Medical Image Registration Based on More Features and Artificial Immune Algorithm

In this paper, an algorithm using point feature and intensity feature combined with the Artificial Immune algorithm is presented. First, the feature points of the two images are extracted by Harris corner detector to reduce the amount of computation. Then, the mutual information (MI) is used to be the similarity measure for MI algorithm based on intensity has excellent robustness and accuracy. Finally, the transformation parameters are calculated by Artificial Immune (AI) algorithm. In the end of this paper, experiments of combining both the feature points and intensity as well as AI algorithm are presented for clinical images. Experiments results show that the algorithm is effective and accurate, and it could reduce the amount of computation largely.