The impact of visual and blind signing on signature biometrics

Abstract

The ubiquitous nature of our digital lifestyle raised many security issues including signature imitation and stealing of our identity. Therefore, there is a need for robust systems to verify or identify the signatory. In this paper, in contradistinction to other researchers working in signature biometrics, we investigate and explore the impact of blind and visual signing in signature biometrics for online signature identification. Experimental performance evaluation, using the publicly available SUSIG signature database, is carried out to provide some new and preliminary insights into the relationship between different practical factors, in particular clarifying the impact on identification performance of the blind and visual signing data collection protocols used to support the signature processing. Our results explored that adaption of the blind or visual signing data collection protocols has impact on the recognition performance less critical than hitherto expected.