Focussing the Beam - A New Laser Illumination Based Data Set Providing Insights to Finger-Vein Recognition

摘要

The vascular pattern inside human fingers has become an emerging biometric trait during the last years, commonly denoted as finger vein recognition. However, the number of publicly available data sets is limited. In order to capture a finger vein data set, a suitable scanner device is needed. The design of such a scanner device is crucial if it comes to image quality, robustness against external influences during the capturing process and consequently to a good recognition performance. In this paper we propose two novel, modular designed, multi-purpose finger vein scanners, both able to capture three fingers at once, together with a publicly available finger-vein data set captured with these scanners. One scanner uses common near-infrared LEDs as a light source. The second one is based on a new concept using near-infrared lasers. Near-infrared lasers are not common in finger-vein recognition before despite their advantages especially in touchless operation. Our recognition performance evaluation confirm the good recognition performance that can be achieved using our proposed scanner design and provides some new insights by conducting sex and age-group specific analysis.