The Effects of Grey Levels on Mutual Information Based Medical Image Registration
Jiaru Lin · Beijing shengwu yixue gongcheng · 2003
Medical image registration is an active field currently. Among all the registration methods, mutual information based methods have been accepted as the most accurate and automated method. Essentially, mutual information based methods are based on doing statistics of grey values of images. Therefore, if different representations of grey levels of one images are used, registration accuracy will be affected. In this paper, based on analysis of compression of grey levels for image quality, and based on the affects of mutual information registration method, series of registration tests of multi modes medical images were carried out. Affects on image registration by different grey levels are compared from two sides:registration accuracy and time length for calculation. According to these results and analysis, the recommended grey levels to be used in mutual information registration are given.