Wavelet Transform Based Global Features for Online Signature Recognition
Fayyaz A. Afsar, Muhammad Arif, Umair Farrukh · 2005
This paper presents an efficient algorithm for an online signature verification system that is based on the extraction of global features from the spatial coordinates obtained during the online acquisition of a signature using one dimensional wavelet transform. A k-NN classifier is used for classification purposes. Low error rates obtained for both random and skilled forgeries datasets illustrate the feasibility of the algorithm for an online signature verification system