AI -Driven Drone Surveillance and Threat Detection System Using YOLOV8

Unmanned Aerial Vehicles (UAVs) are widely used in modern surveillance systems. However, monitoring drone video feeds continuously in real time is still a difficult task. Traditional surveillance systems depend on human operators, which can lead to tiredness and slow response during critical situations. This paper presents an intelligent drone surveillance and threat detection system using the YOLOv8 object detection model. The system processes live drone video footage and detects objects in real time. Detected objects are classified into High, Medium, and Low threat levels based on their position and surrounding conditions. The proposed system also includes automatic vehicle counting using a virtual line-crossing method. It can generate instant alerts when any object enters restricted areas. A user-friendly Graphical User Interface (GUI) is developed to display the live surveillance feed, detected threats, and vehicle count information. The system was tested on aerial traffic videos and showed good accuracy with fast detection speed. Results indicate that the proposed model is effective and reliable for real-time aerial security and surveillance applications.