RELIABLE ONLINE HUMAN SIGNATURE VERIFICATION SYSTEMS

On-line dynamic signature verification systems were designed and teste d. A data base of more than 10,000 signatures in (x(t), y(t))-form was acquired using a graphics tablet. We extracted a 42-parameter feature set at first, and advanced to a set of 49 normalized features that to lerate inconsistencies in genuine signatures while retaining the power to discriminate against forgeries. We studied algorithms for selectin g and perhaps orthogonalizing features in accordance with the availabi lity of training data and the level of system complexity. For decision making we studied several classifiers types. A modified version of ou r majority classifier yielded 2.5% equal error rate and, more importan tly, an asymptotic performance of 7% false acceptance rate at zero fal se rejection rate, was robust to the speed of genuine signatures, and used only 15 parameter features.