A Time Dependent Model via Non-Local Operator for Image Restoration
Image is ubiquitous in modern communication. However, during the process like image acquisition and transmission, blur will appear on the reproduced image. This paper presents a time dependent model for image restoration based on the non-local total variation (TV) operator, which is robust to the noise and makes full use of the spatial information distributed in the different image regions. Experiment results demonstrate that the proposed model produces results superior to some existing models in both visual image quality and quantitative measures.
KeywordsImage restoration Inverse problem Variational model Nonlinear diffusion
The authors would like to thank Yifei Lou and Xiaoqun Zhang for supplying the Matlab implementation of their algorithm. This work was supported by the Project of the key National Natural Science Foundation of China under Grant No.60736010, No.60902060, and the Defense Advanced Research Foundation of the General Armaments Department of the PLA under Grant No.9140A01060110 JW0515.
- 3.Zhang K, Zhang TX, Zhang BY (2006) Nonlinear image restoration with adaptive anisotropic regularizing operator. Opt Eng 45:127004 Google Scholar
- 5.Rudin L, Osher S (1994) Total variation based image restoration with free local constraints. In: Procedings of the first IEEE ICIP, Austin, USA, pp 31–35Google Scholar
- 8.Lou YF, Zhang XQ, Osher S, Bertozzi A (2009) Image recovery via nonlocal operators. J Sci Comput 42:185–197 Google Scholar