A Secure Biometric Key Generation Mechanism via Deep Learning and Its Application

Biometric keys are widely used in the digital identity system due to inherent uniqueness of biometrics. However, existing biometric key generation methods may expose data, which will cause users’ traits be permanently unavailable secure authentication system. To enhance its security and privacy, we propose a method based on deep learning this paper. Firstly, prevent information leakage utilize random binary codes represent data adopt model establish relationship between code for each user. Secondly, protect privacy guarantee revocability key, add permutation operation shuffle elements update new key. Thirdly, further reliability construct fuzzy commitment module generate helper without revealing any during enrollment. Three benchmark datasets including ORL, Extended YaleB, CMU-PIE evaluation. The experiment results show our scheme achieves genuine accept rate (GAR) higher than state-of-the-art at 1% false (FAR), meanwhile satisfies properties randomness keys. analyses that can effectively resist leakage, cross-matching, other attacks. Moreover, proposed is applied encryption scenario local computer, takes less 0.5 s complete whole decryption different lengths.