《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (5): 1591-1597.DOI: 10.11772/j.issn.1001-9081.2021050796

• 多媒体计算与计算机仿真 • 上一篇    下一篇

基于输电杆塔区域提取的图像匹配算法

郭可贵1, 曹瑞2, 万能3, 汪晓1, 尹悦1, 唐旭明4, 熊军林2()   

  1. 1.国网安徽省电力有限公司 超高压分公司, 合肥 230601
    2.中国科学技术大学 信息科学技术学院, 合肥 230026
    3.国网安徽省电力有限公司 合肥供电公司, 合肥 230000
    4.国网安徽省电力有限公司 淮南供电公司, 安徽 淮南 232007
  • 收稿日期:2021-05-17 修回日期:2021-12-22 接受日期:2021-12-23 发布日期:2022-03-08 出版日期:2022-05-10
  • 通讯作者: 熊军林
  • 作者简介:郭可贵(1985—),男,安徽淮南人,工程师,主要研究方向:输电线路无人机巡检
    曹瑞(1997—),男,安徽六安人,硕士研究生,主要研究方向:视觉同时定位与建图
    万能(1990—),男,江西临川人,工程师,硕士,主要研究方向:输电线路无人机巡检
    汪晓(1972—),男,安徽合肥人,高级工程师,硕士,主要研究方向:输电线路运维管理
    尹悦(1991—),男,安徽淮南人,工程师,主要研究方向:输电线路无人机巡检
    唐旭明(1965—),男,安徽淮南人,高级工程师,主要研究方向:输电线路无人机巡检
    熊军林(1977—),男,河南郑州人,教授,博士,主要研究方向:自动控制。 xiong77@ustc.edu.cn
  • 基金资助:
    2019—2020年安徽省电力有限公司科技项目(5212F018008S)

Image matching algorithm based on transmission tower area extraction

Kegui GUO1, Rui CAO2, Neng WAN3, Xiao WANG1, Yue YIN1, Xuming TANG4, Junlin XIONG2()   

  1. 1.Ultra High Voltage Company,State Grid Anhui Electric Power Company Limited,Hefei Anhui 230601,China
    2.School of Information Science and Technology,University of Science and Technology of China,Hefei Anhui 230026,China
    3.Hefei Electric Power Supply Company,State Grid Anhui Electric Power Company Limited,Hefei Anhui 230000,China
    4.Huainan Electric Power Supply Company,State Grid Anhui Electric Power Company Limited,Huainan Anhui 232007,China
  • Received:2021-05-17 Revised:2021-12-22 Accepted:2021-12-23 Online:2022-03-08 Published:2022-05-10
  • Contact: Junlin XIONG
  • About author:GUO Kegui, born in 1985, engineer. His research interestsinclude transmission line unmanned aerial vehicle inspection.
    CAO Rui, born in 1997,M. S. candidate. His research interestsinclude visual simultaneous localization and mapping.
    WAN Neng, born in 1990,M. S.,engineer. His research interestsinclude transmission line unmanned aerial vehicle inspection.
    WANG Xiao, born in 1972,M. S.,senior engineer. His researchinterests include transmission line operation and maintenance management.
    YIN Yue, born in 1991,engineer. His research interests includetransmission line unmanned aerial vehicle inspection.
    TANG Xuming, born in 1965,senior engineer. His researchinterests include transmission line unmanned aerial vehicle inspection.
    XIONG Junlin, born in 1977,Ph. D.,professor. His researchinterests include automatic control.
  • Supported by:
    2019—2020 Science and Technology Project of Anhui Electric Power Company Limited(5212F018008S)

摘要:

针对无人机(UAV)视觉定位过程中传统特征提取与匹配算法匹配质量不佳的问题,提出了一种基于输电杆塔区域提取的图像匹配算法。首先,将图像划分为若干相互重叠的网格区域,并对每个区域采用双层金字塔结构提取特征点,从而保证特征点的均匀分布;其次,使用直线分割检测(LSD)算法提取图像中的直线,从而利用输电杆塔的特殊结构得到输电杆塔的支撑区域;最后,在连续图像中对输电杆塔区域与背景区域内的特征点分别进行匹配,以进一步估计相机运动。在旋转和平移估计实验中,与传统的ORB特征提取与匹配算法相比,所提算法的特征匹配准确率提升了10.1个百分点,相对位姿误差的均值降低了0.049;在UAV巡检实验中,采用所提算法进行UAV轨迹估计的相对误差为2.89%,表明该算法可在实时绕塔飞行过程中实现对UAV位姿的鲁棒、精确估计。

关键词: 无人机巡检, 输电杆塔提取, 均匀化特征, 图像匹配, 运动估计

Abstract:

In order to solve the problem of low matching quality of the traditional feature extraction and matching algorithm in Unmanned Aerial Vehicle (UAV) visual localization, a new image matching algorithm based on transmission tower area extraction was proposed. Firstly, the image was divided into several overlapping grid areas, and the feature points were extracted by a two-layer pyramid structure for each area to ensure the uniform distribution of feature points. Then, the Line Segment Detector (LSD) algorithm was used to extract the lines in the images, the transmission tower support areas were extracted on the basis of special structure of transmission tower. Finally, the feature points in the transmission tower areas and the background areas were matched respectively in continuous images to further estimate the camera motion. In the rotation and translation estimation experiment, compared with the traditional Oriented Features from Accelerated Segment Test(FAST) and Rotated Binary Robust Independent Elementary Features (BRIEF) (ORB) feature extraction and matching algorithm, the proposed algorithm has the feature matching accuracy improved by 10.1 percentage points, the mean value of relative pose error reduced by 0.049. In the UAV inspection experiment, the relative error of the UAV trajectory estimation by using the proposed algorithm is 2.89%, which indicates that the proposed algorithm can achieve the robust and accurate estimation of the UAV’s pose during the real-time flying around the tower.

Key words: Unmanned Aerial Vehicle (UAV) inspection, transmission tower extraction, homogenized feature, image matching, motion estimation

中图分类号: