Selective Cancellable Multi-Biometric Template Generation Scheme Based on Multi-Exposure Feature Fusion

This article introduces a new cancellable multi-biometric system based on the combination of selective encryption method and deep-learning-based fusion technology. The biometric face image is treated with an automatic segmentation algorithm (Viola-Jones), selected eye XORed PRNG (Pseudo Random Number Generator) matrix. output array used to create primary template. process changes histogram image. Arnold’s Cat Map superimpose PRN pixels only cat map deformed eyes are encrypted using Advanced Encryption Standard (AES) encrypt data stored in database. In addition, AES master key for same person identity verification verify identity. It created from fingers right hand, integrated into this deep learning can prevent attacks as whole. order avoid damage or fingerprint images, design considers other images.