Fusion of multi-reprsentation palmprint image for automatic personnel identification

The single biometric system may be inadequate in many applications such as public security, access control and banking because of noise in data samples or because of unavailability of a sample at a given time. In order to overcome the limitation of the single biometric, a Multi-representation biometric is used. In this paper we present a multimodal personal identification system using the Histogram of Oriented Gradients (HOG) descriptors as feature extraction. Then two classifiers are used distance classifier and RBF classier to design two subsystems. The experimental results showed that the designed system achieves an excellent identification rate and provides more security than uni-modal biometric system.