An Improved General Particle Swarm Optimization Algorithm for Fast Infrared Image Segmentation

The method of infrared image segmentation based on 2-D maximum fuzzy partition entropy is a typical integer programming problem with huge searching space and many local optima. In order to realize fast infrared image segmentation, an improved general particle swarm optimization algorithm is proposed. The algorithm is based on general particle swarm optimization, and it makes use of adaptive balance searching strategy. When the evolution stops, simulated annealing algorithm is introduced to select the current global optimum to be chaotic optimized for the sake of enhancing local searching ability and overcoming premature convergence. Experiment shows that the algorithm can get segmentation parameters quickly and accurately to realize fast infrared image segmentation.