Analysis of methods for detecting attacks on facial biometric authentication in mobile devices
DOI:
https://doi.org/10.30837/rt.2023.4.215.02Keywords:
attack, biometric authentication, protection, camera, face, neural network, Android, Deepfake, GAN, PyTorch Mobile, TensorFlow LiteAbstract
The object of the research is facial biometric authentication in the context of mobile devices (financial and government applications, etc.) of the Android and IOS platforms. The subject of the research is methods of detecting attacks on biometric authentication by face in mobile devices. The purpose of the work is to investigate the methods of creating and recognizing Deepfake videos, to evaluate the possibility of local recognition of fakes within applications in mobile devices of the Android and IOS platforms.
As a result, the vulnerability of biometric authentication to Deepfake technology was confirmed, methods for recognizing fakes were analyzed, and the effectiveness of using a neural network model for local recognition of Deepfake in a mobile device was experimentally confirmed. Conclusions are drawn about possible ways to optimize the size of the model with an emphasis on maintaining accuracy.
References
Мироненко Є.В., Сєвєрінов О.В. Біометрична ідентифікація і автентифікація особи за геометрією об-личчя. Харків : НТУ «ХПІ», 2020.
Passport-Photo-Biometric-Statistics [Електронний ресурс]. Режим доступу: https://passport-photo.online/blog/biometric-statistics/#gref
Creswell, Antonia; White, Tom; Dumoulin, Vincent; Arulkumaran, Kai; Sengupta, Biswa; Bharath, Anil A. (2018). Generative Adversarial Networks: An Overview // IEEE Signal Processing Magazine, 35(1), 53–65. DOI: 10.1109/MSP.2017.2765202
Hady A. Khalil;Shady A. Maged; (2021). Deepfakes Creation and Detection Using Deep Learning // International Mobile, Intelligent, and Ubiquitous Computing Conference. 2021. DOI: 10.1109/MIUCC52538.2021.9447642
Zainab Zahid, Ammar Haider, Nosheen Sabahat, Asim Tanwir. Vulnerabilities in Biometric Authentication of Smartphones // 2020 IEEE 23rd International Multitopic Conference (INMIC). 2020. DOI: 10.1109/inmic50486.2020.9318094
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).