跳到主要导航 跳到搜索 跳到主要内容

An Improved Lightweight YOLOv5 Network for Defect Detection of Rail Fasteners

  • Zhen Dai
  • , Zhipeng Wang
  • , Limin Jia
  • , Yong Qin
  • , Lei Tong
  • , Jing Cui
  • Beijing Jiaotong University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

The rail fastener is an important infrastructure of railway line system to ensure the safety of railway operation. There is an urgent need for a set of automatic defect inspection scheme for rail fasteners which have high efficiency and accuracy. To this end, this paper proposes a modified lightweight YOLOv5 model considering the application scenario of UAV. We reconstruct the backbone on the basis of ShuffleNetV2 and RepVGG, and switch the detector head to decoupled type from YOLOX. Data augmentation is adopted to address the problem of deficient defect samples. The results show that Yoloxs-lite-s model with ShuffleNetV2 backbone and YOLOX-s head is the most ideal model.

源语言英语
主期刊名2022 Global Reliability and Prognostics and Health Management Conference, PHM-Yantai 2022
编辑Wei Guo, Steven Li
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9781665496315
DOI
出版状态已出版 - 2022
已对外发布
活动2022 Global Reliability and Prognostics and Health Management Conference, PHM-Yantai 2022 - Yantai, 中国
期限: 13 10月 202216 10月 2022

出版系列

姓名2022 Global Reliability and Prognostics and Health Management Conference, PHM-Yantai 2022

会议

会议2022 Global Reliability and Prognostics and Health Management Conference, PHM-Yantai 2022
国家/地区中国
Yantai
时期13/10/2216/10/22

学术指纹

探究 'An Improved Lightweight YOLOv5 Network for Defect Detection of Rail Fasteners' 的科研主题。它们共同构成独一无二的学术指纹。

引用此