Combining fingerprint, palmprint and hand-shape for user authentication

This paper investigates a new approach for personal authentication by combining unique biometric features which can be acquired from hand images alone. The proposed method attempts to improve the performance of fingerprint-based verification system by integrating palmprint and hand-shape features. The matching scores for the fingerprint images are computed using the number of matched minutiae on the overlapping areas while those for palmprint and hand-shape images are based on distance of feature vectors. These matching scores are combined using simple fusion rule which does not require any training. Our experimental results on the database of 100 users achieve promising results and therefore confirm the usefulness of proposed method.