Abstract：The medical image registration method based on particle swarm optimization algorithm and use mutual information as the measure function has the advantages of high accuracy and strong robustness,but it also has some shortcomings such as local convergence and so on.In order to solve these problems,the self-regulating strategy is introduced into the particle swarm algorithm and the multi resolution method is used for image registration.First,using wavelet transform to stratify the source image.Then using improved particle swarm optimization algorithm for image registration,and take the upper layer registration optimization parameter as an initial parameter of a particle swarm at the next layer.It refined layer by layer,finally achieved accurate image registration.The results show that the proposed method can effectively avoid the local convergence problem,the robustness is stronger,and the registration results can reach the sub pixel level.
王栋栋,宋余庆,刘哲(,). 改进粒子群算法与多分辨率结合的医学图像配准[J]. 小型微型计算机系统, 2017, 38(11): 2610-2614.
WANG Dong-dong,SONG Yu-qing,LIU Zhe(,). Medical Image Registration Based on Improved Particle Swarm Optimization Algorithm and Multi-resolution. Journal of Chinese Computer Systems, 2017, 38(11): 2610-2614.