In this paper, a new concept for classifying handwriting data and its analysis for biometric user authentication is presented. The concept's characteristic is the combination of syntax and semantics. It implies a determination of four distinct levels of syntax and semantics to lower complexity and structure information. We demonstrate the concept's impacts on on-line handwritings and the user verification, and clarify the benefit of applying information of higher levels of semantics within the authentication methods. As a result we are able to evaluate techniques for biometric user authentication. Furthermore, we precisely outline and reason a more accurate biometric user authentication system, due to the classification given by the Verifier-Tuple concept. © IFIP International Federation for Information Processing 2005.
CITATION STYLE
Oermann, A., Dittmann, J., & Vielhauer, C. (2005). Verifier-tuple as a classifier for biometric handwriting authentication - Combination of syntax and semantics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3677 LNCS, pp. 170–179). https://doi.org/10.1007/11552055_17
Mendeley helps you to discover research relevant for your work.