FPGA embedded hardware system for finger vein biometric recognition

Vein biometrics is emerging and gaining popularity over other types of biometric systems due to its strengths of low forgery risk, aliveness detection, non-invasive data acquisition as well as stable over long period. This paper presents a novel design of an embedded finger vein biometric recognition hardware system targeted for implementation on a field-programmable gate array (FPGA) platform. A novel streaming architecture for hardware acceleration of window-based image processing application is proposed. To ensure high quality image capture and high recognition accuracy, we introduce an image acquisition subsystem that uses an embedded camera with dynamic illumination based on quality assessment. Experimental results show that the proposed finger vein verification system achieves 0.87% equal error rate (EER) on a database of 500 finger vein images.