MONO-MODAL IMAGE REGISTRATION VIA CORRENTROPY MEASURE

The registration of images is a fundamental task in numerousapplications in medical image processing. Similarity measureis an important key in intensity based image registration.Here, we propose correntropy measure as similarity measurein mono modal setting. Correntropy is a important measurebetween two random variables based on information theoreticlearning and kernel methods. This measure is useful in non-Gaussian signal processing. In this paper, this measure is usedin image registration. Here, we analytically illustrate thatthis measure is robust in presence of spiky noise (impulsivenoise). The experimental results show that the proposed similarityhas better performance than traditional intensity-basedsimilarity measures such as Mutual information and SSD

کلیدواژه ها:

image registration ، similarity measure ، Correntropy ، spiky noise ، Free Form Deformation