Real-Time Intrusion Detection in Cyber-Physical Systems Using YOLO v11 and Facial Recognition

Main Article Content

Boonsong Srichand
Nithizethe Mhuadthongon

Abstract

This research proposes the development of a Cyber-Physical System (CPS) for real-time monitoring and intrusion detection on edge computing architectures to mitigate high false alarm rates (FAR) of traditional systems while addressing VRAM constraints of edge processing hardware and supporting efficient data transmission using event-based communication methods. The prototype combines YOLO11, DeepSORT, and InsightFace models built under the 5C CPS architecture framework. To overcome VRAM constraints, a shared resource architecture with a global GPU lock is proposed. Additionally, mechanisms for identity stabilization and temporal reasoning were implemented to ensure reliable decision-making under changing conditions. The shared resource architecture used 71.23% less VRAM when processing six camera streams at once. This results in lower resource consumption for this type of application. The overall accuracy of the system is 96.25%, and the temporal voting mechanism improves identification accuracy with occlusion by 30.5%. In addition, no false-positive event was observed in the controlled access-control evaluation, corresponding to a 0.00% false-alarm rate under the evaluated conditions. This study shows that efficient hardware resource allocation allows the implementation of complex AI models without model compression. At the same time, context-aware decision-making mechanisms provide dependability in intelligent systems in CPS applications.

Article Details

How to Cite
[1]
B. Srichand and N. Mhuadthongon, “Real-Time Intrusion Detection in Cyber-Physical Systems Using YOLO v11 and Facial Recognition”, ECTI-CIT Transactions, vol. 20, no. 4, pp. 798–812, Oct. 2026.
Section
Research Article

References

[1] W. Ullah, A. Ullah, I. U. Haq, K. Muhammad, M. Sajjad and S. W. Baik, “CNN features with bi-directional LSTM for real-time anomaly detection in surveillance networks,” Multimedia Tools and Applications, vol. 80, pp. 16979–16995, 2021.

[2] C. Chimpimol, N. Mhuadthongon and P. Krachodnok, “Development of a real-time backup battery voltage and environmental monitoring system for a server room using IoT technology,” EUREKA: Physics and Engineering, no. 2, pp. 108–120, 2026.

[3] K. Khurshid, K. Khurshid, M. Usman Hadi, M. Al Bataineh and N. Saeed, “Securing AIoT Surveillance: Techniques, Challenges, and Solutions,” in IEEE Open Journal of the Communications Society, vol. 6, pp. 6517-6550, 2025.

[4] M. L. Ali and Z. Zhang, “The YOLO framework: A comprehensive review of evolution, applications, and benchmarks in object detection,” Computers, vol. 13, no. 12, p. 336, 2024.

[5] S. Kang, Z. Hu, L. Liu, K. Zhang and Z. Cao, “Object detection YOLO algorithms and their industrial applications: Overview and comparative analysis,” Electronics, vol. 14, no. 6, p. 1104, 2025.

[6] R. Pereira, G. Carvalho, L. Garrote and U. J. Nunes, “SORT and Deep-SORT based multiobject tracking for mobile robotics: Evaluation with new data association metrics,” Applied Sciences, vol. 12, no. 3, p. 1319, 2022.

[7] N. Anjeana and K. Anusudha, “Real-time face recognition system based on YOLO and InsightFace,” Multimedia Tools and Applications, vol. 83, pp. 31893–31910, 2024.

[8] R. Xu, S. Razavi and R. Zheng, “Edge Video Analytics: A Survey on Applications, Systems and Enabling Techniques,” in IEEE Communications Surveys & Tutorials, vol. 25, no. 4, pp. 2951-2982, 2023.

[9] S. Wan, S. Ding and C. Chen, “Edge computingenabled video segmentation for real-time traffic monitoring in the Internet of Vehicles,” Pattern Recognition, vol. 121, p. 108146, 2022.

[10] H. -T. Chen, Y. Chiang and H. -Y. Wei, “Edge Computing Resource Management for CrossCamera Video Analytics: Workload and Model Adaptation,” in IEEE Access, vol. 12, pp. 1209812109, 2024.

[11] R. Kalaria, A. S. M. Kayes, W. Rahayu, E. Pardede and A. S. Shahraki, “Adaptive context-aware access control for IoT environments leveraging fog computing,” International Journal of Information Security, vol. 23, no. 4, pp. 3089–3107, 2024.

[12] S. W. Khan, Q. Hafeez, M. I. Khalid, R. Alroobaea, S. Hussain, J. Iqbal, J. Almotiri and S. S. Ullah, “Anomaly detection in traffic surveillance videos using deep learning,” Sensors, vol. 22, no. 17, p. 6563, 2022.

[13] M. Wang and W. Deng, “Deep face recognition: A survey,” Neurocomputing, vol. 429, pp. 215– 244, 2021.

[14] G. Zhang, J. Yin, P. Deng, Y. Sun, L. Zhou and K. Zhang, “Achieving adaptive visual multiobject tracking with unscented Kalman filter,” Sensors, vol. 22, no. 23, p. 9106, 2022.

[15] E. Badidi, K. Moumane and F. E. Ghazi, “Opportunities, Applications, and Challenges of Edge-AI Enabled Video Analytics in Smart Cities: A Systematic Review,” in IEEE Access, vol. 11, pp. 80543-80572, 2023.

[16] S. Kim, C. Kim and S. Kim, “Improving Performance of Real-Time Object Detection in Edge Device Through Concurrent Multi-Frame Processing,” in IEEE Access, vol. 13, pp. 1522-1533, 2025.

[17] X. Kong, Y. Wu, H. Wang and F. Xia, “Edge Computing for Internet of Everything: A Survey,” in IEEE Internet of Things Journal, vol. 9, no. 23, pp. 23472-23485, Dec. 1, 2022.

[18] W. Wu and J. Lai, “Multi Camera Localization Handover Based on YOLO Object Detection Algorithm in Complex Environments,” in IEEE Access, vol. 12, pp. 15236-15250, 2024.

[19] T. Kim, H. Park, Y. Jin, S. -S. Lee and S. Lee, “Partition Placement and Resource Allocation for Multiple DNN-Based Applications in Heterogeneous IoT Environments,” in IEEE Internet of Things Journal, vol. 10, no. 11, pp. 9836-9848, 2023.

[20] X. Hou, J. Zhou, L. Li, P. Cong, Z. Wu and M. Chen, “Latency and Reliability-Aware Dynamic Task Offloading and Scheduling for EnergyHarvesting Systems in Mobile Edge Computing,” in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 45, no. 8, pp. 3942-3956, Aug. 2026.

[21] J. Jiang, G. Xie, J. Cui and M. Guo, “Surface mine personnel object video tracking method based on YOLOv5-DeepSORT algorithm,” Scientific Reports, vol. 15, no. 1, p. 17123, 2025.

[22] Y. Wang and V. Y. Mariano, “A Multi Object Tracking Framework Based on YOLOv8s and Bytetrack Algorithm,” in IEEE Access, vol. 12, pp. 120711-120719, 2024.

[23] Y. Dai, D. Kim and K. Lee, “An advanced approach to object detection and tracking in robotics and autonomous vehicles using YOLOv8 and LiDAR data fusion,” Electronics, vol. 13, no. 12, p. 2250, 2024.

[24] A. Moksyakov, Y. Wu, S. A. Gadsden, J. Yawney and M. AlShabi, “Object detection and tracking with YOLO and the sliding innovation filter,” Sensors, vol. 24, no. 7, p. 2107, 2024.

[25] Z. Chang, S. Liu, X. Xiong, Z. Cai and G. Tu, “A Survey of Recent Advances in Edge-ComputingPowered Artificial Intelligence of Things,” in IEEE Internet of Things Journal, vol. 8, no. 18, pp. 13849-13875, 2021.

[26] M. Jiang and W. Gong, “Bidirectional Bluetooth Backscatter With Edges,” in IEEE Transactions on Mobile Computing, vol. 23, no. 2, pp. 16011612, 2024.

[27] F. Boutros, P. Siebke, M. Klemt, N. Damer, F. Kirchbuchner and A. Kuijper, “PocketNet: Extreme Lightweight Face Recognition Network Using Neural Architecture Search and Multistep Knowledge Distillation,” in IEEE Access, vol. 10, pp. 46823-46833, 2022.

[28] A. George, C. Ecabert, H. O. Shahreza, K. Kotwal and S. Marcel, “EdgeFace: Efficient Face Recognition Model for Edge Devices,” in IEEE Transactions on Biometrics, Behavior, and Identity Science, vol. 6, no. 2, pp. 158-168, 2024.

[29] Y.-H. Chang, F.-C. Wu, and H.-W. Lin, “Design and implementation of ESP32-based edge computing for object detection,” Sensors, vol. 25, no. 6, p. 1656, 2025.

[30] P. Zhang, L. Tan, Z. Yang, F. Huang, L. Sun and H. Peng, “Device-edge collaborative occluded face recognition method based on crossdomain feature fusion,” Digital Communications and Networks, vol. 11, no. 2, pp. 482–492, 2025.

[31] E. Abdullahu, H. Wache, and M. Piangerelli, “Secure and decentralized hybrid multi-face recognition for IoT applications,” Sensors, vol. 25, no. 18, p. 5880, 2025.

[32] M. Sohail et al., “Deep Learning Based Multi Pose Human Face Matching System,” in IEEE Access, vol. 12, pp. 26046-26061, 2024.

[33] H. Lee, S.-H. Park, J.-H. Yoo, S.-H. Jung and J.-H. Huh, “Face recognition at a distance for a stand-alone access control system,” Sensors, vol. 20, no. 3, p. 785, 2020.

[34] F. Hidayat, U. Elviani, G. B. G. Situmorang, M. Z. Ramadhan, F. A. Alunjati and R. F. Sucipto, “Face Recognition for Automatic Border Control: A Systematic Literature Review,” in IEEE Access, vol. 12, pp. 37288-37309, 2024.

[35] S. Vosta and K.-C. Yow, “A CNN-RNN combined structure for real-world violence detection in surveillance cameras,” Applied Sciences, vol. 12, no. 3, p. 1021, 2022.

[36] W. Xu, D. Zhu, R. Deng, K. Yung and A. W. H. Ip, “Violence-YOLO: Enhanced GELAN algorithm for violence detection,” Applied Sciences, vol. 14, no. 15, p. 6712, 2024.

[37] C. Hua, K. Luo, Y. Wu, and R. Shi, “YOLOABD: A multi-scale detection model for pedestrian anomaly behavior detection,” Symmetry, vol. 16, no. 8, p. 1003, 2024.

[38] A. A. Alsabei, T. M. Alsubait and H. H. Alhakami, “Enhancing Crowd Safety at Hajj: Real-Time Detection of Abnormal Behavior Using YOLOv9,” in IEEE Access, vol. 13, pp. 37748-37761, 2025.

[39] K. U. Duja, I. A. Khan and M. Alsuhaibani, “Video Surveillance Anomaly Detection: A Review on Deep Learning Benchmarks,” in IEEE Access, vol. 12, pp. 164811-164842, 2024.

[40] J. Lee, B. Bagheri and H.-A. Kao, “A cyberphysical systems architecture for Industry 4.0based manufacturing systems,” Manufacturing Letters, vol. 3, pp. 18–23, 2015.

[41] Z. Weng, H. Zhuang, H. Li, B. Ramalingam, R. E. Mohan and Z. Lin, “Online Multi-Face Tracking With Multi-Modality Cascaded Matching,” in IEEE Transactions on Circuits and Systems for Video Technology, vol. 33, no. 6, pp. 27382752, 2023.

[42] B. Li, F. Yang and S. Zhang, “Context-aware risk attribute access control,” Mathematics, vol. 12, no. 16, p. 2541, 2024.

[43] NVIDIA, “DeepStream SDK,” NVIDIA Developer, 2023. [Online]. Available: https:// developer.nvidia.com/deepstream-sdk. Accessed: Oct. 25, 2026.

[44] T. Li et al., “ESMO: Joint Frame Scheduling and Model Caching for Edge Video Analytics,” in IEEE Transactions on Parallel and Distributed Systems, vol. 34, no. 8, pp. 2295-2310, 2023.

[45] A. Masoudi and M. Ahmadi, “Auto-Encoders Derivatives on Different Occluded Face Images: Comprehensive Review and New Results,” in IEEE Access, vol. 13, pp. 195080-195103, 2025.