An L0 regularized cartoon-texture decomposition model for restoring images corrupted by blur and impulse noise

作者：

Highlights：

• We propose a novel L0 regularized image decomposition model based on coupled sparse representation for image restoration.

• We propose a weighted L1–L2 fidelity term for image deblurring and impulse noise removal.

• We develop an optimization method based on an alternating strategy with the half-quadratic splitting method to solve the proposed multi-variable minimization problem.

• We conduct extensive experiments on images with various characteristics and present detailed objective and subjective comparisons with existing state-of-the-art methods.

摘要

•We propose a novel L0 regularized image decomposition model based on coupled sparse representation for image restoration.•We propose a weighted L1–L2 fidelity term for image deblurring and impulse noise removal.•We develop an optimization method based on an alternating strategy with the half-quadratic splitting method to solve the proposed multi-variable minimization problem.•We conduct extensive experiments on images with various characteristics and present detailed objective and subjective comparisons with existing state-of-the-art methods.

论文关键词：L0 sparsity,Cartoon-texture,Image deblurring,Impulse noise removal

论文评审过程：Received 14 August 2019, Revised 22 November 2019, Accepted 23 December 2019, Available online 27 December 2019, Version of Record 2 January 2020.

论文官网地址：https://doi.org/10.1016/j.image.2019.115762