The use of slow-speed low-altitude unmanned aerial vehicles (UAVs) is becoming increasingly widespread, and the characteristics of low-altitude, slow-speed, and small targets make them difficult to be captured by traditional radar detection means. The method of photoelectric following has become a hot issue in current anti-UAVs field. In this paper, we focus on UAV tracking and aiming, and design a UAV tracking and aiming system based on a motion platform, which uses position closed-loop PID control algorithm to control the servo turntable DC motor and captures target images with a high-resolution camera. Then, the YOLO_v4 and OpenCV algorithms are used for image recognition. Besides, this system is based on STM32 tracking system integrated control software for overall system control. Finally, the generality, rationality and stability of the experimental system are verified through real tests. It is shown that YOLO_v4 recognition algorithm has strong background adaptability but slow recognition rate, while OpenCV algorithm has fast recognition rate but is only applicable to pure sky background. We can choose suitable algorithms for different sky background conditions.