基于粒子群优化算法的多模态医学图像刚性配准

提出了一种基于轮廓特征点及利用PSO(粒子群优化)求解多模态医学图像自动配准新方法．首先采用数学形态学中腐蚀和膨胀算法对图像进行预处理，用区域生长法提取图像的边缘；再用subtractive聚类算法提取出轮廓特征点，将两个特征点集的均方根极小值作为配准准则，然后用PSO算法求解空间变换参数．该算法适用于多模态医学图像配准，与其他算法相比，PSO算法具有操作方便、可靠性好、不易陷入局部极值等优点。

A method for multimodality medical image registration based on particle swarm optimization

Multimodality medical image registration plays an important role in clinical diagnosis and therapy planning. A novel automatic registration method based on contour feature points and particle swarm optimization (PSO) is presented. Firstly, the images are pretreated by using rust and expansion operations, and the edges of images are detected by using region growth method; then an algorithm called subtractive is used to educe the contour feature points; the minimum value of the root-mean-square for the two feature point sets is regarded as the registration principle; finally, the translation parameters are calculated by using PSO algorithm. The experiments help illustrate that this method is efficient for multimodality medical image registration and has many advantages, such as simplicity, robustness and can avoid getting into local minimum, etc.