Evaluation of strengths and weaknesses of dynamic handwritten signature recognition against forgeries

摘要

This paper proposes a methodology for evaluating the robustness of handwritten signature biometrics against forgeries. A particular emphasis has been applied in consider the viability of the attacks under operational environments. In order to carry out this work, not only a methodology but a toolbox has been developed. Such toolbox stores data, not only from a signing pad connected to a computer, but also from touch-screen devices such as smartphones and tablets. The proposed methodology has been tested by performing a robustness evaluation of a DTW-based algorithm, and with a database of genuine signatures acquired using a STU-500 pad as an input for the evaluation toolbox. The results will proof the validity of the methodology proposed, as well as show some of the strengths and weaknesses of this biometric modality. It is important to note that those strengths and weaknesses may only apply to the particular algorithm used in the evaluation, and may differ for other algorithms, in particular those that incorporates Presentation Attack Detection (PAD) mechanisms.