EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Novel Defect Detection and Quantification Method for Stone Arch Bridges Based on UAV Photogrammetry and Deep Learning

Fan Sun, Qing Chun, Chengwen Zhang

Structural Control and Health Monitoring · 2026-01 · 2026 (1)

DOI: 10.1155/stc/7264726 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Structural Control and Health Monitoring, online erschienen am 10.11.2026. Verfasst von Fan Sun und 2 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ancient stone arch bridges in China possess significant historical, cultural, and scientific value, yet they are increasingly threatened by material deterioration, structural damage, and environmental factors. Traditional visual inspection methods are time‐consuming, labor‐intensive, subjective, and difficult to quantify. To address these challenges, this study proposes an automated framework for defect detection and quantification of stone arch bridges, integrating UAV photogrammetry, SfM‐MVS 3D reconstruction, orthophoto and cylindrical projection generation, and deep learning–based instance segmentation. This framework enables direct pixel‐to‐physical dimension conversion for defect area and crack length measurement. By slicing and annotating high‐resolution projected images, a dataset covering five typical defect categories—efflorescence, staining, cracks, weathering, and vegetation—was constructed. In addition, an improved YOLO26s‐seg model incorporating AFPN and ASFF was adopted for defect detection and segmentation. Experimental results show that the improved model achieved a Box mAP@0.5 of 76.91% and a Mask mAP@0.5 of 76.87%, outperforming the baseline while reducing the parameter count and computational cost. The proposed framework can directly quantify defect area and crack length, providing reliable automated technical support for the inspection and maintenance of ancient stone arch bridges.

Förderung

  • National Key Research and Development Program of China

Zitieren

APA

Sun, F., Chun, Q., & Zhang, C. (2026). A Novel Defect Detection and Quantification Method for Stone Arch Bridges Based on UAV Photogrammetry and Deep Learning. Structural Control and Health Monitoring, 2026(1). https://doi.org/10.1155/stc/7264726

Vancouver

Sun F, Chun Q, Zhang C. A Novel Defect Detection and Quantification Method for Stone Arch Bridges Based on UAV Photogrammetry and Deep Learning. Structural Control and Health Monitoring. 2026;2026(1). doi:10.1155/stc/7264726

BibTeX
@article{Sun2026,
  author = {Sun, Fan and Chun, Qing and Zhang, Chengwen},
  title = {A Novel Defect Detection and Quantification Method for Stone Arch Bridges Based on UAV Photogrammetry and Deep Learning},
  journal = {Structural Control and Health Monitoring},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/stc/7264726}
}
Kostenlos starten →