A Modified Particle Swarm Optimization Algorithm

Particle swarm optimization is a new heuristic global optimization algorithm based on swarm intelligence. The algorithm is simple, easy to implement and has good performance of optimization. Now it has been applied in many fields. However, when optimizing multidimensional and multimodal functions, the basic particle swarm optimization is apt to be trapped in local optima, which is called premature. This paper proposes a modified optimization method (MPSO), which considers for convergence speed and search capacity. This modified algorithm has stronger exploitation ability, so it can prevent premature well. Simulation results show that this modified algorithm performs better performance. It is used in segmentation of infrared image. The experimental results show that the modified PSO not only realizes the image segmentation well, but also improves the speed greatly.