Forensic Validation of Biometrics Using Dynamic Handwritten Signatures

Abstract

Forensic examination of handwritten signatures is an important task that has been used to resolve conflicts for centuries. The incorporation of new technologies into the process of signing documents has created new challenges for this task. In particular, the use of electronic capture devices may compromise the capabilities of forensic examination. However, the forensic examination may not be challenged and may even be improved if, in addition to the signature graph, the temporal signals that are generated during the process of signing are captured. In biometric terms, the acquisition and processing of such temporal signals are referred to as dynamic signature biometric recognition. Unfortunately, the data are captured in a format that a forensic examiner is unable to understand. Therefore, there is a need of adapting this information to allow a forensic examiner to manipulate it and obtain the required measurements. This paper explains this need using the design and development of a desktop application as the guiding thread. After covering this need, a forensic examiner can extract the relevant graphometric features that are necessary for applying graphonomics to signatures and determining the authenticity of a questioned signature compared with a certain signature.