Total variation based image deblurring with nonlocal self-similarity constraint

Abstract

An improved total variation (TV) based image deblurring method combining nonlocal (NL) self-similarity constraint is proposed. TV based image deblurring methods are effective in restoring image structures from blurred image data. However, they tend to over smooth the local image details. This drawback is overcome by combining a local TV regulariser with a NL self-similarity constraint, which helps to sharpen image edges and restore the fine details. The local TV model and NL self-similarity constraint are complementary to each other, making the proposed approach highly effective in removing blur while preserving image edges. Experimental results validate the performance of the proposed approach in both evaluation indexes and visual quality.