Numerous industries, including defense, logistics, surveillance, agriculture, and entertainment, have undergone radical change as a result of the quick development of Unmanned Aerial Vehicles (UAVs), also referred to as drones. However, the increasing deployment of drones has introduced significant challenges in ensuring airspace security, privacy, and safety, particularly due to their resemblance to birds and other aerial objects when viewed from a distance. Accurate and real-time detection and classification of drones have thus become critical to mitigating threats such as unauthorized surveillance, espionage, and potential attacks. Using cutting-edge deep learning techniques, this study suggests a reliable drone identification and classification system that makes use of the YOLOv8 (You Only Look Once version 8) object detection algorithm. YOLOv8 is notable for striking a compromise between accuracy and real-time performance, which makes it ideal for dynamic settings. An extensive and varied collection of drone photos— including fixed-wing, single-rotor, and multi-rotor drones— was cited, and the PASCAL VOC format was used for annotation. To guarantee the best possible model performance, the dataset was preprocessed, enhanced, and divided into training, validation, and test sets. To enable users to upload photos or stream videos for real-time drone identification, the trained YOLOv8 model was implemented in a Flask-based web application.