A Comparative Study of Hand Recognition Systems

Hand-based recognition represents a key biometric technology with a wide range of potential applications both in industry and government. By far, many different hand-based recognition algorithms have been developed. This paper presents a comparative study to evaluate the performance of three state of the art hand-based recognition methods. Using the University of Nevada at Reno (UNR) and the University of Notre Dame (UND) hand databases, we compare a geometric-based method, a component-based approach using Zernike moments, and an algorithm employing 3D finger surface features. Both recognition and authentication experiments have been conducted to investigate the performance and robustness of the three methods. Our experimental results show that Zernike descriptors yield features that are more robust and accurate compared to hand geometric features and 3D finger surface features.