基于图像处理的无人机电力巡线航向偏差检测The Course Deviation Detection of Power Line Tracking UAV based on Image Processing
杨坚,徐硕,林中圣
YANG Jian,XU Shuo,LIN Zhongsheng
摘要(Abstract)：
无人机航向偏差检测是无人机电力线智能巡检的关键技术之一。通过对图像中电力线的特征分析,提出一种电力线图像检测并获得航向偏差的方法。首先对原始图像进行预处理,减少背景对电力线检测的干扰;其次采用Sobel算法进行边缘检测,通过形态学处理滤除二值图中的背景噪声并增强电力线边缘;然后利用Hough变换实现电力线的快速提取;最后基于检测结果拟合出当前电力线方向,获得航向偏差。实验结果表明,该算法可以有效减少图像背景干扰,电力线检测速度达到12～24帧/s,实现直线型电力线检测并获得航向偏差,航向准确率达到98.67%,位置准确率达到97.45%,具有检测速度快、准确率高等优点。
The detection of course deviation is the key technology of intelligent inspection of UAV power line based on image processing technology. An image detecting and course deviation obtaining method is proposed by analyzing the characteristics of power line image. Firstly, the original image is preprocessed to reduce the interference of background to power line detection. Then, the Sobel algorithm is used for edge detection. The background noise in the binary image is filtered and the edge of the power line is enhanced through the morphological processing algorithm. Finally, the Hough transform is used to achieve fast extraction of power lines.The course deviation is obtained based on the fitted power lines. The experimental results indicate that the algorithm can effectively reduce the interference of image background. The power line detection speed can reach 12～24 frames per second. The power lines can be detected and the course deviation can be obtained.The course accuracy rate and position accuracy rate reach 98.67% and 97.45%, respectively. It has the advantages of fast detection speed and high accuracy.
关键词(KeyWords)：
无人机;航向偏差;Hough变换;Sobel算子
UAV;course deviation;hough transform;sobel operator
基金项目(Foundation): 国网浙江省电力有限公司科技项目（5211TZ1800LL）
作者(Author):
杨坚,徐硕,林中圣
YANG Jian,XU Shuo,LIN Zhongsheng
DOI: 10.19585/j.zjdl.202109015
参考文献(References)：
- [1]赵鹏，邓春，袁亦超.应用直升飞机巡检输电线路[J].华北电力技术，2002(10):1-3.
- [2]陈晓兵，马玉林，徐祖舰.无人飞机输电线路巡线技术探讨[J].南方电网技术，2008,2(6):59-61.
- [3]彭向阳，刘正军，麦晓明，等.无人机电力线路安全巡检系统及关键技术[J].遥感信息，2015,30(1):51-57.
- [4]杨敏祥，宋巍，赵雪松，等.几种输电线路巡检方式的比较———浅析无人直升机巡检的优势[J].智能电网，2015,5(3):148-154.
- [5]ZHOU G,YUAN J,YEN I L,et al.Robust real-time UAV based power line detection and tracking[A].In:2016 IEEE International Conference on Image Processing(ICIP)[C]//Phoenix,AZ,USA,IEEE,2016:744-748.
- [6]张科比，雷勇.基于摄像头的输电线无人机自主巡线方案设计[J].电子技术应用，2017,43(4):83-85.
- [7]李全民，张运楚，李俊策.架空输电线路巡线机器人的视觉导航[J].计算机工程与应用，2007,43(19):221-224.
- [8]汤明文，戴礼豪，林朝辉，等.无人机在电力线路巡视中的应用[J].中国电力，2013,46(3):35-38.
- [9]钱金菊，韩正伟，易琳，等.图像处理技术在无人机电力线路巡检中的应用[J].电子技术与软件工程，2017(15):72-73.
- [10]姒绍辉，胡伏原，顾亚军，等.一种基于不规则区域的高斯滤波去噪算法[J].计算机科学，2014,41(11):313-316.
- [11]CHEN L C,PAPANDREOU G,KOKKINOS I,et al.DeepLab:semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected CRFs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848.
- [12]赵江曼，孟建良.高压输电线路航拍图像目标边缘检测[J].中国电力，2018,51(2):27-32.
- [13]沈德海，张龙昌，鄂旭.基于Sobel算子的改进边缘检测算法[J].信息技术，2015,39(4):15-18.
- [14]DUDA R O,HART P E.Use of the Hough transformation to detect lines and curves in pictures[J].Communication of the ACM,1972,15(1):11-15.
- [15]张秋雁，杨忠，陈聪，等.一种复杂环境下的电力线检测方法[J].应用科技，2019,46(5):51-56.