科技创新与应用2025，Vol.15Issue(20)：7-11,5.DOI:10.19981/j.CN23-1581/G3.2025.20.002
基于深度学习的用电网络绝缘子识别定位技术
司桂行 1刘明 1朱向荣 1马兆兴1
作者信息
- 1. 中铁十局集团电务工程有限公司,济南 250101
- 折叠
摘要
Abstract
At present,most of the identification and positioning of power line insulators rely on manual work,which is costly and inefficient.The development of deep learning theory has greatly promoted the intelligent identification and positioning of insulators.This paper takes the image recognition of insulators in power supply networks as the research object.Based on the deep learning method and combined with the application characteristics of the power supply safety detection and monitoring device system,the application in intelligent recognition of power supply safety detection and monitoring images is studied.In this paper,two convolutional networks are combined.The new algorithm reduces the generation of redundant windows and uses a more intelligent sliding mechanism to improve positioning accuracy.This improvement makes the unmanned monitoring process of insulators more efficient,and to a certain extent provides technical support for unmanned real-time monitoring of power networks.关键词
用电网络/接触网/绝缘子/深度学习/定位辨识Key words
electricity network/contact line/insulator/deep learning/positioning and identification分类
信息技术与安全科学引用本文复制引用
司桂行,刘明,朱向荣,马兆兴..基于深度学习的用电网络绝缘子识别定位技术[J].科技创新与应用,2025,15(20):7-11,5.
