Image registration using curvature scale space corner and pseudo-Zernike moments

In this paper, an novel image registration method was proposed by using curvature scale space (CSS) corner and the pseudo-Zernike moments. It is composed of three main steps: firstly extraction of feature points using curvature scale space corner, then obtaining the correspondence between the feature points of the reference and input image based on the pseudo-Zernike moments defined on these feature points neighborhood, finally estimating the transformation parameters mapping the input image to the reference one by applying RANSAC. Experimental results demonstrate that the proposed algorithm is robust to translation, rotation, and slight scaling.