Multi-Scale Image Deblurring Based on Local Region Selection and Image Block Classification
Chenchen Yu, Hai-Peng Tan, Qiang Chen · 2017
A multi-scale image deblurring method based on local region selection and image block classification is proposed in this paper. Firstly, optimal local region is automatically chosen from the original blurred image and built to pyramid images. Then, the blind restoration problem is solved by using a Variational Bayesian model that is constructed by considering the blur kernel sparse features as prior knowledge. The coarse deblurring results are obtained using the non-blind deconvolution based on image gradient sparsity. Finally, image blocks in the coarse results are classified and adaptively filtered in order to obtain the fine deblurring results. The experimental results on natural, remote sensing and medical images demonstrate that the proposed method can effectively remove fuzziness while preserving the edges and the details.