期刊
NEUROCOMPUTING
卷 443, 期 -, 页码 272-291出版社
ELSEVIER
DOI: 10.1016/j.neucom.2021.02.053
关键词
Wavelet tight frame; Single image restoration; Adaptive hard thresholding; Nonconvex nonsmooth optimization; Alternating minimization
资金
- NSFC [12001005, 11701079, 61806134, 62076170]
- Anhui University [Y040418173]
- Natural Science Research Project of Anhui Universities [KJ2019A0005, KJ2019A0032]
- Science and Technology Planning Project of Sichuan Province [2020YFG0324]
- Fund of Sichuan UniversityTomorrow Advancing Life
- Natural Science Foundation of Anhui Province [2008085QF286]
- Fundamental Research Funds for the Central Universities [2412020FZ023]
- Jilin Provincial Department of Education [JJKH20190293KJ]
- Sichuan University Innovation Spark Project [2019SCUH0007]
This paper proposes a novel tight frame-based algorithm that simultaneously exploits sparsity and support priors for image restoration. Experimental results show that the proposed method is more effective than standard algorithms and outperforms other methods in both objective and perceptual quality.
Image restoration problems, i.e., recovery of an original high-quality image from the degraded observation, arise in various science and engineer areas. Over the past decades, the framelet-based methods are particularly investigated and adopted, owing to the excellent ability of sparse approximating the piecewise-smooth functions such as natural images. In this paper, we propose a novel tight frame-based l(2)-relaxed truncated l(0) analysis-sparsity model that simultaneously exploiting the sparsity and support priors. The resulting nonconvex nonsmooth optimization problem is addressed by using the proposed proximal alternating adaptive hard thresholding (PAAHT) method. We also proved that the sequence generated by the proposed algorithm sublinearly converges. Numerical experiments on several typical image restoration problems demonstrate that the proposed method is more effective than the standard sparsity-inducing algorithms and outperforms several state-of-the-art methods in both objective and perceptual quality. (C) 2021 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据