无人机目标追踪系统设计与实现

Design and Implementation of UAV Target Tracking System

随着无人机技术的逐渐成熟,无人机也越来越贴合人们的生活。四轴无人飞行器因为小巧灵活深受人们的喜爱,但现实生活中无人机的操控往往需要花费大量的学习时间成本,同时四轴无人机在动态场景中检测和跟踪非常困难。基于此,设计一套基于DeepSORT目标跟踪算法和YOLO-V3目标检测算法的无人机目标追踪系统,实现目标追踪、轨迹推演、手势控制功能,让人们可通过手势来操控无人机,实现目标追踪等复杂繁琐的任务。

With the gradual maturity of UAV technology,UAV is more and more suitable for people’s life.Four-axis unmanned aerial vehicle(UAV)is loved by people because of its small and flexible size.But the control of UAV in real life often requires a lot of learning time cost,and the detection and tracking of four-axis UAV in dynamic scenes is very difficult.Therefore,this paper designs a UAV target tracking system based on DeepSORT target tracking algorithm and YOLO-V3 target detection algorithm to achieve target tracking,trajectory inference,gesture control functions,so that people can use gestures to control the UAV,target tracking and other complex tasks.