Finger Vein Pattern-based Authentication using Geometrical Properties of Corner Points

Abstract

Securing an organization with proper authentication is a challenge. Intruders are always active and pose thread to anyone. And, attaining cost-effective system with high accuracy is a management perspective. Hence, in this paper we propose a low-cost reliable biometric system based on finger vein patterns. The proposed algorithm initially captures the finger vein image and is preprocessed to remove possible noise and distortion. Geometrical features like corner points and their location are extracted. These features will play the role in authenticating any individual. The simulation results of the proposed algorithm have shown the FAR as 4.72%, FRR as 1.9% and the overall performance as 98.11%.