Restoration of color images subjected to interchannel blurring

This paper deals with the restoration of colored images, distorted by both intra- and inter-channel blur, and corrupted by additive white Gaussian noise. The image is modeled as a Markov Random Field (MRF), and color image restoration is cast as a maximum a posteriori (MAP) estimation problem. We propose a First Order Interchannel Interaction (FOII) model for image restoration. Simulated annealing algorithm is then used to minimize the posterior energy function. We compare the simulation results of the conventional non-interaction (NI) approach and the proposed FOII approach. The proposed model is fairly general, and the results are satisfactory even when interchannel degradation parameter is unknown.