EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Insulator Surface Defect Detection Method Based on Graph Feature Diffusion Distillation

Shucai Li, Na Zhang, Gang Yang, Yannong Hou, Xingzhong Zhang

Journal of Imaging · 10.06.2025 · 11 (6) · 190

DOI: 10.3390/jimaging11060190 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 10.06.2025. Verfasst von Shucai Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: State Grid Shanxi Electric Power Company Lvliang Power Supply Company, Lvliang 033000, China. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Science and Technology Project of State Grid Shanxi Electric Power Company, Science and Technology Project of State Grid Shanxi Electric Power Company.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aiming at the difficulties of scarcity of defect samples on the surface of power insulators, irregular morphology and insufficient pixel-level localization accuracy, this paper proposes a defect detection method based on graph feature diffusion distillation named GFDD. The feature bias problem is alleviated by constructing a dual-division teachers architecture with graph feature consistency constraints, while the cross-layer feature fusion module is utilized to dynamically aggregate multi-scale information to reduce redundancy; the diffusion distillation mechanism is designed to break through the traditional single-layer feature transfer limitation, and the global context modeling capability is enhanced by fusing deep semantics and shallow details through channel attention. In the self-built dataset, GFDD achieves 96.6% Pi.AUROC, 97.7% Im.AUROC and 95.1% F1-score, which is 2.4–3.2% higher than the existing optimal methods; it maintains excellent generalization and robustness in multiple public dataset tests. The method provides a high-precision solution for automated inspection of insulator surface defect and has certain engineering value.

Schlagwörter

Stichwörter der Autor:innen

defect detectiongraph featuresknowledge distillationteacher-student networksunsupervised learning

Autor:innen und Institutionen

  • State Grid Shanxi Electric Power Company Lvliang Power Supply Company, Lvliang 033000, China.
  • State Grid Shanxi Electric Power Company Electric Power Research Institute, Taiyuan 030001, China.
  • State Grid Shanxi Electric Power Company Ultra High Voltage Substation Branch, Taiyuan 030021, China.
  • School of Software, Taiyuan University of Technology, Taiyuan 030001, China.

Förderung

  • The Science and Technology Project of State Grid Shanxi Electric Power Company
  • Science and Technology Project of State Grid Shanxi Electric Power Company

Zitieren

APA

Li, S., Zhang, N., Yang, G., Hou, Y., & Zhang, X. (2025). Insulator Surface Defect Detection Method Based on Graph Feature Diffusion Distillation. Journal of Imaging, 11(6), 190. https://doi.org/10.3390/jimaging11060190

Vancouver

Li S, Zhang N, Yang G, Hou Y, Zhang X. Insulator Surface Defect Detection Method Based on Graph Feature Diffusion Distillation. Journal of Imaging. 2025;11(6):190. doi:10.3390/jimaging11060190

BibTeX
@article{Li2025,
  author = {Li, Shucai and Zhang, Na and Yang, Gang and Hou, Yannong and Zhang, Xingzhong},
  title = {Insulator Surface Defect Detection Method Based on Graph Feature Diffusion Distillation},
  journal = {Journal of Imaging},
  year = {2025},
  volume = {11},
  number = {6},
  pages = {190},
  doi = {10.3390/jimaging11060190}
}
Kostenlos starten →