Locally Weighted Enhanced DTW Based Online Signature Verification

In this paper, we present enhance Dynamic Time Warping (DTW) based method on the online Signature Verification (SV) by employing the code-vectors generated from Vector Quantization (VQ) to match the aligned pairs in the warping path. The DTW based method is used to compute the distance score between the test signature and the genuine enrolled signatures, for the decision making. In order to improve the results, we conducted the evaluation using MCYT-100 database and obtained an Equal Error Rate (EER) of 1.55% and SVC-2004 database provides the EER of 2.73%. In this work, we exploit the characteristics of the warping path for online SV and obtained enhanced efficacy of the system.