Abstract:

—The rise in complexity of modern warfare, combined with the need for real-time situational awareness, has led to the use of Artificial Intelligence (AI) in defense systems. This paper presents an AI-powered Military Intrusive Detection and Target Acquisition System. It is designed to detect, identify, and track unauthorized intrusions in restricted military areas. The system uses computer vision and deep learning models along with hardware-level edge computing to provide automated surveil- lance.By using the YOLOv5 model integrated with Raspberry Pi and OpenCV, the system achieves real-time object detection with a precision rate of 94%. Alerts are generated through a Flask- based backend to instantly notify the control unit. Experimental results show that the system reliably recognizes potential threats and effectively reduces false alarms.