电力安全技术2024,Vol.26Issue(11) :36-40.

基于YOLOv8模型的电力线路局部缺陷检测

Local Defect Detection of Power Lines Based on YOLOv8 Model

程义皓 王玲芝
电力安全技术2024,Vol.26Issue(11) :36-40.

基于YOLOv8模型的电力线路局部缺陷检测

Local Defect Detection of Power Lines Based on YOLOv8 Model

程义皓 1王玲芝1
扫码查看

作者信息

  • 1. 西安邮电大学自动化学院,陕西 西安 710121
  • 折叠

摘要

针对电力线路运行过程中存在的较高故障发生率,提出一种基于深度学习的目标检测算法,选取YOLOv8模型作为核心检测模型,基于7772张图像对该模型进行训练,再对300张电力线路缺陷图像进行测试试验.试验结果表明,所采用的YOLOv8模型能够较准确地检测出电力线路的局部缺陷,可以辅助人工进行电力巡检,减少人工资源的浪费,提高巡检效率.

Abstract

In allusion to the relatively high failure rate in the running of power lines,a target detection algorithm based on deep learning is proposed,which selects a YOLOv8 model as the core detection model and train the model based on more than 7772 images,and then tests 300 power line defect images.The test results show that the YOLOv8 model adopted is available to accurately detect the local defects of power lines,and assist in manual power inspection,reduce the overuse of manual resources,and improve the efficiency of inspection.

关键词

电力线路/局部缺陷检测/YOLOv8模型/图像预处理

Key words

power line/local defect detection/YOLOv8 model/image preprocessing

引用本文复制引用

出版年

2024
电力安全技术
中国电机工程学会安全技术专业委员会 苏州热工研究院有限公司

电力安全技术

影响因子:0.351
ISSN:1008-6226
段落导航相关论文