Abstract: Mutual information is widely used in registration and optimization algorithms are used to optimize mutual information.A new optimization algorithm，which is a hybrid of Particle Swarm Optimization and Simplex，is proposed in the paper.PSO excels in global searching，while Simplex excels in local searching.When they are hybridded together，a more accurate result can be gotten.Meanwhile，the hybrid algorithm is easy to realize.Furthermore，as mutual information based registration requires too many computations，wavelet transform is introduced into registration.A good result is shown in experiment.

摘要： 互信息被广泛地应用于图像配准，通常采用优化算法来优化互信息。提出了粒子群优化算法和单纯形法相结合的混合优化算法，该混合算法实现简单，粒子群优化算法具有极强的全局搜索能力，能够有效地跳出局部极值，而单纯形法又能够有效地 进行局部搜索，将两种方法相结合进行图像配准能大大提高配准精度。为了减少运算量，提出了利用小波变换的方法来缩小搜索范围，配准精度达到亚像素级。