Wavelet based unsupervised variational Bayesian image reconstruction approach

In this paper, we present a variational Bayesian approach in the wavelet domain for linear image reconstruction problems. This approach is based on a Gaussian Scale Mixture prior and an improved variational Bayesian approximation method. Its main advantages are that it is unsupervised and can be used to solve various linear inverse problems. We show the good performance of our approach through comparisons with state of the art approaches on a deconvolution problem.