Multispectral image context classification using stochastic relaxation

A new multispectral image context classification, which is based on a stochastic relaxation algorithm and Markov-Gibbs random field, is presented. The implementation of the relaxation algorithm is related to a form of optimization programming using annealing. The authors motivate a Bayesian context decision rule, and a Markov-Gibbs model for the original Landsat MSS (multispectral scanner) image is introduced, and then develop a new contextual classification algorithm, in which maximizing the posterior probability (MAP) is based on stochastic relaxation, an annealing optimization method. Finally, experimental results that are based on simulated and real multispectral remote sensing images to Fhow how classification accuracy is greatly improved are presented. The algorithm is highly parallel and exploits the equivalence between Gibbs distribution\ and Markov random fields (MRF).