A Method for Capturing the Finger-Vein Image Using Nonuniform Intensity Infrared Light

In the finger vein authentication, one problem is that the variable finger brightness makes the veins unclear. For the same reason, it is difficult to get the intact finger-vein pattern which contains more features. In this paper, we developed an experimental rule to adjust the intensity of the light, which can solve the problem mentioned. We propose three criterions, the standard deviation of the gray level, total length of the vein and the number of bifurcations, to judge the quality of the image. In our experiment, our method makes the brightness of the finger more uniform, thus the standard deviation of the gray level decrease by 48.4% on average. The numbers of two features, the veins and its bifurcations, increase by 44.1% and 31.4% respectively. Using our method, we have built a finger vein image database which has 600 images.