Research And Application Of Multi-modal Medical Image Registration Algorithm

This paper briefly introduced the subject of medical image registration, such as the purpose, the significance and the research status in this field. And made a Careful study and analysis of the similarity measure and optimization algorithms in medical image registration. Made some experiment in order to verify the accuracy and reliability of normalized mutual information.This paper proposed an image registration method that based on improved genetic algorithm. It proved that the importance of every gene in the chromosome is different. This method improves the algorithm’s accuracy and search speed through setting different mutation of probability on every gene in the chromosomal gene cluster.Then experiment on improved GA and standard GA shows that the performance of the improved method is higher than the original method. The improved method has strong robustness.Finally, pointed out some problems that should be solved in the field of medical image registration in the future based on the summary of the literature that I have read.