蚁群算法和Powell法结合的多分辨率三维图像配准

基于互信息的配准方法具有精度高,鲁棒性强的特点,成为近年来图像配准研究的热点.但基于互信息的目标函数存在许多局部极值,为配准的优化过程带来了很大的困难.该文提出了一种蚁群算法和Powell法相结合的多分辨率搜索优化算法.该算法以互信息作为相似性测度,采用基于小波变换的多分辨率策略,将蚁群算法与Powell法结合起来对三维的CT,MR图像进行了配准.实验结果表明,这种方法能够有效地克服互信息函数的局部极值,大大地提高了配准精度,达到亚像素级.

Multiresolution 3D Image Registration Using Hybrid Ant Colony Algorithm and Powell’s Method

Image registration based on mutual information is of high accuracy and robustness. Hence, it has received much attention these years. Unfortunately, the mutual information function is generally not a smooth function but one containing many local maxima, which has a large influence on optimization. This paper proposes a registration method based on wavelet representation. In this method the mutual information is used as the similarity measure and a hybrid algorithm combined by ant colony algorithm and Powell’s method as the search technique. This method is applied to the 3D registration of CT and MR. Experiments shows that this registration method could efficiently restrain local maxima of mutual information function and the subvoxel accuracy can be achieved.