Near infrared hand vein image acquisition and ROI extraction algorithm

In this paper, a hand vein acquisition system is devised, and a new coordinate localization-based algorithm for extracting the region of interest (ROI) of hand vein images is proposed for this system. This algorithm uses the low and right boundaries of hands as the y-axis and x-axis, respectively, to build a coordinate system to determine the position of the ROI of hand vein images. The acquisition system gets the hand vein images with low resolution and the proposed algorithm can perform very well for low-resolution images. We use the designed acquisition device to collect a hand vein database. This database and another database are used to conduct ROI extraction and hand vein recognition experiments. The experimental results show that the ROI images can be successfully and effectively extracted from different hand vein images, and the proposed algorithm can achieve high recognition accuracies.