A study of blind image restoration by fast image deconvolution utilizing hyper-laplacian prior
Masanao Sawada, Kyosuke Ohkoshi, Tomio Goto, Satoshi Hirano, Masaru Sakurai · 2013
In this paper, we propose a novel blind deconvolution method in which PSF (Point Spread Function) estimations and regularized deconvolution are calculated alternately. Our proposed method utilizes the hyper-Laplacian prior for deconvolution processes. Experimental results show that our proposed method improves picture quality, especially dark portions, and reduces the computation time by almost 1/3rd that of the conventional method.