Privacy-preserving method for face recognition based on homomorphic encryption
PLOS ONE · 11.02.2025 · 20 (2) · e0314656
DOI: 10.1371/journal.pone.0314656 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 11.02.2025. Verfasst von Zhigang Song und 7 weiteren Autor:innen. Beteiligte Einrichtung: The Academy of Digital China, Fuzhou, Fujian, China. Hauptthemen laut MEDLINE: Computer Security, Automated Facial Recognition, Privacy, Facial Recognition. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: REGIONAL DEVELOPMENT PROJECT OF FUJIAN PROVINCIAL DEPARTMENT OF SCIENCE AND TECHNOLOGY.
Abstract (Originalsprache)
In recent years, facial recognition technology has been widely adopted in modern society. However, the plaintext storage, computation, and transmission of facial data have posed significant risks of information leakage. To address this issue, this paper proposes a facial recognition framework based on approximate homomorphic encryption (HE_FaceNet), aimed at effectively mitigating privacy leaks during the facial recognition process. The framework first utilizes a pre-trained model to extract facial feature templates, which are then encrypted. The encrypted templates are matched using Euclidean distance, with the final recognition being performed after decryption. However, the time-consuming nature of homomorphic encryption calculations limits the practical applicability of the HE_FaceNet framework. To overcome this limitation, this paper introduces an optimization scheme based on clustering algorithms to accelerate the facial recognition process within the HE_FaceNet framework. By grouping similar faces into clusters through clustering analysis, the efficiency of searching encrypted feature values is significantly improved. Performance analysis indicates that the HE_FaceNet framework successfully protects facial data privacy while maintaining high recognition accuracy, and the optimization scheme demonstrates high accuracy and significant computational efficiency across facial datasets of varying sizes.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- The Academy of Digital China, Fuzhou, Fujian, China.
- College of Computer and Data Science, Fuzhou University, Fuzhou, Fujian, China.
- School of Business, East China University of Science and Technology, Shanghai, China.
- School of Computing and Information Systems, Singapore Management University, Singapore.
ORCID-Profile: Zeli Wang
Förderung
- REGIONAL DEVELOPMENT PROJECT OF FUJIAN PROVINCIAL DEPARTMENT OF SCIENCE AND TECHNOLOGY
Zitieren
APA
Song, Z., Wang, G., Yang, W., Li, Y., Yu, Y., Wang, Z., Zheng, X., & Yang, Y. (2025). Privacy-preserving method for face recognition based on homomorphic encryption. PLOS ONE, 20(2), e0314656. https://doi.org/10.1371/journal.pone.0314656
Vancouver
Song Z, Wang G, Yang W, Li Y, Yu Y, Wang Z, et al. Privacy-preserving method for face recognition based on homomorphic encryption. PLOS ONE. 2025;20(2):e0314656. doi:10.1371/journal.pone.0314656
BibTeX
@article{Song2025,
author = {Song, Zhigang and Wang, Gong and Yang, Wenqin and Li, Yunliang and Yu, Yinsheng and Wang, Zeli and Zheng, Xianghan and Yang, Yang},
title = {Privacy-preserving method for face recognition based on homomorphic encryption},
journal = {PLOS ONE},
year = {2025},
volume = {20},
number = {2},
pages = {e0314656},
doi = {10.1371/journal.pone.0314656}
}
EUVIMED