Image Source Separation Using Color Channel Dependencies

We investigate the problem of source separation in images in the Bayesian framework using the color channel dependencies. As a case in point we consider the source separation of color images which have dependence between its components. A Markov Random Field (MRF) is used for modeling of the inter and intra-source local correlations. We. resort to Gibbs sampling algorithm for obtaining the MAP estimate of the sources since non-Gaussian priors are adopted. We test the performance of the proposed method both oil synthetic color texture mixtures and a realistic color scene captured with a spurious reflection.