Abstract :

With The Rapid Expansion Of Large-scale Biometric Identification Systems, Ensuring Both Privacy Protection And Computational Efficiency Has Become A Critical Challenge. Conventional Biometric Indexing Methods Often Rely On Exhaustive Search Strategies Or Unprotected Feature Representations, Leading To High Computational Overhead And Increased Privacy Risks. To Address These Limitations, This Work Proposes A Novel Privacy-preserving Multibiometric Identification Framework That Integrates Secure Template Protection With An Efficient Indexing Mechanism. The Proposed System Leverages Frequent Binary Pattern Analysis On Protected Biometric Templates To Enable Effective Workload Reduction While Maintaining High Recognition Accuracy. Unlike Traditional Approaches That Are Tailored To A Single Biometric Modality, The Proposed Mechanism Is Designed To Be Modality-agnostic And Supports The Fusion Of Multiple Biometric Traits Such As Face, Fingerprint, And Iris At Both Feature And Representation Levels. Secure Cancellable Transformations Are Applied To Biometric Embedding’s, Ensuring Irreversibility, Unlink Ability, And Renewability Of Biometric Data. Experimental Analysis Demonstrates That The Proposed Indexing Strategy Significantly Reduces The Number Of Required Template Comparisons Compared To Exhaustive Search Methods, While Simultaneously Improving Identification Performance At High-security Operating Thresholds. The Results Confirm That The Proposed Mechanism Achieves An Effective Trade-off Between Privacy Preservation, Computational Efficiency, And Biometric Accuracy, Making It Suitable For Deployment In Large-scale And Security-critical Biometric Systems.

Published:

20-5-2026

Issue:

Vol. 26 No. 5 (2026)

Page Nos:

1274-1279

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite