A New Palmprint Identification Technique Based on a Two–Stage Neural Network Classifier

Palmprint is one of the relatively new physiological biometrics due to its stable and unique characteristics. The rich texture information of palmprint offers one of the powerful means in the field of personal recognition. The proposed system is based on geometrical features and texture features extracted using kernel principal components analysis (K-PCA). In the coarse-level stage, the hand geometrical features are applied in the SOMNN to select a small set for further matching, and in the fine-level matching, texture features are input into the BPNN for final identification. The experimental results show the effectiveness and reliability of the proposed approach.