Group Sparse Representation for Restoring Blurred Images with Cauchy Noise

Abstract

For the image restoration problem, recent variational approaches exploiting nonlocal information of an image have demonstrated significant improvements compared with traditional methods utilizing local features. Hence, we propose a new variational model based on the sparse representation of image groups, to recover blurred images with Cauchy noise. To achieve efficient and stable performance, an alternating optimization scheme with a novel initialization technique is used. Experimental results suggest that the proposed method outperforms other methods in terms of both visual perception and numerical indexes.