INCORPORATE TV − L ∞ MODEL WITH SPARSE REPRESENTATIONS FOR IMAGE DENOISING, A POST-PROCESSING APPROACH

Sparse representations of images have revoked remarkable in-terest recently. The assumption that natural images admit a sparse decomposition over a redundant dictionary leads to ef-ficient algorithm for image processing. In particular, the K-SVD method has been recently proposed and shown to per-form very well for gray-scale and color image denoising task ([1],[2]). Meanwhile, the TV − l ∞ model with special choice of dictionary has been proved to be very effective for image restoration([3],[4]). In this paper, we propose a hybrid model which combines these two methods and may be regarded as a post-processing procedure for K-SVD. Due to the excellent work of K-SVD and the fact that TV −l ∞ can reconstruct lost information quickly, this hybrid model lead to a new state-of-art denoising performance. Index Terms — Sparse representation, total variation, im-age denoising, dictionary, K-SVD, post-processin