The Registration And Fusion For Multi-modal Medical Image

With development of modern medical imaging systems,medicalimages become more and more important in medical applications. It brings diagnostics,computer aided surgery,focus monitoring with support of high quality information.However,there are new problems for us to solve. Medical image registration is a way of eliminating the incon-stancy. The term of image registration'comes from electronic engineering, which was defined as:image registration is a course of processing two or more images obtained from different sensors, different time,different angle or so no to align them with best possible match.Medicine image registration is one of the solutions to the problem. The so-called medicine image registration, actually is to seek the best mapping from a series image space to another one. Medicine image registration, whether in the arena of the clinical medical or the computer, is the hot research topic at present. As a multi-disciplinary overlapping research area, Medicine image registration is an important application of the computer graphics and the imagery processing in the bio-medical engineering domain. It involves digital image processing, the computer graphics as well as the knowledge in the domain of medicine and has important application in aspects and so on diagnostic medicine, surgery plan and analog simulation as well as radiotherapy plan. Therefore, it is of vital academic significance and application value to research the medicine image registration. This article, embarked from the two-dimensional medicine image's rigid transformation, has studied multimoding medicine image registration method based on mutual information. The mutual information law is to weigh the similarity measure function of two image registrations based on mutual information. When two image registrations match, mutual nformation comes to the maximum. This method has few manual intervention, and only relies on image's information. It does not need any supposition or the prior medicine knowledge, and does not need to carry on pretreatments to the image, such as characteristic point extraction, tissue typing and so on.This article realized the fusion of the CT and the MRI images with the application of the method of spatial structure fusion based on the pixel of the images. The method of spatial structure fusion is to get a combination of some respective part of the CT and the MRI images with the consideration of different space position. Two images with different nature coexisted in the new one, in which there is no increase of gray level. Through image fusion, we can combine the merit or the complementarity of the image and finally develop the technology to achieve new image with more abundant information.