Histogram Transformation for Inter-Modality Image Registration

This paper describes an efficient image transformation method based on histogram information and some prior knowledge of tissue expression in different modalities for registration of images from different image aquisition methods, such as CT and MRI. The metric defined in this paper represents a computationally efficient alternative to traditional, mutual information based methods. The optimization step in the registration algorithm can be carried out either by using a normalized cross-correlation function or as a simple image subtraction, which makes it suitable for hardware implementations.