An improved hand vein image acquisition method based on the proposed image quality evaluation system

Jun Wang, Guoqing Wang, Ming Li, Wenhui Yu, HaoTian

School of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou, 221116, China

In the part of hand vein image acquisition of vein recognition, we basically cannot get the images that are suitable for later process due to both the subjective factors and the objective factors. After a deep analysis on the main reasons including the different thickness of hands and mass of things on the hand, we propose the design of hand vein acquisition device based on the adaptive modulation control. The key part of the design is the proposed double-processing image quality assessment system: the first part of which involves the process of quality assessment based on the effective information of multidimensional histogram and the normalized information entropy; the second part involves adding the grey image variance and quantities of cross points with the proper weights, then two process will be made according to the final score: if the score is lower than the set threshold then image will be discarded, and on the other hand the brightness degree of the LED groups of the device will be adaptively adjusted according to the D-value of the score and the threshold so that we can get the high-quality hand vein image for later feature extraction and matching and get a higher recognition rate.