EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Defect-Mask2Former: An Improved Semantic Segmentation Model for Precise Small-Sized Defect Detection on Large-Sized Timbers

Mingming Qin iD, Hongxu Li iD, Yuxiang Huang, Xingyu Tong, Zhihong Liang

Sensors · 06.04.2026 · 26 (7) · 2254

DOI: 10.3390/s26072254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.04.2026. Verfasst von Mingming Qin und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650224, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Yunnan Province Science and Technology Department, Yunnan Provincial Department of Education, Southwest Forestry University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The precise segmentation of small-sized defects on wood surfaces is critical for the quality grading of glued laminated timber (GLT). Existing semantic segmentation models face core bottlenecks in this context: high miss rates, blurred boundary localization, and excessive size measurement errors. To address these issues, this paper proposes an improved Defect-Mask2Former model that integrates an Attention-Guided Pyramid Enhancement (AGPE) module and a Defect Boundary Calibration and Correction (DBCC) module. Through synergistic optimization, the model achieved pixel-level precise segmentation. To support model training and validation, a custom image acquisition device was designed, and the PlankDefSeg dataset was constructed, comprising 3500 pixel-level annotated images covering five defect types across six industrial wood species. Experimental results demonstrate that on the PlankDefSeg dataset, Defect-Mask2Former achieved a mean Intersection over Union (mIoU) of 85.34% for small-sized defects, a 17.84% improvement over the baseline Mask2Former. The miss rate was reduced from 20.78% to 5.83%, and the size measurement error was only 2.86%, strictly meeting the ≤3% accuracy requirement of the GB/T26899-2022 standard. The model achieved an inference speed of 27.6 FPS, satisfying real-time detection needs. By integrating the model into the GLT grading workflow, a grading accuracy of 94.3% was achieved, and the processing time per timber was reduced from 30 s to 1.5 s, a 20-fold efficiency improvement. This study provides reliable technical support for intelligent GLT quality grading and offers a reference solution for other industrial surface defect segmentation tasks.

Schlagwörter

Stichwörter der Autor:innen

attention-guided pyramid enhancementdefect boundary calibrationglued laminated timber gradingsmall-sized defectswood defect segmentation

Autor:innen und Institutionen

  • College of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650224, China.
  • College of Materials and Chemical Engineering, Southwest Forestry University, Kunming 650224, China.

ORCID-Profile: Mingming Qin, Hongxu Li

Förderung

  • Yunnan Province Science and Technology Department
  • Yunnan Provincial Department of Education
  • Southwest Forestry University
  • Key R&D Program Project of Yunnan Provincial
  • Yunnan Fundamental Research Projects
  • Yunnan Provincial Department of Education Scientific Research Fund Project

Zitieren

APA

Qin, M., Li, H., Huang, Y., Tong, X., & Liang, Z. (2026). Defect-Mask2Former: An Improved Semantic Segmentation Model for Precise Small-Sized Defect Detection on Large-Sized Timbers. Sensors, 26(7), 2254. https://doi.org/10.3390/s26072254

Vancouver

Qin M, Li H, Huang Y, Tong X, Liang Z. Defect-Mask2Former: An Improved Semantic Segmentation Model for Precise Small-Sized Defect Detection on Large-Sized Timbers. Sensors. 2026;26(7):2254. doi:10.3390/s26072254

BibTeX
@article{Qin2026,
  author = {Qin, Mingming and Li, Hongxu and Huang, Yuxiang and Tong, Xingyu and Liang, Zhihong},
  title = {Defect-Mask2Former: An Improved Semantic Segmentation Model for Precise Small-Sized Defect Detection on Large-Sized Timbers},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {7},
  pages = {2254},
  doi = {10.3390/s26072254}
}
Kostenlos starten →