摘要
输电线受风力等自然条件作用会发生舞动,针对这一问题,运用了一种基于机器视觉的输电线舞动检测方案。该方案在对图像预处理后,使用了一种结合背景差法和自适应阈值分割法的新方法来提取输电线目标的特征值,经透视变换校正特征值后,最终插值重建出输电线舞动曲线。多个工况下的传感器数据与重建曲线对应位置数据对比表明,该方法可以有效提取特征值,并最终直观地呈现整条输电线路的垂直舞动情况,实现了误差小于50mm的输电线目标垂直运动检测。
Power lines would wave under natural conditions such as wind.To solve this problem,power transmission line galloping detection scheme based on machine vision is applied.In this scheme,after image preprocessing,a new method combining background difference method and adaptive threshold segmentation method is used to extract the eigenvalue of the transmission line target.After the eigenvalue is corrected by perspective transformation,the transmission line wave curve is reconstructed by interpolation.The comparison between the sensor data and the corresponding position data of the reconstruction curve under multiple tests shows that the method can effectively extract the eigenvalue,and finally intuitively present the vertical galloping of the whole transmission line,realizing the detection of the vertical movement of the transmission line target with error less than 50 mm.
作者
张浩亚
金永
李海涛
刘佳鑫
Zhang Haoya;Jin Yong;Li Haitao;Liu Jiaxin(College of Information and Communication Engineering,North University of China,Taiyuan 030051.China;College of Science,North University of China.Taiyuan 030051,China)
出处
《国外电子测量技术》
2020年第8期1-4,共4页
Foreign Electronic Measurement Technology
基金
国家自然科学基金(11702257,11602233)
山西省自然科学基金(201801D121112)
山西省研究生创新项目(2020SY367)资助。
关键词
输电线舞动
直方图均衡化
阈值分割
背景差
透视变换
三次样条插值
transmission lines galloping
histogram equalization
threshold segmentation
background subtraction
perspective transformation
cubic spline interpolation
