摘要
随着无人机技术的日益发展和计算机视觉原理的广泛应用,无人机的视觉部分逐渐成为其核心竞争力的一部分。本文详细分析了Lucas–Kanade光流算法,以及其在无人机的室内定位方面的研究于应用,同时对相关算法进行优化处理,相较于传统的算法,运行速度提升22.8%。无人机的室内定位精度,在自稳定状态下能够在半径为15cm的圆内,若外加水平方向上干扰,能够迅速回复原位置。
With the increasing development of drone technology and the widespread application of computer vision principles,the visual part of the drone has gradually become part of its core competitiveness.This article analyzes in detail the Lucas–Kanade optical flow algorithm and its research and application in indoor positioning of drones,and optimizes related algorithms.Compared with traditional algorithms,the running speed is increased by 23%.The indoor positioning accuracy of the UAV can be in a circle with a radius of 15cm in a self-stable state,and if it is disturbed in the horizontal direction,it can quickly return to its original position.
作者
金正康
秦工
李朝阳
李申
Jin Zhengkang;Qin Gong;Li Chaoyang;Li Shen(School of intelligent manufacturing,Jianghan University,Wuhan Hubei 430000)
出处
《电子测试》
2020年第19期52-55,共4页
Electronic Test
基金
湖北省高等院校2019年省级大学生创新训练项目。
关键词
光流
无人机
室内定位
滤波算法
optical flow
UAV
indoor positioning
filtering algorithm
