4.2 Article

Key Parts of Transmission Line Detection Using Improved YOLO v3

期刊

出版社

ZARKA PRIVATE UNIV
DOI: 10.34028/iajit/18/6/1

关键词

Deep learning; YOLO v3; electric tower; insulator

资金

  1. National Natural Science Foundations of China [61671412]
  2. Zhejiang Provincial Natural Science Foundation of China [LY19F010002, LY21F010014]
  3. Fundamental and Commonweal Projects of Zhejiang Province [LGN20F010001]
  4. General Scientific Research Project of Zhejiang Education Department [Y201941122]
  5. Ningbo Municipal Projects for Leading and Top Talents [NBLJ201801006]
  6. Natural Science Foundation of Ningbo, China [2018A610053, 202003N4323]
  7. Innovation and consulting project from Ninghai Power Supply Company, State Grid Corporation of Zhejiang, China
  8. Fundamental Research Funds for Zhejiang Provincial Colleges and Universities
  9. School-level Research and Innovation Team of Zhejiang Wanli University

向作者/读者索取更多资源

This paper proposes a deep learning detection model based on YOLO v3, which improves detection accuracy, flops, and speed by simplifying the neural network structure and optimizing the dataset during training.
Unmanned Aerial Vehicle (UAV) inspection has become one of main methods for current transmission line inspection, but there are still some shortcomings such as slow detection speed, low efficiency, and inability for low light environment. To address these issues, this paper proposes a deep learning detection model based on You Only Look Once (YOLO) v3. On the one hand, the neural network structure is simplified, that is the three feature maps of YOLO v3 are pruned into two to meet specific detection requirements. Meanwhile, the K-means++ clustering method is used to calculate the anchor value of the data set to improve the detection accuracy. On the other hand, 1000 sets of power tower and insulator data sets are collected, which are inverted and scaled to expand the data set, and are fully optimized by adding different illumination and viewing angles. The experimental results show that this model using improved YOLO v3 can effectively improve the detection accuracy by 6.0%, flops by 8.4%, and the detection speed by about 6.0%.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.2
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据