Contour-based multisensor image registration with rigid transformation

A contour-based multisensor image registration algorithm is presented. The geometrical deformation between the input images is supposed to be rigid transformation. The characteristic of this approach is that the registration parameters are calculated according to the centroids and long axes of matched contour pairs in the input images. It is robust as it overcomes the difficulties of control point detection and correspondence in normal feature-based registration techniques. Salient contours are firstly extracted from the input images. Then, contour matching is performed to obtain the closed contour matches and open contour matches. The open contour matches are transferred into closed contour matches. The registration parameters are estimated according to the centroids and the angles of long axes of the closed contour matches. Experiments using real data show that the proposed algorithm works well in multisensor image registration with rigid transformation.