Study of Medical Image Registration Based on Second-Order Mutual Information

Second-order MI (mutual information) is an improved measure of first-order MI in image registration, enhancing the robustness and accuracy of first-order MI. However, in the process of implementing registration based on second-order MI by Rueckert et al., the related techniques can still be improved. So, in this paper, we study the affection of grey levels and neighborhood information on the technology of image registration based on second-order MI from some experiments and get the best grey levels and neighborhood information to improve this technology. At last, the effects of noise on second- order MI are analyzed to explain some conclusions drawn in our paper.