Image registration using Markov random coefficient and geometric transformation fields

Abstract — Image Registration is central to different appli-cations such as medical analysis, biomedical systems, image guidance, etc. In this paper we propose a new algorithm for multi-modal image registration. A Bayesian formulation is presented in which a likelihood term is defined using an observation model based on coefficient and geometric fields. These coefficients, that represent the local intensity polynomial transformations, as the local geometric transformations, are modelled as prior infor-mation by means of Markov random fields. This probabilistic approach allows one to find optimal estimators by minimizing an energy function in terms of both fields, making the registration between the images possible. I