Image deconvolution using a stochastic differential equation approach

Abstract: We consider the problem of image deconvolution. We foccus on a Bayesian approach which consists of maximizing an energy obtained by a Markov Random Field modeling. MRFs are classically optimized by a MCMC sampler embeded into a simulated annealing scheme. In a previous work, we have shown that, in the context of image denoising, a diffusion process can outperform the MCMC approach in term of computational time. Herein, we extend this approach to the case of deconvolution. We first study the case where the kernel is known. Then, we address the myopic and blind deconvolutions.