Security and Privacy in Biometric Authentication: Advancements and Risks Across Platforms

Title

Security and Privacy in Biometric Authentication: Advancements and Risks Across Platforms

Creator

Benny, Sandra; Das, Deepthi; Rajesh Kanna, R.

Description

Biometric identification systems like fingerprint, face, iris, and voice recognition have revolutionized user authentication with more convenience and increased security over conventional password based authentication. Nonetheless, the fact that biometric features are immutable introduces unprecedented issues around privacy, spoofing, and database compromise because once biometric data leaks out, it cannot be withdrawn or substituted. This work advances empirical testing with analytical analysis to test biometric authentication for face, fingerprint, and iris modalities. Models based on convolutional neural networks (CNNs) were used and evaluated on benchmarked datasets, and findings indicated that face and iris recognition performed nearly perfectly with zero false acceptance and rejection rates in controlled environments, while fingerprint recognition performed poorly because of dataset size and quality constraints. The conclusions identify the significance of data preparation and variation in ascertaining the reliability of biometric information. In an effort to cross-verify the experiments further, case studies of Aadhaar, Apple Face ID, and Biostar 2 also clarified the threats of central storage, spoofing, and regulatory loopholes. The research concludes by suggesting privacy preserving frameworks, encryption, and multimodal methods for securing the future of biometric authentication. 2025 IEEE.

Source

Proceedings - 2025 International Conference on Transformative Computing Technologies, ICTCT 2025;pp.254-259

Date

01-01-2025

Publisher

Institute of Electrical and Electronics Engineers Inc.

Subject

Biometric authentication; Face recognition; Fingerprint recognition; Iris recognition; Privacy and security

Coverage

Benny S., Christ University Computer Science, Bengaluru, India; Das D., Christ University Statistics and Data Science, Bengaluru, India; Rajesh Kanna R., Christ University Computer Science, Bengaluru, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISBN: 979-833159195-3;

Format

online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1109/ICTCT69201.2025.00056

https://www.scopus.com/pages/publications/105037440350?origin=resultslist