EUVIMEDEUVIMED entdecken
Crossref

MGBFNet: a multi-scale fault detection network for transmission lines based on gated mechanism and bidirectional feature fusion

Shuai Hao, Sirui Mao, Xu Ma, Tianrui Qi, Shiao Fan, Junhao Zhao, Tianqi Li

Measurement Science and Technology · 27.03.2026 · 37 (14) · 145403

DOI: 10.1088/1361-6501/ae537b ↗

Kurz gefasst

Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 09.04.2026. Verfasst von Shuai Hao und 6 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Gefördert von: Natural Science Basic Research Program of Shaanxi Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the challenge of multi-scale fault targets that compromise detection accuracy in complex environments, this paper proposes MGBFNet, a multi-scale fault detection network for transmission lines based on gated mechanism and bidirectional feature fusion. First, to address the issue of low accuracy in multi-scale object detection under complex environmental interference, a gated multi-order aggregation module is proposed, which enhances multi-scale feature perception by combining an agent-based global context aggregation mechanism with parallel multi-order convolutions. Second, to tackle target occlusion caused by varied shooting angles, a bidirectional information exchange and fusion module is designed. Driven by a frequency-aware dual-stream architecture, this module hierarchically exchanges complementary semantics between high- and low-frequency feature representations, thereby actively guiding the robust reconstruction of missing local details. Third, to alleviate the problem of low accuracy caused by unbalanced sample numbers, the Slide loss function is incorporated to dynamically balance the weights of easy and hard samples. Finally, to rigorously evaluate the effectiveness of the proposed approach, a dataset encompassing fault samples from twelve distinct categories was constructed, and extensive comparative experiments were performed against ten well-established detection methods. According to the experimental results, the proposed method can detect multi-scale fault targets on power transmission lines in complex environments, achieving a precision of 95.4% and an mAP@0.5 of 93.2%.

Förderung

  • Natural Science Basic Research Program of Shaanxi Province

Zitieren

APA

Hao, S., Mao, S., Ma, X., Qi, T., Fan, S., Zhao, J., & Li, T. (2026). MGBFNet: a multi-scale fault detection network for transmission lines based on gated mechanism and bidirectional feature fusion. Measurement Science and Technology, 37(14), 145403. https://doi.org/10.1088/1361-6501/ae537b

Vancouver

Hao S, Mao S, Ma X, Qi T, Fan S, Zhao J, et al. MGBFNet: a multi-scale fault detection network for transmission lines based on gated mechanism and bidirectional feature fusion. Measurement Science and Technology. 2026;37(14):145403. doi:10.1088/1361-6501/ae537b

BibTeX
@article{Hao2026,
  author = {Hao, Shuai and Mao, Sirui and Ma, Xu and Qi, Tianrui and Fan, Shiao and Zhao, Junhao and Li, Tianqi},
  title = {MGBFNet: a multi-scale fault detection network for transmission lines based on gated mechanism and bidirectional feature fusion},
  journal = {Measurement Science and Technology},
  year = {2026},
  volume = {37},
  number = {14},
  pages = {145403},
  doi = {10.1088/1361-6501/ae537b}
}
Kostenlos starten →