The Non-Rigid Medical Image Registration Based on the Histogram Decomposition and Improved Energy Function

Abstract

A new method of non-rigid image registration is proposed in this paper. The algorithm uses the local spatial intensity histograms to design a new type of attribute vector for each point in an intensity image and also uses the Newton's action-reaction law method form a symmetric relationship, which tends to improve the registration by establishing symmetric image interdependence. The new energy function uses the image local histogram information and boundary information and the matching by the new energy function makes both the global and local image matched well. From the experiments we can find the algorithm proposed in this paper has an high accuracy in non-rigid image registration.