The identification and detection of military aircraft hold significant importance across various domains, encompassing aviation safety, border security, and aerial surveillance. With strategic decision-making heavily reliant on accurate aircraft identification, the task remains inherently challenging. This study introduces an innovative approach utilizing You Only Look Once version 5 (YOLOv5), a cutting-edge object detection framework, to address this challenge. Leveraging the multi-detection heads of the YOLOv5 architecture, the model demonstrates the capability to detect aircraft across diverse scales, spanning from small to large objects. Through the prediction of bounding boxes and class labels, crucial information for both identification and localization of aircraft is obtained. Based on the results, the analysis led to the conclusion that the proposed model produced accuracy of 87%. The proposed methodology attains a notable balance between high detection accuracy and real-time performance, rendering it well-suited for a wide range of applications including airspace surveillance, aircraft tracking, and safety enforcement.
Real-Time Military Aircraft Detection using YOLOv5
2024-05-17
422373 byte
Conference paper
Electronic Resource
English