Advanced Finger Vein Authentication: Detection and Matching Techniques for Enhanced Biometric Security

Abstract

Finger vein authentication is a secure biometric recognition method leveraging the unique and internal structure of finger veins. This paper explores advancements in detection and matching techniques to enhance accuracy and security. Near-infrared (NIR) imaging, preprocessing methods, and deep learning-based segmentation improve vein pattern extraction. Matching processes utilize traditional distance metrics and machine learning models like CNNs and Siamese networks for robust authentication. Challenges such as image quality, spoofing threats, and computational efficiency are addressed through AI-driven solutions and multi-modal biometric integration. Experimental analysis using public datasets demonstrates improved performance with deep learning models. Future directions include edge AI, blockchain identity verification, and hybrid biometrics to strengthen security. This study provides insights into optimizing finger vein authentication for secure identity verification systems.