BioDeepHash: Mapping Biometrics into a Stable Code

TL;DR

BioDeepHash introduces a novel biometric template protection framework that unifies deep hashing and cryptographic hashing to map multiple biometric samples from the same user to a stable code without relying on error-correcting codes. A deep fuzzy hashing (DFH) model with Gaussian-classwise loss, regression loss, and quantization loss generates robust, binary codes that are then transformed by an application-specific XOR string into cancelable templates, finally hashed with SHA3-512 for strong security. The system achieves high genuine acceptance rates and exceptionally low false acceptance rates on iris and facial datasets, while satisfying irreversibility, revocability, and unlinkability under a full-disclosure threat model. By avoiding helper data and ECC, BioDeepHash reduces privacy risks and enables revocation via XOR string updates, offering a practical, secure approach for biometric template protection with strong real-world implications for biometric security and privacy.

Abstract

With the wide application of biometrics, more and more attention has been paid to the security of biometric templates. However most of existing biometric template protection (BTP) methods have some security problems, e.g. the problem that protected templates leak part of the original biometric data (exists in Cancelable Biometrics (CB)), the use of error-correcting codes (ECC) leads to decodable attack, statistical attack (exists in Biometric Cryptosystems (BCS)), the inability to achieve revocability (exists in methods using Neural Network (NN) to learn pre-defined templates), the inability to use cryptographic hash to guarantee strong security (exists in CB and methods using NN to learn latent templates). In this paper, we propose a framework called BioDeepHash based on deep hashing and cryptographic hashing to address the above four problems, where different biometric data of the same user are mapped to a stable code using deep hashing instead of predefined binary codes thus avoiding the use of ECC. An application-specific binary string is employed to achieve revocability. Then cryptographic hashing is used to get the final protected template to ensure strong security. Ultimately our framework achieves not storing any data that would leak part of the original biometric data. We also conduct extensive experiments on facial and iris datasets. Our method achieves an improvement of 10.12$\%$ on the average Genuine Acceptance Rate (GAR) for iris data and 3.12$\%$ for facial data compared to existing methods. In addition, BioDeepHash achieves extremely low False Acceptance Rate (FAR), i.e. 0$\%$ FAR on the iris dataset and the highest FAR on the facial dataset is only 0.0002$\%$.

Figures (6)

• Figure 1: Research Motivation
• Figure 2: An illustration of the proposed framework
• Figure 3: Model Architecture
• Figure 4: Unlinkability of BioDeephash on CASIA-IrisV4-Lamp
• Figure 5: GAR and FAR ($\%$) for different $L$ on Lamp and Thousand