Drone Classification and Identification Using Computer Vision, Machine Learning and YOLO V8 Algorithm

Drones have increasingly been exploited by adversaries in border regions to transport contraband such as drugs and weapons. The widespread adoption of drone technology across various sectors has brought significant advantages, but also raised critical concerns involving security, privacy, and regulatory oversight. With the growing density of drones operating in shared airspace, it becomes essential to develop reliable systems for drone recognition and classification to promote safe airspace management. This paper presents a novel technique that integrates computer vision and machine learning to identify and categorize drones effectively. Utilizing advanced deep learning models, our approach extracts distinctive features from images of drones captured by surveillance systems or sensors. These features are then classified into specific drone categories using the YOLOv8 algorithm, enabling real-time detection and monitoring. We validate our method on a diverse dataset comprising multiple drone types and varying environmental scenarios. The experimental findings reveal robust performance and scalability, with high accuracy for both identification and classification tasks. Additionally, we explore the relevance of our system in critical applications such as law enforcement, border patrol, infrastructure surveillance, and emergency management. This work advances drone detection methodologies and addresses key challenges in integrating drones safely within civilian airspace.