SEMANTIC SEGMENTATION OF FACIAL IMAGES IN BIOMETRIC AUTHENTICATION SYSTEMS OF PERSONNEL OF CRITICAL INFRASTRUCTURE FACILITIES
DOI:
https://doi.org/10.28925/2663-4023.2025.28.816Keywords:
neural network model, semantic segmentation, information protection, critical infrastructure facility, information security, biometric authentication, person recognitionAbstract
The problem of the article is to increase the efficiency of biometric authentication of personnel of critical infrastructure facilities. It is shown that one of the main directions of increasing efficiency is to improve the procedure for highlighting facial contours in the test image, the result of which in most known cases is the definition of a rectangular area covering the face. Such a result does not provide accurate highlighting of facial contours and interference during video recording, in particular personal protective equipment, hair and glasses. To overcome these limitations, it is advisable to use neural network semantic segmentation tools, which allow you to accurately highlight facial contours, the eye area, as well as areas with overlaps or background elements, which significantly increases the accuracy of face recognition in biometric systems. At the same time, the results of the analysis of modern scientific and practical solutions in the field of semantic segmentation show that most of them do not provide the possibility of effective functioning in the conditions of critical infrastructure facilities, which is primarily explained by the imperfection of methodological support. In order to overcome the above-mentioned shortcomings, the article proposes a model of semantic segmentation of facial images, which is based on an encoder-decoder neural network architecture with the ability to adapt design parameters to the conditions of application on critical infrastructure objects. Based on this model, a method for determining the architectural parameters of a neural network model has been developed, which involves a sequential assessment of the task conditions, selection of the basic architecture, adjustment of the encoder, decoder and training parameters, evaluation of efficiency and adaptive modification of the model structure. The method allows taking into account the influence of a number of factors, in particular, the spatial characteristics of the segmentation elements, class imbalance, lighting variability, limitations on computing resources and regulatory requirements. Experimental studies have shown that the use of the proposed method allows reducing the volume of necessary experiments by 2 times and achieving facial image segmentation accuracy at the level of 0.9, which exceeds the indicators of existing analogues by approximately 10–20%.
Downloads
References
Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017). Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE transactions on pattern analysis and machine intelligence, 39(12), 2481–2495. https://doi.org/10.1109/TPAMI.2016.2644615
Bazarevsky, V., Kartynnik, Y., Vakunov, A., Raveendran, K., & Grundmann, M. (2019). Blazeface: Sub-millisecond neural face detection on mobile gpus. CVPR Workshop on Computer Vision for Augmented and Virtual Reality. https://doi.org/10.48550/arxiv.1907.05047
Chen, L. C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2017). Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4), 834–848. https://doi.org/10.1109/TPAMI.2017.2699184
Deng, J., Guo, J., Ververas, E., Kotsia, I., & Zafeiriou, S. (2020). Retinaface: Single-shot multi-level face localisation in the wild. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 5203–5212.
Lin, G., Milan, A., Shen, C., & Reid, I. (2017). Refinenet: Multi-path refinement networks for high-resolution semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, 1925–1934. https://doi.org/10.48550/arXiv.1611.06612
Liu, S., Shi, J., Liang, J., & Yang, M. H. (2017). Face parsing via recurrent propagation. BMVC. https://doi.org/10.48550/arXiv.1708.01936
Ranjan, R., Bansal, A., Zheng, J., Xu, H., Gleason, J., Lu, B., Castillo, C. & Chellappa, R. (2019). A Fast and Accurate System for Face Detection, Identification, and Verification. In IEEE Transactions on Biometrics, Behavior, and Identity Science, 1(2), 82–96. https://doi.org/10.1109/TBIOM.2019.2908436
Schroff, F., Kalenichenko, D. & Philbin, J. (2015). FaceNet: A unified embedding for face recognition and clustering. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 815–823. https://doi.org/10.1109/CVPR.2015.7298682
Tereikovskyi, I., Hu, Z., Chernyshev, D., Tereikovska, L., Korystin, O., & Tereikovskyi, O. (2022). The method of semantic image segmentation using neural networks. International Journal of Image, Graphics and Signal Processing, 10(6), 1, 14(6), 1–14. https://doi.org/10.5815/ijigsp.2022.06.01
Tereikovskyi, I., Korchenko, O., Bushuyev, S., Tereikovskyi, O., Ziubina, R. & Veselska, O. (2023). A Neural Network Model for Object Mask Detection in Medical Images. International Journal of Electronics and Telecommunication, 69(1), 41–46. https://doi.org/10.24425/ijet.2023.144329
Zhang, K., Zhang, Z., Li, Z., & Qiao, Y. (2016). Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks. IEEE signal processing letters, 23(10), 1499–1503. https://doi.org/10.1109/LSP.2016.2603342
Zhao, H., Shi, J., Qi, X., Wang, X., & Jia, J. (2017). Pyramid scene parsing network. In Proceedings of the IEEE conference on computer vision and pattern recognition, 6230–6239. https://doi.org/10.1109/CVPR.2017.660
Korchenko O., & Tereikovskyi O. (2023). Analysis and evaluation of biometric authentication means based on the image of the face and iris of the staff of critical infrastructure facilities. Cybersecurity: Education, Science, Technique. 1(21), 136–148. https://doi.org/10.28925/2663-4023.2023.21.136148
Korchenko, O., Tereikovskyi, O. (2024). Modular neural network model for biometric authentication of the personnel of critical infrastructure facilities using facial images and iris. Information security, 30(2), 339–347. https://doi.org/10.18372/2225-5036.30.19247
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Олександр Корченко, Олег Терейковський
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
