摘要
绝缘子是电力架空线路传输系统中的重要部件,绝缘子的损坏会对电力传输造成极大的负面影响。为了实现复杂背景下绝缘子的快速、准确识别,文章采用卷积神经网络模型,对所有感兴趣的区域进行分类,实现对绝缘子的智能检测,使计算机能够快速地从航拍图像中识别出绝缘子,从而使对绝缘子的故障检测更加准确,实现了在复杂背景下的绝缘子串的语义分割研究。实验结果表明,该方法与已有方法相比,不仅能够在不同光照条件、不同拍摄角度以及复杂背景干扰下实现绝缘子的识别与分割,并且处理时间短、精度高和鲁棒性强。
Insulator is an important part in power transmission overhead line system, the damage to the insulator will cause great negative impact to the power transmission. In order to realize the rapid and accurate identification of insulators under complex background, this paper adopts the convolutional neural network model to classify all areas of interest and realize the intelligent detection of insulators, so that the computer can quickly identify insulators from aerial photos, thus making the fault detection of insulators more accurate and realizing the semantic segmentation research of insulator strings under complex background.The experimental results show that the proposed method compared with existing methods, not only can be taken in different light conditions, different Angle and achieved the identification of insulator and segmentation under complex background interference, and the processing time is short, high accuracy and strong robustness.
作者
李柏阳
Li Boyang(School of Electrical and Electronic Engineering,Hubei University of Technology,Wuhaa 430068,China)
出处
《信息通信》
2020年第12期113-115,共3页
Information & Communications
关键词
绝缘子
卷积神经网络
深度学习
语义分割
故障检测
insulator
convolutional neural network
deep learning
sematic segmentation
fault detection
作者简介
李柏阳(1994-),男,湖北襄阳人,硕士,研究方向:机器学习。
