Print Email Facebook Twitter Performance of optimal registration estimators Title Performance of optimal registration estimators Author Pham, T.Q. Bezuijen, M. Van Vliet, L.J. Schutte, K. Luengo Hendriks, C.L. Faculty Applied Sciences Department Quantitative Imaging Group Date 2005-05-25 Abstract This paper derives a theoretical limit for image registration and presents an iterative estimator that achieves the limit. The variance of any parametric registration is bounded by the Cramer-Rao bound (CRB). This bound is signal-dependent and is proportional to the variance of input noise. Since most available registration techniques are biased, they are not optimal. The bias, however, can be reduced to practically zero by an iterative gradientbased estimator. In the proximity of a solution, this estimator converges to the CRB with a quadratic rate. Images can be brought close to each other, thus speedup the registration process, by a coarse-to-fine multi-scale registration. The performance of iterative registration is finally shown to significantly increase image resolution from multiple low resolution images under translational motions. Subject registration limitCramer-Rao boundregistration bias,super-resolution To reference this document use: http://resolver.tudelft.nl/uuid:a8ba0d08-edf6-4135-8371-39a12e6c53db Publisher SPIE ISSN 0277-786X Source Proceedings of SPIE, 2005 vol. 5817 Part of collection Institutional Repository Document type conference paper Rights (c)2005 Pham, T.Q., Bezuijen, M., Van Vliet, L.J., Schutte, K., Luengo Hendriks, C.L. Files PDF PerformancePham.pdf 499.05 KB Close viewer /islandora/object/uuid:a8ba0d08-edf6-4135-8371-39a12e6c53db/datastream/OBJ/view