EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Yolov8-CA-EYHL: A bridge crack detection algorithm based on improved YOLOv8

Xianwei Zhu iD, He Chao, Yahui Zhang

PLOS One · 22.09.2026 · 21 (9) · e0358872

DOI: 10.1371/journal.pone.0358872 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 22.09.2026. Verfasst von Xianwei Zhu und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical and Electrical Engineering, Zhengzhou University of Industrial Technology, Zhengzhou, China. Hauptthemen laut MEDLINE: Algorithms, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Henan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Automated bridge crack detection is challenging because cracks often exhibit weak contrast, irregular morphology, slender structures, and strong interference from complex surface textures. To address these issues, this study proposes a YOLOv8-CA-EYHL framework that combines filtering-equalization preprocessing with coordinate attention (CA) and enhanced high- and low-frequency feature learning modules. The filtering-equalization strategy improves crack-background contrast while suppressing background interference. CA enhances spatially directional feature representation, while the YHL module strengthens local and high-frequency crack features. The EYHL module further establishes multi-scale feature interaction and spatial recalibration to improve the representation of irregular and slender cracks. Ablation experiments demonstrate that the performance improvement results from the complementary contributions of preprocessing and the proposed architectural components. The proposed model achieves an mAP50 of 93.1% and an mAP50-95 of 63.5%, compared with 64.4% and 42.8%, respectively, for YOLOv8s. It requires 32.6 B FLOPs and 13.7 M parameters while maintaining an inference speed of approximately 76 FPS. Repeated experiments with different random seeds and evaluation on an independent external dataset provide further evidence of the stability and generalisation potential of the proposed framework.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AlgorithmsImage Processing, Computer-AssistedDetection Algorithms

Autor:innen und Institutionen

  • School of Mechanical and Electrical Engineering, Zhengzhou University of Industrial Technology, Zhengzhou, China.
  • School of Intelligent Manufacturing and Electrical Engineering, Nanyang Normal University, Henan, China.
  • School of Intelligent Manufacturing, Nanyang Institute of Technology, Henan, China.

ORCID-Profile: Xianwei Zhu

Förderung

  • Natural Science Foundation of Henan Province

Zitieren

APA

Zhu, X., Chao, H., & Zhang, Y. (2026). Yolov8-CA-EYHL: A bridge crack detection algorithm based on improved YOLOv8. PLOS One, 21(9), e0358872. https://doi.org/10.1371/journal.pone.0358872

Vancouver

Zhu X, Chao H, Zhang Y. Yolov8-CA-EYHL: A bridge crack detection algorithm based on improved YOLOv8. PLOS One. 2026;21(9):e0358872. doi:10.1371/journal.pone.0358872

BibTeX
@article{Zhu2026,
  author = {Zhu, Xianwei and Chao, He and Zhang, Yahui},
  title = {Yolov8-CA-EYHL: A bridge crack detection algorithm based on improved YOLOv8},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {9},
  pages = {e0358872},
  doi = {10.1371/journal.pone.0358872}
}
Kostenlos starten →