摘要
Medical image registration is becoming more and more important in diagnosis, surgical planning, guiding treatment and monitoring disease progressing, it can align different images to get more valuable information. The programming and development of image registration is rather troublesome, Insight Toolkit (ITK) is a very professional tool, but it is not easy to use. In this paper, first we introduce the ITK architecture and registration framework in detail, and installation under Windows operating system and configuration using python language. Secondly, we introduce the main image registration algorithms of ITK through experiments, including Mattes Mutual Information registration, centered transforms, Multi- Resolution registration, Deformable registration based on FEM and so on. Finally, we give the calculation process and experimental results. With the help of ITK framework, different algorithms are compared and verified, and suitable algorithms are selected for application in engineering, which can shorten the software development cycle and greatly reduce the development cost.