Geometric Centroids and their Relative Distances for Off-line Signature Verification

In this paper, we propose a new approach for symbolic representation of off-line signatures based on relative distances between centroids useful for verification. Distances between centroids of off-line signatures are used to form an interval valued symbolic feature vector for representing signatures. A method of off-line signature verification based on the symbolic representation is presented. We investigate the feasibility of the proposed representation scheme for signature verification on a MCYT_ signature database. We cluster similar signatures in each class and also investigate the cluster based symbolic representation for signature verification. Unlike other signature verification methods, the proposed method is simple and efficient. Several experiments are conducted to demonstrate the efficacy of the proposed scheme.