Automatic biometric identification system by hand geometry

Abstract

In this paper we propose a novel and simple method to recognize individuals based on their hand-palm geometry. A compact set of parameters has been extracted and reduced using different transformations. Two classification methods have been implemented: Neural Networks (NN) based on the commonly used Multilayer Perceptron (MLP) and the most nearby neighbor classifier (KNN). Results show that not complex algorithms are required in the classification phase to obtain high recognition values. In our simulations, rates beyond 99% have been achieved.