基于数字图像处理的电力线异物识别方法研究  

Research on foreign object recognition method of power line based on digital image processing

作　　者：梁新福 罗日成[1] 党世轩 周晶[1] 阳冠菲 LIANG Xinfu;LUO Richeng;DANG Shixuan;ZHOU Jing;YANG Guanfei(Changsha University of Science and Technology,Changsha 410114)

机构地区：[1]长沙理工大学,长沙410114

出　　处：《电气技术》2022年第2期73-78,共6页Electrical Engineering

基　　金：广东电网有限责任公司科技项目资助(GDKJXM20184782)。

摘　　要：针对架空线路异物搭挂的潜在威胁,本文提出一种快速识别航拍图像中异物的方法。首先运用直线段检测(LSD)算法从预处理后的图像中提取电力线,并基于架空导线上异物特征设计针对异物的识别算法,提取导线异物;接着对其进行边缘检测,提取缺陷画框并在巡检图像中标识;最后,通过多组样本数据对算法实用性能进行评估。实验结果表明,该方法能对输电导线上搭挂异物的情况进行有效识别,在无人机电力巡检中具有一定的应用价值。Aiming at the potential threat of foreign objects hanging on overhead lines,a method for quickly identifying foreign objects in aerial images is proposed.First uses the linear segment detection(LSD)algorithm to extract the power line from the preprocessed image,and designs a foreign body recognition algorithm based on the foreign body feature on the overhead wire to extract the wire foreign body;then performs edge detection on it,extracts the defect frame and identifies it in the inspection image;finally,the practical performance of the algorithm is evaluated through multiple sets of sample data.The experimental results show that this method can effectively identify foreign objects hanging on the transmission wires,and it has great application value in the power inspection of unmanned aerial vehicles(UAVs).

关 键 词：输电线路 异物识别 直线段检测(LSD) 边缘检测 

分 类 号：TM75[电气工程—电力系统及自动化] TP391.41[自动化与计算机技术—计算机应用技术]