基于形态特征的输电线接续管视觉检测方法  

Visual inspection method of transmission line connection pipe based on morphological characteristics

作　　者：江维 林先晶 张宏伟 邹德华 JIANG Wei;LIN Xianjing;ZHANG Hongwei;ZOU Dehua(School of Mechanical Engineering and Automation,Wuhan Textile University,Wuhan 430073,China;Hunan Province Key Laboratory of Intelligent Live Working Technology and Equipment(ROBOT)(State Grid Hunan Ultra-High Voltage Transmission Company),Changsha 420100,China;Live Inspection and Intelligent Operation Technology State Grid Corporation Laboratory(State Grid Hunan Ultra-High Voltage Transmission Company),Changsha 420100,China)

机构地区：[1]武汉纺织大学机械工程与自动化学院,湖北武汉430073 [2]智能带电作业技术及装备(机器人)湖南省重点实验室(国网湖南省电力有限公司超高压输电公司),湖南长沙420100 [3]带电巡检与智能作业技术国家电网公司实验室(国网湖南省电力有限公司超高压输电公司),湖南长沙420100

出　　处：《河南工程学院学报（自然科学版）》2024年第3期39-43,共5页Journal of Henan University of Engineering：Natural Science Edition

基　　金：湖北省自然科学基金项目(2018CFB273);湖北省教育厅科研计划项目(B2019067);湖南省电力有限公司科技项目(5216AJ210005)。

摘　　要：接续管是一种用于大档距输电导线连接的重要输电线金具,其压接质量和内部结构属于隐性工程,一旦接续管出现故障,会导致整条输电线路断裂。为保证输电线路正常稳定运行,需要定期对输电线路接续管进行检修维护,而作业前必须定位到接续管。基于此,提出了一种基于几何形态特征的输电导线接续管视觉检测方法。首先用轮廓和边缘特征检测间接表征接续管外观形态,用直线特征检测定位接续管在图像中所辖区域,用角点特征检测定位各个方向灰度有明显变化的图像凸点,然后提取接续管的轮廓特征、边缘特征、直线特征和角点特征,用多特征信息融合定位检测接续管,最后使用OpenCV计算机视觉函数库和NumPy数学函数库进行仿真,结果表明基于几何形态特征能够实现接续管的检测和识别,满足电力运维检修需求。Connection pipe is an important transmission line tooling used for large distance transmission wire connection,its crimping quality and internal structure are hidden engineering,once the connection pipe fails,the entire transmission line may break,leading to extreme accident.Therefore,to ensure the normal and stable operation of transmission lines need to regularly repair and maintain the transmission line connection pipe,before the operation,the first task is to locate the connection pipe.Based on this,this paper proposes a visual inspection method based on geometric morphological features of transmission line splice tube,which can indirectly characterize the appearance of the splice tube according to the contour and edge detection,locate the area of the connection pipe in the image by straight line detection,and locate the image bumps with obvious changes in grayscale in each direction by corner point detection.In this paper,we extract the contour features,edge features,straight-line features,and corner point features of the tube and fuse them to locate and detect the tube.Finally,using OpenCV computer vision function library and NumPy mathematical function library for algorithm simulation research,simulation results show that based on geometric morphological features can achieve the detection and identification of the connection tube and meeting power operation and maintenance requirements.

关 键 词：接续管 几何形态 角点检测 视觉识别 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]