Research on key technologies of UAV real-time recognition and tracking based on YOLOv4

With the development of the Internet of Things (IoT) technology and the fifth generation of mobile communication technology (5G), unmanned aerial vehicle (UAV) has been widely used in urban life, such as UAV photography and monitoring, and UAV delivery. However, UAV targets are generally small in size, and characteristics between multiple UAVs are not obvious. Interference from other flying objects and complex electromagnetic environment bring great challenges to accurate detection and stable tracking of UAV targets. Therefore, according to the characteristics of urban environment, it is very necessary to establish a high-efficiency UAV target real-time detection and tracking system. In our study, a UAV detection and tracking method with YOLOv4 as the target detector and Deep-SORT as the target tracker is proposed, which can realize the high-efficiency and high-precision identification and tracking of multiple UAV targets.