An Image Denoising Method Based on Group Sparsity and Low Rank

• 作者: ZHAO Yu ZHANG Tao
• 来源期刊: 武汉大学自然科学学报（英文版）
• 年，卷（期）: 2021,（4）
• 所属期刊栏目: Computer Science
• 页码范围: 349-357
• 页数: 9页
• 分类号: TP751
• 语种: 英文
• DOI: 10.19823/j.cnki.1007-1202.2021.0044

摘要

In this paper,we propose an image denoising method combining the priors of non-local self similarity (NSS),low rank and group sparsity.In the proposed scheme,the image is decom-posed into overlapping patches,and then these patches are classi-fied by the K-means clustering.Patches in each cluster are stacked into a matrix and then are decomposed into low frequency com-ponent and high frequency component through 2-D wavelet trans-form.Intuitively,the low frequency component should be a low rank matrix.We show that the high frequency component can be recovered by weighted mixed norm minimization which is also known as group sparse model.Then we propose an image denois-ing model using nuclear norm and weighted mixed norm as regu-larizers to enforce the priors on the low and high frequency.The proposed model can be solved efficiently in the framework of alternating direction multiplier method (ADMM) algorithm.Sev-eral experiments are carried out to verify the performance of the proposed model.