This project presents an AI-based drone detection system designed to identify, classify, and track unauthorized drones in real time. With the growing use of drones in agriculture, surveillance, delivery, and disaster management, the risk of misuse in sensitive areas such as airports, military zones, and public events has increased significantly. To address this issue, the proposed system uses artificial intelligence and machine learning techniques to detect drones accurately and distinguish them from birds, aircraft, and other aerial objects. The model is trained using drone image datasets and deep learning, particularly convolutional neural networks, to improve detection reliability. The system processes images and video streams from camera inputs and provides instant alerts when a drone is detected. In addition, it can support monitoring through a web-based interface and hardware components such as ESP32-CAM, LCD, buzzer, and relay modules. The use of data augmentation and diverse training samples improves performance under varying lighting conditions and backgrounds. Overall, the proposed solution offers a low-cost, efficient, and intelligent approach to drone surveillance, enhancing security, reducing false alarms, and supporting safer monitoring of critical and public spaces.
Publication Date: 2026-06-17