iterative shrinkage-thresholding (IST) algorithm, called backtracking-based adaptive IST (BAIST), for image compressive sensing (CS) reconstruction. For increasing iterations, IST usually yields a smoothing of the solution and runs into prematurity. To add back more details, the BAIST method backtracks to the previous noisy image using L2 norm minimization, i.e., minimizing the Euclidean distance between the current solution and the previous ones. Through this modification, the BAIST method achieves superior performance while maintaining the low complexity of IST-type methods. Also, BAIST takes a nonlocal regularization with an adaptive regularizor to automatically detect the sparsity level of an image. Experimental results show that our algorithm outperforms the original IST method and several excellent CS techniques.

MSC:

68U10 Computing methodologies for image processing
68P30 Coding and information theory (compaction, compression, models of communication, encoding schemes, etc.) (aspects in computer science)

Software:

TwIST
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Huang, J.; Zhang, S.; Metaxas, D.; Efficient MR Image Reconstruction for Compressed MR Imaging; Med. Image Anal.: 2011; Volume 15 ,670-679.
[2] Mahrous, H.; Ward, R.; Block Sparse Compressed Sensing of Electroencephalogram (EEG) Signals by Exploiting Linear and Non-Linear Dependencies; Sensors: 2016; Volume 16 .
[3] Chen, C.; Huang, J.; Exploiting the wavelet structure in Compressed Sensing MRI; Magn. Reson. Imaging: 2014; Volume 32 ,1377-1389.
[4] Burns, B.L.; Wilson, N.E.; Thomas, M.A.; Group Sparse Reconstruction of Multi-Dimensional Spectroscopic Imaging in Human Brain in Vivo; Algorithms: 2014; Volume 7 ,276-294. · Zbl 07042207
[5] Som, S.; Schniter, P.; Compressive imaging using approximate message passing and a markov-tree prior; IEEE Trans. Signal Process.: 2012; Volume 60 ,3439-3448. · Zbl 1393.94714
[6] Dong, W.; Shi, G.; Li, X.; Ma, Y.; Huang, F.; Compressive sensing via nonlocal low-rank regularization; IEEE Trans. Image Process.: 2014; Volume 23 ,3618-3632. · Zbl 1374.94085
[7] Metzler, C.; Maleki, A.; Baraniuk, R.G.; From denoising to compressed sensing; IEEE Trans. Inf. Theory: 2016; Volume 62 ,5117-5144. · Zbl 1359.94137
[8] Egiazarian, K.; Foi, A.; Katkovnik, V.; Compressed sensing image reconstruction via recursive spatially adaptive filtering; Proceedings of the IEEE International Conference on Image Processing (ICIP): ; ,I-549-I-552.
[9] Shi, J.; Zheng, X.; Yang, W.; Robust sparse representation for incomplete and noisy data; Information: 2015; Volume 6 ,287-299.
[10] Dabov, K.; Foi, A.; Katkovnik, V.; Egiazarian, K.; Image denoising by sparse 3D transform-domain collaborative filtering; IEEE Trans. Image Process.: 2007; Volume 16 ,2080-2095.
[11] Bioucas-Dias, J.; Figueiredo, M.; A new TwIST: Two-step iterative shrinkage/thresholding algorithms for image restoration; IEEE Trans. Image Process.: 2007; Volume 16 ,2992-3004.
[12] Sun, T.; Cheng, L.; Reweighted fast iterative shrinkage thresholding algorithm with restarts for l1-l1, minimisation; IET Signal Process.: 2016; Volume 10 ,28-36.
[13] Daubechies, I.; Defriese, M.; DeMol, C.; An iterative thresholding algorithm for linear inverse problems with a sparsity constraint; Commun. Pure Appl. Math.: 2004; Volume 57 ,1413-1457. · Zbl 1077.65055
[14] Beck, A.; Teboulle, M.; A fast iterative shrinkage-thresholding algorithm for linear inverse problems; SIAM J. Imaging Sci.: 2009; Volume 2 ,183-202. · Zbl 1175.94009
[15] Afonso, M.; Bioucas-Dias, J.; Figueiredo, M.; Fast image recovery using variable splitting and constrained optimization; IEEE Trans. Image Process.: 2010; Volume 19 ,2345-2356. · Zbl 1371.94018
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.