Image Blind Deblurring using Luminance Minimization Prior
Xiangxia Li, Fang Liu, Junjian Feng, Yanhong Chen · 2023
We propose a novel image prior for image blind deblurring. Observing that the luminance component of YUV colour space has very low intensity value in the clear images. Based on this properties, we propose a novel luminance minimization prior for blind image deblurring. In particular, a L0-regularized term is applied to enforce the sparsity of the luminance component. Then an efficient algorithm is presented for solving the proposed formulation, which obtains the clear image. To validate the proposed approach, we conduct extensive experiments on deblurring images. The results show the effectiveness of the proposed method for image blind deblurring.