Accuracy of the MTsengv algorithm for on-line biometric signature classification

The first part of the paper describes the MTsengv algorithm for on-line handwritten signatures verification. MTsengv algorithm was created by modifying the Tseng algorithm, in order to increase the accuracy of signature verification. After that, an algorithm for determining the decision threshold is presented. The algorithm is based on False Positive and False Negative error analysis. The second part of the paper describes an experiment in which the performance of a MTseng algorithm was determined using True Positive, True Negative, False Positive and False Negative errors. The mentioned errors are shown tabularly. Finally, the performed comparative analysis of the results shows a higher accuracy of the MTsengv algorithm..