A MAP Approach for Image Deblurring Based on Sparsity Prior and Laplacian Mixture Modeling

This paper addresses the image deblurring problem, where a known linear space-invariant point-spread function (PSF) is to be deconvoluted from a given blurry image, with addictive zero-mean white and homogeneous Gaussian additive noise. We propose a novel MAP image deblurring method based on sparse representation and Laplacian mixture modeling. The research contents mainly include: (1) Establish the deblurring equation by maximizing the posterior probability of representation vector of the target image, (2) Study the parameters estimation algorithms, such as the construction of redundant dictionary and the estimation of covariance matrix, (3) Develop an iterative scheme to solve the deblurring equation. Experimental results show that the proposed algorithm achieves a very competitive performance both in subjective visual quality and PSNR value, compared with other state-of-the-art deblurring algorithms.