Satellite image classification using a modified Metropolis dynamics

摘要: A pseudo-stochastic variation of the Metropolis dynamics for combinatorial optimization in image classification using Markov random fields is presented. At high temperature, behavior algorithm similar to stochastic ones. However, if temperature less than a certain threshold, it becomes deterministic. The length phase controlled by constant threshold used modified dynamics. yields an approximate but usually good solution problem. runs on connection machine. It applied standard pixel problem; objective and subjective comparisons with other algorithms have been made. >