Support vector machine for hand geometry-based identity verification system

A new approach to classifying the hand geometry features for personal verification is presented. This paper attempts to improve the performance of hand geometry-based systems by applying the Support Vector Machine (SVM) to the template classification task. We also compare the SVM-based approach to vector distance-based and neural network-based approaches as well as to other systems described in the literature and to the popular commercial systems. The testing results show that our system is competitive to the other known solutions.© (2006) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.