Medical image fusion fuses the complementary information from different medical images, which makes the fused images contain clearer details and richer information than any original ones. And it provides more direct, comprehensive and clearer judgement to lesions observation and diagnosis so as to improve the detection rate of disease. Based on CT and MRI images, this dissertation concentrates the research on the miage fusion technology of wavelet transform.With the beginning, the dissertation describes the background of medical image fusion technology, the current study situation at home and abroad, and the basic concepts and application in the medical field of medical image fusion, makes a brief introduction to the imaging principles and characteristics of common medical images, and fucuses on the imaging principles and characteristics of CT and MRI. Then the dissertation makes comparative analyses on the present fusion algorithms, including principal component analysis, pyramid decomposition and wavelet decompositon, and on the different fusion rules, and introduces the present evaluation system of fused images. Finally, the last two chapters respectively introduce the fusion algorithm presented in the dissertation——image fusion based on discrete stationary wavelet transform, principal component analysis and nonsubsampled Contourlet.The experiment shows that the algorithm presented in the dissertation can better fuse CT, MRI images and the fused images have clear bones and soft tissue. Compared to the present fusion algorithms, the one in the dissertation owns more advantages. |