Blind image deblurring using elastic-net based rank prior

In this paper, we propose a new image prior for blind image deblurring. The proposed prior exploits similar patches of an image and it is based on an elastic-net regularization of singular values. We quantitatively verify that it favors clear images over blurred images. This property is able to facilitate the kernel estimation in the conventional maximum a posterior (MAP) framework. Based on this prior, we develop an efficient optimization method to solve the proposed model. The proposed method does not require any complex filtering strategies to select salient edges which are critical to the state-of-the-art deblurring algorithms. We also extend the prior to deal with non-uniform image deblurring problem. Quantitative and qualitative experimental evaluations demonstrate that the proposed algorithm performs favorably against the state-of-the-art deblurring methods.