Phase Congruency Corner Detection as Biometric Feature

Palmprint is one of the most reliable physiological biometric due to its unique and permanent features like geometry features, principal line features, wrinkles, ridges, delta point features, texture features and points. Here, Phase Congruency Corner Detector method is proposed as a corner detection technique to extract palmprint features in the form of corners. The palmprint features extracted are binarized and then matched using Hamming distance similarity measurement with sliding window method. By using hamming distance method for corner matching, the comparison time is non-dependent on the number of corners detected. So, the comparison (matching) time will be constant irrespective of the number of corners. Experiments were developed on a database of 600 images from 100 individuals, with five image samples per individual for training and one image sample per individual for testing. The proposed approach is applied on PolyU database and the experimental results demonstrate that the proposed approach has good discriminating effect.