Hybrid Fingerprint Matching on Programmable Smart Cards

• Autores: Tommaso Cucinotta, Ricardo Brigo, Marco Di Natale
• Localización: Trust and Privacy in Digital Business: First International Conference,TrustBus 2004 Zaragoza, Spain, August 30 - September 1, 2004 Proceedings / coord. por Sokratis Katsikas, Günther Pernul, Javier López, 2004, ISBN 3540229191, págs. 232-241
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper presents a hybrid fingerprint matching algorithm combining two heterogeneous schemes, namely the texture-vector and minutiae-based methods. The proposed technique has been designed in order to run on a programmable smart card, with image processing and feature extraction performed on the host, and matching performed by the card device. The two matching algorithms have been carefully tuned in order to achieve an acceptable performance despite the computation and memory constraints. Given the high level of intrinsic security that smart cards already have, and the interactive nature of target applications, the complexity of the problem has been greatly reduced, making such an approach feasible. This is validated by the experimental results we show, gathered from an implementation onto a Java Card device, where acceptable false acceptance and rejection rates are achieved at the cost of a reasonable response time of the device.