The unmanned aerial vehicle (UAV) monitoring platform has the advantages of flexibility, traceability, convenient storage and wide-view, and it has become an effective tool in the field of intelligent transportation systems. It is still a challenge to extract the traffic parameters from the UAV-videos accurately. First, this paper proposes a YOLO-X deep learning approach to detect vehicles using the DJI Phantom 4RTK UAV and vehicle on-board unit. The UAV is used to monitor the traffic situation at the altitudes of 30m, 50m, and 100m, and the vehicle on-board unit is used to record the accurate speed of the experiment vehicle. The newly announced YOLO-X deep learning framework and the video frame-based detection algorithm are incorporated in this approach, which can acquire the results of vehicle detection and speed calculation. Then, a field experiment was conducted in Tianjin City, which had six scenarios, i.e., UAV flight height (30m, 50m, 100m), and vehicle speed (40km, 60km). Next, the YOLO-X deep learning model was trained based on the collected data, and the aerial data processed by the deep learning neural network was compared with the ground vehicle data. The results show that the accuracy of target detection is 92.58%, the accuracy of target vehicle speed detection is 94%, which are more robust and reliable than other algorithms, this demonstrates that the proposed approach is promising for traffic target detection. In addition, the limitations of this approach were discussed.