Registration based on mutual information is a typical method in medical image registration. Mutual information is a common similarity measure in image registration, which has excellent robustness and accuracy, but large calculation amount makes it difficult to be applied to clinics. In this paper, a registration algorithm based on multi-resolution and hybrid optimization is adopted to implement 2-dimension monomodal and multimodal registrations of MRI and CT images of human heads with different numbers of gray bins. Results of experiments show that registration precisions have not notable change with 32, 64 gray bins, compared with 256 gray bins, whereas the computation costs decrease remarkably.