Optimization strategies for data collections used in evaluating dynamic signature authentication systems

The present article addresses a key issue in evaluating performances of biometric authentication systems, which is the data collection optimization. This particular case study is represented by a system using the dynamic signature as the characteristic to be analyzed. The first 2 sections describe a data acquisition system and also the procedure used to accomplish the acquisition process. Next we present a number of optimization strategies of the data set, that we will demonstrate they can lead to a notable performance increase. Further, we describe an evaluation system and we finally present the obtained results proving that the optimization measures were successful.