文献类型：会议论文

英文题名：Image deconvolution using 1 sparse regularization

作者：He, Ning[1]; Zhang, Qi[1]; Chi, Yue[1]; Lu, Ke[2]

第一作者：何宁

机构：[1] Beijing Key Laboratory of Information Service Engineering, College of Information Technology, Beijing Union University, Beijing, 100101, China; [2] University of Chinese Academy of Sciences, Beijing, 100049, China

第一机构：北京联合大学北京市信息服务工程重点实验室|北京联合大学智慧城市学院

会议论文集：ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service

会议日期：August 19, 2015 - August 21, 2015

会议地点：Zhangjiajie, Hunan, China

语种：英文

外文关键词：Data handling - Deconvolution - Iterative methods

摘要：This paper studies sparse regularization image deconvolution scheme over the space of measures. This regularization method is the natural extension of the 1 norm of vectors to the setting of measures. The proposed model is composed of data fitting term and regularization term. The regularization parameter controls the degree of sparsity of the solution, 1 fitting term promotes sparsity, and the 2 data fitting term promotes smoothness while preserving edges. We use convex analysis tools and get a simple iterative algorithm. Extensive experiments on image deblurring and image inpainting verify the effectiveness of the proposed algorithm. ? 2015 ACM.