Sparse Adaptive Iteratively-Weighted Thresholding Algorithm (SAITA) for L p -Regularization Using the Multiple Sub-Dictionary Representation

Abstract

Both L1/2 and L2/3 are two typical non-convex regularizations of Lp (0<p<1), which can be employed to obtain a sparser solution than the L1 regularization. Recently, the multiple-state sparse transformation strategy has been developed to exploit the sparsity in L1 regularization for sparse signal recovery, which combines the iterative reweighted algorithms. To further exploit the sparse structure of signal and image, this paper adopts multiple dictionary sparse transform strategies for the two typical cases p∈{1/2, 2/3} based on an iterative Lp thresholding algorithm and then proposes a sparse adaptive iterative-weighted Lp thresholding algorithm (SAITA). Moreover, a simple yet effective regularization parameter is proposed to weight each sub-dictionary-based Lp regularizer. Simulation results have shown that the proposed SAITA not only performs better than the corresponding L1 algorithms but can also obtain a better recovery performance and achieve faster convergence than the conventional single-dictionary sparse transform-based Lp case. Moreover, we conduct some applications about sparse image recovery and obtain good results by comparison with relative work.

Nonconvex Nonsmooth Low-Rank Minimization for Peneralized Image Compressed Sensing via Group Sparse Representation

“…Hence, to appropriate the 0 -norm by nonconvex function will achieve a more accurate solution [27] [28]. Typical nonconvex surrogate function including -norm [29][30][31],…”

A Convex Constraint Variational Method for Restoring Blurred Images in the Presence of Alpha-Stable Noises

“…A wide range of approaches has been proposed to remove additive Gaussian noise [ 4 , 5 , 6 ]. However, many other noises, such as impulse noise [ 7 , 8 , 9 , 10 , 11 , 12 ], multiplicative noise [ 13 , 14 ], Poisson noise [ 15 , 16 , 17 ], Cauchy noise [ 18 , 19 ], and Rician noise [ 20 ], commonly appear in the real world and thus are studied by many researchers. Another impulsive noise is often caused by alpha-stable noise, which normally appears in many applications, such as wireless communication systems, synthetic aperture radar (SAR) images, biomedical images, and medical ultrasound images [ 21 , 22 ].…”

“…Although the ROF deblurring and denoising model is a very useful deblurring and denoising approach with additive Gaussian noise, it does not achieve good performance in the scenario of non-Guassian environments. As a result, many kinds of variational models based on TV have been proposed for restoring clean images from blurred and non-Guassian noise distribution, such as that of impulse noise [ 7 , 8 , 9 , 10 , 11 , 12 ], multiplicative noise [ 13 , 14 ], Poisson noise [ 15 ], Cauchy noise [ 18 , 19 ], and Rician noise [ 20 ]. Based on different noise distributions, and data fidelity terms, one can obtain appropriate variational models for image denoising and deblurring in the presence of different noises.…”

GARLM: Greedy Autocorrelation Retrieval Levenberg–Marquardt Algorithm for Improving Sparse Phase Retrieval

“…It utilizes a linear combination of a small number of elements to describe the signal in some bases or dictionaries, simplifying the representation of complex signals [39,40]. Hence it is widely used in multiple image processing fields [41][42][43]. More recently, several robust phase recovery algorithms combating Gaussian noise have been proposed for targeting the noise in the measured values.…”