Chapter 2: Registration

Title: Feature-based deformable image registration with RANSAC based search correspondence

Author(s): Thomas Colleu, Claude Cariou, Jian-Kun Shen, Bogdan J. Matuszewski, Lik-Kwan Shark

Address: Ecole Nationale Superieure De Sciences Appliquees et de Technologies (ENSSAT), 6, Rue de Kerampont, BP 80518, 22305 Lannion cedex, France | Ecole Nationale Superieure De Sciences Appliquees et de Technologies (ENSSAT), 6, Rue de Kerampont, BP 80518, 22305 Lannion cedex, France | ADSIP Research Centre, University of Central Lancashire, Preston PR1 2HE, UK | ADSIP Research Centre, University of Central Lancashire, Preston PR1 2HE, UK | ADSIP Research Centre, University of Central Lancashire, Preston PR1 2HE, UK

Reference: Atlantic Europe Conference on Remote Imaging and Spectroscopy pp. 57 - 64

Abstract/Summary: The paper presents an algorithm for deformable image registration based on point features extracted from input images using the Harris corner detector. The correspondence between the points extracted from different images is established using the RANdom SAmple Consensus (RANSAC) method with the affine and perspective global transformation used to model the deformations. The initial correspondence is established using an enumerative search with a rotation invariant cross-correlation measure. Based on the estimated corresponding set of points, further refinement of the displacement field is achieved through application of the multilevel B-spline technique applied to the overlapping image area. The performance of this method is demonstrated in this paper using two pairs of remote sensing images. The method is also evaluated against the global polynomial transformation and the registration results are compared with the results achieved using the ENVI software.

Order a copy of this article