Volume : 5, Issue : 6, June - 2016

A Convolutional Neural Network Method for Palm Vein Recognition

You Jinghan, Liang Zuoyu, Cui Lunan

Abstract :

<p>&lt;p&gt;&amp;nbsp;In recent years, identity authentication by using biometric features has received more and more attention. Unlike the traditional finger print recognition, the blood vein image lies under the skin and is difficult to forge. Therefore, identity authentication by employing blood vein image is with higher security than using finger print. In this study, we have proposed a method, which is based on convolutional neural network and aims at recognizing the identity of certain palm vein images. The convolutional neural network was proved to achieve good performance in image recognition tasks. Comprehensive experiments have demonstrated that the proposed method achieves high success rates in identity authentication and should have a wide range of applications&lt;/p&gt;</p>