Privacy-Preserving Iris Recognition: Performance Challenges and Outlook

TL;DR

This paper examines the performance challenges of privacy-preserving iris recognition using Fully Homomorphic Encryption (FHE) and proposes a scalable framework that maintains accuracy but incurs high computational overhead.

Contribution

It introduces a scalable privacy-preserving iris recognition framework based on FHE that aligns with ISO standards and evaluates its performance on a large dataset.

Findings

The framework achieves accuracy similar to non-encrypted iris recognition.

It incurs approximately 120,000 times higher computational cost for pairwise comparisons.

Two-level schemes are necessary for scalable 1-N template comparisons.

Abstract

Iris-based biometric identification is increasingly recognized for its significant accuracy and long-term stability compared to other biometric modalities such as fingerprints or facial features. However, all biometric modalities are highly sensitive data that raise serious privacy and security concerns, particularly in decentralized and untrusted environments. While Fully Homomorphic Encryption (FHE) has emerged as a promising solution for protecting sensitive data during computation, existing privacy-preserving iris recognition systems face significant performance limitations that hinder their practical deployment. This paper investigates the performance challenges of the current landscape of privacy-preserving iris recognition systems using FHE. Based on these insights, we outline a scalable privacy-preserving framework that aligns with all the requirements specified in the ISO/IEC 24745 standard. Leveraging the Open Iris library, our approach starts with robust iris segmentation, followed by normalization and feature extraction using Gabor filters to generate iris codes. We then apply binary masking to filter out unreliable regions and perform matching using Hamming distance on encrypted iris codes. The accuracy and performance of our proposed privacy-preserving framework is evaluated on the CASIA-Iris-Thousand dataset. Results show that our privacy-preserving framework yields very similar accuracy to the cleartext equivalent, but a much higher computational overhead with respect to pairwise iris template comparisons, of $\sim 120\,000 \times$. This points towards the need for the deployment of two-level schemes in the context of scalable $1-N$ template comparisons.