EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Real-Time Cascade Framework for UAV-Based Insulator Defect Detection with Attention-Guided Lightweight CNN

Zeliha Doğan Ersoy iD, Mustafa Gelmez iD, Durmuş Ersoy iD, M Erdem Isenkul iD, Fırat Kaçar iD, Ali Ataş iD

Sensors · 13.08.2026 · 26 (16) · 5126

DOI: 10.3390/s26165126 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 13.08.2026. Verfasst von Zeliha Doğan Ersoy und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Engineering, Istanbul Esenyurt University, Istanbul 34510, Turkey. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Istanbul University-Cerrahpaşa, Istanbul University-Cerrahpasa.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The physical condition of insulators on electrical transmission lines is critical for system reliability. Early fault detection such as broken discs or flashovers prevents outages and accidents. Traditional inspection methods are costly and difficult, driving demand for UAV-based autonomous systems. This article proposes a novel dataset enhancement method and a cascade deep learning model for detecting and classifying insulator disc conditions. Existing datasets were enriched by approximately 30% with wide-background, real-field images to improve model generalization. After comparative analyses, YOLO was selected for detection due to superior performance, while MobileNetV2 was chosen as the classifier for its processing speed advantage. CBAM attention module and Focal Loss function were integrated to boost classification performance and handle class imbalance. The resulting YOLO + MobileNetV2 model achieved 99.76% overall accuracy across “Normal”, “Broken”, and “Flashover” classes, with an F1-score of 0.99 for broken discs, operating with 185.60 ms latency, confirming its viability for real-time inspection.

Schlagwörter

Stichwörter der Autor:innen

CBAMMobileNetV2YOLOdataset enhancementdeep learningfocal lossinsulator fault detection

Autor:innen und Institutionen

  • Department of Computer Engineering, Istanbul Esenyurt University, Istanbul 34510, Turkey.
  • Department of Electric and Electronics Engineering, Istanbul University-Cerrahpasa, Istanbul 34310, Turkey.
  • Department of Research and Development, Mamur Technology Systems Inc., Istanbul 34590, Turkey.
  • Department of Computer Engineering, Istanbul University-Cerrahpasa, Istanbul 34310, Turkey.
  • Department of Motor Vehicles and Transportation Technologies, Istanbul Topkapi University, Istanbul 34020, Turkey.

ORCID-Profile: Zeliha Doğan Ersoy, Mustafa Gelmez, Durmuş Ersoy, M Erdem Isenkul, Fırat Kaçar, Ali Ataş

Förderung

  • Istanbul University-Cerrahpaşa
  • Istanbul University-Cerrahpasa

Zitieren

APA

Ersoy, Z. D., Gelmez, M., Ersoy, D., Isenkul, M. E., Kaçar, F., & Ataş, A. (2026). A Real-Time Cascade Framework for UAV-Based Insulator Defect Detection with Attention-Guided Lightweight CNN. Sensors, 26(16), 5126. https://doi.org/10.3390/s26165126

Vancouver

Ersoy ZD, Gelmez M, Ersoy D, Isenkul ME, Kaçar F, Ataş A. A Real-Time Cascade Framework for UAV-Based Insulator Defect Detection with Attention-Guided Lightweight CNN. Sensors. 2026;26(16):5126. doi:10.3390/s26165126

BibTeX
@article{Ersoy2026,
  author = {Ersoy, Zeliha Doğan and Gelmez, Mustafa and Ersoy, Durmuş and Isenkul, M Erdem and Kaçar, Fırat and Ataş, Ali},
  title = {A Real-Time Cascade Framework for UAV-Based Insulator Defect Detection with Attention-Guided Lightweight CNN},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {5126},
  doi = {10.3390/s26165126}
}
Kostenlos starten →