Feature-level fusion for effective palmprint authentication

A feature-level fusion approach is proposed for improving the efficiency of palmprint identification. Multiple Gabor filters are employed to extract the phase information on a palmprint image, which is then merged according to a fusion rule to produce a single feature called the Fusion Code. The similarity of two Fusion Codes is measured by their normalized hamming distance. A database containing 7,752 palmprint images from 386 different palms is used to validate the performance of the proposed method. Empirically comparing our previous non-fusion approach and the proposed method, improvement in verification is ensured.