Multi-modal Medical Image Registration Based on Gradient of Mutual Information and Hybrid Genetic Algorithm

Genetic algorithm is easy to fall into premature and Powell algorithm depends on the initial value, according to the characteristics of these two algorithms, in this paper, we combined the two optimization algorithms, the optimal parameter solution which is obtained by genetic algorithm is used as the initial value in Powell algorithm, and then Powell algorithm is used to obtain global optimization parameter solution. In the process of optimizing, the maximum gradient of mutual information is used for measuring the similarity between the two registered images. The experiment shows that the registration algorithm proposed in this paper can do more effectively and more accurately work than the optimizing algorithm based on mutual information and genetic algorithm. And the registration accuracy could achieve sub-pixel.