EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

WFD-YOLO: A Hybrid YOLO Architecture with Frequency-Domain Guidance for Weld Defect Segmentation

Shuo Wang, Mingwei Li, Feng Xue, Hongxia Zhang, Dagong Jia

Applied Sciences · 14.06.2026 · 16 (12) · 6019

DOI: 10.3390/app16126019 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 14.06.2026. Verfasst von Shuo Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, National Major Scientific Instruments and Equipments Development Project, Tianjin Metrology Technology Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Precise segmentation of weld defects offers clearer advantages over simple localization in the modern manufacturing, which can improve reliability in high-density weld zones. In order to improve the segmentation mean Average Precision (mAP) and inference speed, we propose a hybrid WFD-YOLO that employs a wavelet-based frequency down-sampling (WFD) module, a lightweight channel-thresholding attention (CTA), and a dedicated P2 small-object layer for weld defect segmentation, where the WFD module is used for suppressing aliasing while preserving low-frequency structural details, the CTA module is used for reducing the impact of background and noise during defect segmentation, and the dedicated P2 small-object layer is used for giving explicit sensitivity to minor defects like porosity and spatters. The upgraded model improves precision by 3.5%, recall by 7.8%, mAP@0.5 by 7.3%, and mAP@0.5–0.95 by 2.7% over the original YOLO11n-seg, while achieving an inference speed of 303 FPS. The segmentation mAP for porosity and spatters, which represent the most challenging defect categories, is improved by 16% and 15.8%, respectively. These performance gains position the hybrid WFD-YOLO network as an industry-deployable tool for safety-critical weld inspection, compatible with high-speed automated welding production lines.

Förderung

  • National Key Research and Development Program of China
  • National Major Scientific Instruments and Equipments Development Project
  • Tianjin Metrology Technology Project
  • Natural Science Foundation of Tianjin

Zitieren

APA

Wang, S., Li, M., Xue, F., Zhang, H., & Jia, D. (2026). WFD-YOLO: A Hybrid YOLO Architecture with Frequency-Domain Guidance for Weld Defect Segmentation. Applied Sciences, 16(12), 6019. https://doi.org/10.3390/app16126019

Vancouver

Wang S, Li M, Xue F, Zhang H, Jia D. WFD-YOLO: A Hybrid YOLO Architecture with Frequency-Domain Guidance for Weld Defect Segmentation. Applied Sciences. 2026;16(12):6019. doi:10.3390/app16126019

BibTeX
@article{Wang2026,
  author = {Wang, Shuo and Li, Mingwei and Xue, Feng and Zhang, Hongxia and Jia, Dagong},
  title = {WFD-YOLO: A Hybrid YOLO Architecture with Frequency-Domain Guidance for Weld Defect Segmentation},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {12},
  pages = {6019},
  doi = {10.3390/app16126019}
}
Kostenlos starten →