A New Single Image Deblurring Algorithm Using Hyper Laplacian Priors
Chia-Feng Chang, Jiunn‐Lin Wu · Frontiers in artificial intelligence and applications · 2015
Motion blur is an ill-posed problem which has been addressed for a long time in digital photographing. It usually happened under the circumstance that surrounding light is not enough. Several methods had been proposed for this problem. However, ringing is inevitable artifacts arising in the deconvolu-tion stage. To suppress undesirable artifacts, regularization based methods have been proposed using natural image priors to overcome the ill-posedness of de-convolution problem. These studies suggest that priors based on natural image statistics can regularize deblurring problems to yield better results.