To read this content please select one of the options below:

RFID-transformer recognition system (RTRS): enhancing privacy in facial recognition with transformer models

Zeyuan Wang (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)
He Xu (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)
Manman Zhang (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)
Zhaorui Cai (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)
Yongyuan Chen (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)

Sensor Review

ISSN: 0260-2288

Article publication date: 27 September 2024

24

Abstract

Purpose

This paper aims to present a novel approach to facial recognition that enhances privacy by using radio frequency identification (RFID) technology combined with transformer models, eliminating the need for visual data and thus reducing privacy risks associated with traditional image-based systems.

Design/methodology/approach

The proposed RFID-transformer recognition system (RTRS) uses RFID technology to capture signal features such as phase and received signal strength indicator, which are then processed by a transformer model. The model is specifically designed to handle structured RFID data, capturing subtle patterns and dependencies to achieve accurate biometric recognition. The system’s performance was validated through comprehensive experiments involving different environmental conditions and user scenarios.

Findings

The experimental results demonstrate that the RTRS system achieves a recognition accuracy of 98.91%, maintaining robust performance across various challenging conditions, including low-light environments and changes in face orientation. In addition, the system provides a high level of privacy preservation by avoiding the collection and storage of visual data.

Originality/value

To the best of the authors’ knowledge, this work introduces the first RFID-based facial recognition system that fully leverages transformer models, offering a privacy-preserving alternative to traditional image-based methods. The system’s ability to perform accurately in diverse scenarios while ensuring user privacy makes it a significant advancement in biometric technology.

Keywords

Citation

Wang, Z., Xu, H., Zhang, M., Cai, Z. and Chen, Y. (2024), "RFID-transformer recognition system (RTRS): enhancing privacy in facial recognition with transformer models", Sensor Review, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/SR-07-2024-0659

Publisher

:

Emerald Publishing Limited

Copyright © 2024, Emerald Publishing Limited

Related articles