Medical Image Registration Based on SVD and Modified PSNR

Medical image registration plays a crucial role in clinical diagnosis, treatment, quality assurance, evaluation of curative efficacy and so on. In this paper, by computing the medical image moments, the centroid is obtained, and according to the rotational invariance of the singular values of the matrix of the medical image coordinates, the rotation angles of the reference and floating images are computed respectively, on the foundation of which the initial values for registering the images are generated. When searching the optimal geometric transformation parameters, the modified peak signal-to-noise ratio (MPSNR) is selected as the similarity measure, and the simplex method as multi-parameter optimization. The experimental results show that, this proposed method has a fairly simple implementation, a low computational load, a fast registration and good registration accuracy. It also can effectively avoid trapping in the local optimum. Also, the improved iterative closest point algorithm is introduced. The results reveal that the measure deriving the initial values for registration from SVD is potent strategy.