Improvement of Japanese signature verification by combined segmentation-verification approach

Abstract

This paper proposes a signature verification technique called combined segmentation-verification based on off-line features and on-line features. We use three different off-line feature vectors extracted from full name Japanese signature image and from the sub-images of the first name and the last name. The Mahalanobis distance for each off-line feature vector is calculated for signature verification. The on-line feature based technique employs dynamic programming (DP) matching technique for time series data of the fullname signature and first name and last name. The final decision (verification) is performed by SVM (Support Vector Machine) based on the three Mahalanobis distances and three dissimilarity of the DP matching. In the evaluation test the proposed technique achieved 3.35% EER (Equal Error Rate) with even FRR (False Acceptance Rate) and FAR (False Rejection Rate), which is 3.10% lower than the best EER obtained by the individual technique. This result shows that the proposed combined segmentation-verification approach improves Japanese signature verification accuracy significantly.