Detecting Forged Handwriting with Wavelets and Statistics

In this paper, we present a method of analyzing handwriting samples (distinguishing between authentic samples and forgeries) and address the question of whether or not the handwriting of different individuals contains a mathematical signature. The method discussed is a modified version of a similar approach towards art presented in ``A Digital Technique for Art Authentication\u27\u27 by Lyu et al. We give a short history of art and handwriting forgery and characteristics examined in order to determine authenticity. We also examine the rationale behind the Lyu et al.\u27s method and the changes that we have made in our application to handwriting analysis