An Improved Particle Swarm Optimization Algorithm for Image Matching
Ru An, Chen Chun-ye, Huilin Wang · 2009
Image matching is widely applied in the areas of pattern recognition, computer vision, medicine, remote sensing, aircraft navigation and movement tracking. In this paper, an improved particle swarm optimization algorithm based on variable swarm population size and mutual information as similarity measure function is proposed for image matching. The aim is to enhance the overall performance of image matching. The proposed scheme adjusts the population size in terms of the diversity of the population. The algorithm presented is compared with the exhaustive search based on mutual information, and standard PSO. Remote sensing images captured by different sensors with different resolutions are as testing data. It is proved that the algorithm the paper suggested is effective for image matching.