摘要
针对配电线路导线断股散股缺陷的检测问题,分析现有方法的不足和导线缺陷的突出特点,具体阐述了导线缺陷检测的算法和流程,提出了一种基于图像处理技术的导线缺陷检测方法。利用导线的形状和灰度特征提取复杂背景中的导线,人工提取导线缺陷特征,并训练支持向量机分类器在导线区域中检测存在断股散股缺陷的区域。实验结果表明,提出的导线断股散股缺陷检测方法能够准确检测导线缺陷,自动化性能良好,可应用于配电线路的巡检工作。
Aiming at the problem of detecting the broken strand and loose strand defect of distribution line conductor,based on the analysis of the deficiency of existing detection methods and the prominent features of power line defects,this paper expounds the algorithm and process of power line defects detection,and comes up with the method of power line defects detection based on image processing technology.Extracting the power line in the complex background with the shape and gray feature of it,and then extracting the feature manually according to the defect,at last,training the SVM classifier to detect the defect region of distribution circuit.The experimental results show that the method proposed in this paper can accurately detect the conductor and its defects,has good automatic performance,and can be used in the power distribution line inspection.
作者
王照
黄若榕
周煜
WANG Zhao;HUANG Ruorong;ZHOU Yu(Guangdong Bai-yun Power Supply Bureau,Guangzhou 510080,China)
出处
《电视技术》
2020年第6期79-83,88,共6页
Video Engineering
基金
中国南方电网有限责任公司科技项目(GZHKJXM20180046)。
关键词
图像处理
断股散股
导线缺陷检测
配电线路
image processing
broken strand and loose strand
wire defect detection
distribution line
