Iterative Total Variation Image Deblurring with Varying Regularized Parameter
Binbin Hao, Jianguang Zhu, Yan Hao · 2014
Total variation based model is one of the most effective method for image restoration. In this paper, we consider the total variation (TV) based regularization method and evaluate the regularization parameter for the TV based iterative forward-backward splitting (IFBS) approach. Different parameters with different iterations are obtained. The proposed adaptive iterative forward-backward splitting method does not need to know the initial value of the regularization parameter and does not require any information about the perturbation process. Experimental results demonstrate that the adaptive parameter method is efficient and provide competitive performance.