文献类型：会议论文

英文题名：Image denoising using fractional-order non-local TV model

作者：He, Ning[1]; Lu, Ke[2]; Wang, Jin-Bao[3]

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; [3] Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing, 100101, China

第一机构：北京联合大学智慧城市学院

会议论文集：ICIMCS 2014 - Proceedings of the 6th International Conference on Internet Multimedia Computing and Service

会议日期：July 10, 2014 - July 12, 2014

会议地点：Xiamen, China

语种：英文

外文关键词：Computation theory - Equations of motion - Graph algorithms - Lagrange multipliers - Numerical methods

摘要：A fractional order non-local denoising algorithm and the implementation of fractional order non-local filter are introduced and discussed in this paper. The proposed regularize image denoising problem using a total variation (TV) prior on a graph, which based on the Riemann-Liouville numerical algorithm of fractional differential. The solution of this model is obtained by solving the minimum of the fractional functional. An efficient numerical method based on the dual theory is introduced to solve the corresponding Euler-Lagrange equation. The relationship between the regularization parameter and fractional-order is analyzed, which provides an important basis for the design of optimal strategy. The experiments demonstrate the validity of this fractional-order non-local TV model. The results of emulation experiments show that our algorithm's performance is prior to the Gaussian smoothing filter. Copyright 2014 ACM.