Robust image restoration via random projection and partial sorted ℓ_p norm

详细信息    查看全文

• 作者：Han Pan ; ^{hanpan@sjtu.edu.cn}Author Vitae ; Zhongliang Jing ^{zljing@sjtu.edu.cn}Author Vitae ; Minzhe Li ^{lmznuaa@126.com}Author Vitae
• 关键词：Image restoration ; Low rank approximation ; Random projection
• 刊名：Neurocomputing
• 出版年：2017
• 出版时间：26 January 2017
• 年：2017
• 卷：222
• 期：Complete
• 页码：72-80
• 全文大小：1130 K
• 卷排序：222

文摘

Image restoration plays an important role in video technology. In this paper, a robust image and video denoising method, based on random projection and partial sorted ℓ_{p-norm, is proposed. First, the input signal is decomposed into two components: a low rank component and a sparse component. The low rank component is approximated by random projection. Second, the sparse one is recovered by partial sorted ℓp-norm. A generalized iterative thresholding shrinkage solver is developed for the resulting problem. Some theoretical results about sparse random projection are provided. Numerical experiments for mixed Gaussian and random value impulsive noise demonstrated that the proposed method outperforms some state-of-art restoration methods, both quantitatively and visually.}