Non-Blind Image Deblurring Method Using Shear High Order Total Variation Norm

In this paper, we propose a shear high-order gradient (SHOG) operator by combining the and (HOG) operator. Compared with HOG operator, proposed SHOG can incorporate more directionality detect abundant edge information. Based on extend total variation (TV) norm to (SHOTV), then SHOTV deblurring model. We also study some properties of show that matrices are Block Circulant Blocks (BCCB) when angle is [see formula in PDF]. The model solved efficiently alternating direction method multipliers (ADMM). Experimental results demonstrate outperforms state-of-the-art non-blind methods both objective perceptual quality.