A study on blind image restoration of blurred images using R-map
Satoshi Motohashi, Takahiro Nagata, Tomio Goto, Reo Aoki, Haifeng Chen · 2018 International Workshop on Advanced Image Technology (IWAIT) · 2018
Image restoration which restores a clear image from a single blur image is a difficult problem of estimating two unknowns: a point spread function (PSF) and its ideal image. In this paper, we propose a novel blind deconvolution method to alternately estimate PSF and the latent image. And we incorporate the gradient reliability map (R-map) that enables edge selection appropriate for PSF estimation processing. This method improves restoration performance by excluding noise that adversely affects the estimation, and the experimental results show that robustness is improved in our proposed method.