This paper aims at the problem of vehicle detection in UAV aerial videos and tries to investigate the performance of different kinds of object detection algorithms under such a condition. We chose ViBe, HOG+SVM, Faster R-CNN and YOLOv3 as the typical detection algorithms. After applying them on UAV aerial videos with different height and traffic scenes and testing their precision and recall, we make a theoretical analysis of their performance. Our experiments showed that Faster R-CNN is the best method. Meanwhile, ViBe, with the advantages of less calculation and relatively low hardware requirements, is a good choice for a simple scene where the vehicles keep moving all the time.

>With the maturity and popularity of UAV (unmanned aerial vehicle) technology [1–3], UAV video is becoming an effective supplement to the fixed monitoring video [4, 5]. In the aspect of traffic information acquisition, the advantages of UAV are obvious. UAV can fly not only between the buildings, but also on the freeway, and even into the tunnel, showing the unique flexibility and maneuverability. UAV can control its hovering position artificially and has a high angle shot to get more comprehensive and clearer video data.In some emergent situation, such as evacuation caused by typhoons and earthquakes, there are lots of countryside areas without fixed road monitoring cameras. UAV is undoubtedly the best choice to

利用无人机航拍获取交通信息具有灵活性好、机动性高等优势,能在突发事件、应急处理等环境发挥重要作用.提出一种针对无人视航拍视频进行车辆检测的方法.针对无人机视频的偏转问题,采用基于SURF算法的图像配准方法,将无人机视频的偏转根据标准帧进行矫正.通过基于像素的自适应分割算法将背景和运动车辆分隔,检测出视频中运动的车辆目标.实验表明:本方法匹配的准确率达到90％,能在较大的场景范围内检测出运动车辆,召回率(Re),精度(Pr),假阴性率(FNR)达到82.3％,90.1％,17.3％.

The synchronized operating system of multistepping motors is widely applied in industry. In this paper, a method of Fuzzy-PID control based on fuzzy control educed from the conventional PID control method is introduced. The stability of synchronized operation of multistepping motors can be improved with this method. The simulation analysis based on practical stepping motor parameters shows that this method is simpler and not only has less overshoot but also a shorter setting time than the conventional PID controller. At the same time, its steady-state characteristics and robustness are preferable. This method is easy to implement and has good performance making it possible to apply Fuzzy-PID control in many practical situations.