IMAGE REGISTRATION BASED ON LIFTING PROCESS AND GENETIC OPTIMIZATION: AN APPLICATION TO DENTAL IMAGING

摘要

In this paper, a new method for the registration of unimodal 2-D medical images is presented. The proposed method can be dealt with as an object-based registration method without imposing the often limiting prerequisite of segmentation in order to detect the boundary of the object of interest. This is achieved by augmenting the dimensionality of the problem from two-dimensional gray scaling matching to three-dimensional surface matching using the process of lifting in combination with an existing scheme for surface matching. The pseudo-three-dimensional transformation that matches the lifted images incorporates some advantageous characteristics in comparison with the simple two-dimensional affine transformation: Automatic anisotropic correction of brightness-contrast differences and introduction of some non-affine (projective) effects that very often are needed when matching radiographic modalities.