摘要
针对无人机航拍巡检操作繁琐、效率低下等问题,提出了一种基于iOS设备和图像跟踪算法的无人机自主航拍巡检控制终端。使用无人机控制软件开发包搭建了终端与无人机、航拍云台、无线图像传输设备、无人机控制器之间的通信链路。基于FFmpeg实现视频解码,提出了基于自适应尺度的时空上下文跟踪算法,引入尺度参数变化率以在线纠正更新尺度参数的过激变化,大幅增强了航拍云台在复杂背景下的跟踪稳定性,实现了无人机航拍系统对不同尺度目标的自主、稳定跟踪,提高了航拍巡检效率。
A new design scheme of autonomous aerial inspection client based on iOS devices and object tracking algorithm for the complex and inefficient operation of traditional inspection system is presented. The communication link between the UAV, gimbal, wireless transmission system and the UAV controller is completed by Mobile SDK. Use FFmpeg lib in the decoding process. The a- daptive scale STC (spatio-temporal context) algorithm which can update scale parameters according to the tracking windows di- mension changing for object tracking is proposed. The experimental results show that the new client enhances the tracking stability comparing with the original inspection system in complex background situation, improves the efficiency of UVA inspection system.
出处
《电视技术》
北大核心
2016年第10期55-59,共5页
Video Engineering
关键词
无人机巡检
时空上下文
IOS
软件开发包
自适应尺度
unmanned aerial vehicle inspection
spatio-temporal context
lOS
mobile SDK
adaptive scale