Implementation of Fixed-Length Template Protection Based on Homomorphic Encryption With Application to Signature Biometrics

Marta Gomez-Barrero, Julian Fierrez, Javier Galbally, Emanuele Maiorana, Patrizio Campisi; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2016, pp. 191-198

Given the high sensitivity of biometric data, any information leakage poses severe security and privacy risks. This fact raises the need to protect biometric templates so that no information can be learned from them, preserving at the same time the unprotected system's performance and speed. We propose a new efficient biometric template protection scheme based on homomorphic probabilistic encryption for fixed-length templates, where only encrypted data is handled. Under a fully reproducible experimental framework, we analyse different distance measures for the particular case of on-line signature, showing that all requirements described in the ISO/IEC 24745 standard on biometric information protection are met with no performance degradation and at a low computational cost. Furthermore, the proposed approach is robust to hill-climbing and inverse-biometrics attacks.