Abstract

This research focuses on developing an advanced UAV detection system using state-of-theart YOLOv8 and YOLOv9 models. By training these models on diverse datasets, we aim to enhance real-time detection accuracy and precision under various environmental conditions. The system is designed to improve UAV safety and minimize bird collisions. Through rigorous benchmarking, we demonstrate significant improvements in detection performance, accuracy, and computational efficiency compared to existing approaches. A user-friendly web interface, built using HTML, CSS, and Flask, provides real-time visualization of detection results and system performance. This innovative system holds immense potential for applications in defense, airspace safety, wildlife conservation, and other UAV-dependent sectors.