Multimodal biometric person recognition system based on fingerprint & Finger-Knuckle-Print using correlation filter classifier

Biometrics is an effective technology for personnel identity recognition, but uni-modal biometric systems which use a single trait for recognition will suffer from problems like noisy sensor data, non-universality, lack of distinctiveness of the biometric trait, and spoof attacks. These problems can be tackled by using multi-biometrics in the system. Hand-based person recognition provides a reliable, low-cost and user-friendly viable solution for a range of access control applications. As one of the most popular biometric traits, fingerprints (FP) are widely used in personal recognition. However, a novel hand-based biometric feature, Finger-Knuckle-Print (FKP), has attracted an increasing amount of attention. In this paper, FP and FKP are integrated in order to construct an efficient multi-biometric recognition system based on matching score level and image level fusion. In this study we use the minimum average correlation energy (MACE) and Unconstrained MACE (UMACE) filters in conjunction with two correlation plane performance measures, max peak value and peak-to-sidelobe ratio, to determine the effectiveness of this method. The experimental results showed that the designed system achieves an excellent recognition rate on the Hong Kong polytechnic university (PolyU) FKP and high resolution fingerprint database.