Novel similarity measure based on arithmetic-geometric mean distance for multimodal image registration
SHI Yong-gang, ZOU Mou-yan (Institute of Electronics, Chinese Academy of Sciences, Beijing100080, China)
According to the relation between the intensity joint probability distribution function of two images and the similarity between images, a new similarity measure for multimodal image registration is proposed which is based on the distance between the arithmetic mean and the geometric mean of two probability distribution functions. Unlike information theoretic registration measures, the new measure do not require the condition of absolute continuity to be satisfied by the probability distribution involved. The results of experiment show that the new similarity measure is more tolerable to noise and requires less computational cost than the ones based on the information theory.