基于AI的输电线路导线断散股缺陷检测方法

AI-based Method of Detecting Broken Strand Defect of Transmission Line Conductor

针对现有方法对输电线路导线断散股缺陷检测时存在的缺陷检测准确率较低的问题,提出了基于AI技术的输电线路导线断散股缺陷检测方法。首先对所采集的输电线路导线图像进行降噪和增强等处理;然后利用LBG算法完成K-means算法的优化,应用改进后的算法对导线图像进行分割提取;最后采用AI技术中的卷积神经网络算法,实现对输电线路导线断散股缺陷的检测。实验结果表明,采用所提方法对导线断散股检测时,其检测准确率均高于95%,优于对比方法,具有一定应用价值。

Aiming at the problem of low accuracy in defect detection of broken strands in transmission line conductors using existing methods,an AI based method for detecting broken strands in transmission line conductors is proposed.Firstly,denoise and enhance the collected images of the transmission line wires;Then,the LBG algorithm is used to optimize the K-means algorithm,and the improved algorithm is applied to segment and extract the wire image;Finally,the convolutional neural network algorithm in AI technology is used to detect broken strand defects in transmission line conductors.The experimental results show that the detection accuracy of the proposed method for detecting broken strands of wires is higher than 95%,which is superior to the comparison method and has certain application value.