Total variation with overlapping group sparsity for speckle noise reduction

详细信息    查看全文

• 作者：Jun Liu^a ; ^b ; ^{junliucd@163.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Ting-Zhu Huang^b ; ^{tingzhuhuang@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Gang Liu^bAuthor Vitae ; Si Wang^bAuthor Vitae ; Xiao-Guang Lv^cAuthor Vitae
• 关键词：Total variation ; Speckle noise ; Synthetic aperture radar ; SSIM
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：5 December 2016
• 年：2016
• 卷：216
• 期：Complete
• 页码：502-513
• 全文大小：4446 K

文摘

Staircase effect usually happens on the total variation (TV) regularized solutions, while the overlapping group sparsity total variation (OGSTV) as a regularization has been proved to be effective for alleviating this drawback. For coherent imaging systems, such as the synthetic aperture radar, the acquired images are corrupted by speckles. In this paper, we propose a speckle noise reduction model based on the regularization of OGSTV. Under the framework of efficient alternating direction method of multipliers, we develop the corresponding algorithm for solving the proposed model. Numerical experiments are presented to illustrate the superiority of the proposed model and efficiency of the corresponding algorithm.