Multi-instance Finger Vein-based Authentication with Secured Templates

The illegitimate access to biometric templates is one of the major issues be handled for authentication systems. In this work, we propose use two instances finger vein images which inherits advantages a robust multi-modal system without needing different sensors. Two local texture feature extraction methods are experimented on standard finger-vein datasets. Fused discriminating features with reduced dimension lowers down computational cost. A cancelable template protection scheme as Gaussian Random Projection based Index-of-Max then applied embedding privacy and security templates. Foremost properties like revocability, non-invertibility unlinkability observed significantly obeyed by proposed considerable performance. Recognition performance compared some previously executed systems less complex overperforming combined basis protection. Thus, utilizes multiple evidence provides balanced respect authentication,