Abstract: Biometric System has been dynamically materializing in various industries for the past few years and it is ongoing to roll to grant higher security features for access control system. Several types of unimodal biometric systems have been expanded. Though, these systems are only able to offer low to middle range of security aspect. Therefore, for higher security aspect, the fusion of two or more unimodal biometrics (multiple modalities) is needed. This study proposes a multimodal Biometric System for person recognition using Multi-Spectral Hand Image Acquisition and Fusion of Finger knuckle scores to Hand Geometry and PalmPrint scores. PalmPrint, Hand Geometry and Finger-Knuckle-Print (FKP) textures are concurrently obtained using Multi-Spectral Hand Image Acquisition from the users facade standardized textured 3D hand for recognizing the similarity between the hand postures. The features of FKP are mined using the scale invariant feature transform (SIQT) and the speeded up robust features (SSF). The three modalities are combined and the fusion is applied at the matching-score level by using weighted sum rule method. The experimental results illustrated that the proposed system improves the metrics of similarity scores in terms of rate and time of similarity verification, minimize size of hand data features and reduction of templates in multimodality matching.
A. Kirthika and S. Arumugam, 2013. Fusion of Finger Knuckle Scores to Hand Geometry and Palm Print Scores using Multi-Spectral Hand Image Acquisition. Asian Journal of Information Technology, 12: 312-317.