The CFVD reflection-type finger-vein image database with evaluation baseline

Abstract

In this paper, we describe the reflection-type finger-vein image database named by CFVD for biometrics research, including its acquisition, contents and evaluation baseline. The main contributions of this work include the following points: providing the worldwide researchers with reflection-type finger-vein recognition uniform database and ground-truth evaluation baseline. Currently, the CFVD database contains 1345 images of 130 fingers from 13 individuals ( 10 males and 3 females ). Based on this database, the researchers can evaluate the performance of their algorithms for reflection-type finger-vein recognition. As the first reflection-type finger-vein image database available in the public domain, we believe that it will promote the development of finger-vein recognition techniques. We are keeping enlarging this database by including the finger-vein data of additional people. © Springer International Publishing 2013.