Off-line signature verification using feature based image registration

Abstract

This paper discusses the issues of off-line signature verification and shows that signature verification methods can and should be improved by applying feature based image registration methods. First the different aspects and problems of signature verification are discussed in conjunction with off-line analysis methods. It is assumed that area based registration methods fail to improve their results because the high level of abstraction they use. To overcome this limitations a feature based image registration method is introduced, altered step by step to get a better matching between the endpoints of two signatures. © 2008 Springer-Verlag Berlin Heidelberg.