An adaptive algorithm for image de-noising based on fuzzy Gibbs random fields

Because of the flexible cliques and effective prior models, Gibbs Random Field (GRF) has gained more and more attentions in image processing. However, in those GRF-based image denoising algorithms, Gibbs distribution binary potential clique parameter, beta, can't be changed adaptively with different area features when we adopt fuzzy Gibbs random field for image de-noising. The article shows an adaptive algorithm to alter the value of beta. The approach can automatically decrease beta to keep details near the object edges and increase beta to suppress noises in smooth areas. Based on several simulation cases, the proposed adaptive algorithm is compared with the standard GRF algorithm, and the results show that the new algorithm behaves better in identifying and resolving capability.