Enhancing Biometric Authentication through Score-Level Fusion of Gait and Palm Vein Modalities

Biometric authentication is evolving itself by adapting into new technology, eliminating the traditional authentication techniques such like fingerprints, Iris Recognition, Retina Scanning and facial recognition by gait and palm vein recognition. This paper aims to expand around the combination of gait and palm vein recognition which collectively offer additional accuracy and safe results. The Gait recognition system uses the unique walking pattern of an individual for recognition. Using infrared light, the vein patterns beneath the skin are recognized by the palm vein identification system. Combination of both will overcome the limitations of the conventional biometric recognition technology, such as the susceptibility to spoofing and environmental variability. The proposed model employs Fisher's Linear Discriminant Analysis (FLDA) for classification and Principal Component Analysis (PCA) for feature extraction in gait recognition, alongside preprocessing, feature extraction, and matching in palm vein recognition. Recognition performance is improved by normalization and score fusion in the integrated biometric system.