Open Access Open Access  Restricted Access Subscription Access

AI-Based Real-Time Border Surveillance System Using LBPH Face Recognition and YOLOv8 Detection

Parth S. Patil, P. V. Lokhande, A. S. Mali, S. T. Patil

Abstract


Border security remains a critical challenge, demanding continuous monitoring, rapid threat detection, and accurate identification of individuals and vehicles entering restricted military zones. Traditional human-based surveillance systems often suffer from fatigue, limited coverage, and delayed decision-making. To address these limitations, this project proposes an AI-powered real-time military border surveillance system integrating face recognition and military vehicle detection using advanced computer vision and deep learning techniques. The first module employs Haar Cascade and LBPH face recognition to authenticate authorised military personnel and automatically identify unauthorised individuals. Upon detecting an unknown face, the system captures the image, retrieves the GPS-based location, and sends an alert email to security authorities. The second module utilises a YOLOv8 object detection model trained specifically for military vehicles, enabling accurate real-time detection and distance estimation using image-based measurements. The entire system is deployed on a Raspberry Pi–based edge computing platform, supporting 24/7 autonomous operation with minimal human intervention. The integration of real-time alerts, location tracking, automated evidence capture, and lightweight on-device inference significantly enhances situational awareness and strengthens border surveillance capabilities. This dual module AI system demonstrates a reliable, scalable, and efficient solution for modernising military security infrastructure and reducing reliance on manual monitoring.


Full Text:

PDF

References


M. Chen, W. Liu, T. Wang, S. Zhang, and A. Liu, “A game-based deep reinforcement learning approach for energy-efficient computation in MEC systems,” Knowledge-Based Systems, vol. 235, Nov. 2022, Art. no. 107660.

M. Chen, A. Liu, W. Liu, K. Ota, M. Dong, and N. N. Xiong, “RDRL: A recurrent deep reinforcement learning scheme for dynamic spectrum access in reconfigurable wireless networks,” IEEE Transactions on Network Science and Engineering, vol. 9, no. 2, pp. 364–376, Mar. 2022, doi: 10.1109/TNSE.2021.3117565.

P. Gupta, B. Pareek, G. Singal, and D. V. Rao, “Edge device-based military vehicle detection and classification from UAV,” Multimedia Tools and Applications, vol. 81, no. 14, pp. 19813–19834, 2022.

Y. Ren, A. Liu, X. Mao, and F. Li, “An intelligent charging scheme maximizing the utility for rechargeable network in smart city,” Pervasive and Mobile Computing, vol. 77, Oct. 2021, Art. no. 101457.

B. Janakiramaiah, G. Kalyani, A. Karuna, L. Prasad, and M. Krishna, “Military object detection in defense using multi-level capsule networks,” Soft Computing, pp. 1–15, Jun. 2021, doi: 10.1007/s00500-021-05912-0.


Refbacks

  • There are currently no refbacks.