EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Container damage detection based on lightweight U-Net and vision transformer

Li Yali, Jintao Wang, Ye Fang, Sun Heting, Jinyu Hao, Feng Zhao

PeerJ Computer Science · 20.04.2026 · 12 · e3770

DOI: 10.7717/peerj-cs.3770 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 20.04.2026. Verfasst von Li Yali und 5 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Central Guidance on Local Science and Technology Development Fund of Shandong Province, Technology Innovation Program of Yantai City.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Container structural integrity is vital to global trade safety and operational efficiency; however, manual damage inspection is time-consuming and error-prone, particularly for minor defects in complex port environments. Although machine vision has provided promising solutions, existing methods remain limited by difficulties in multi-scale damage detection under cluttered backgrounds, high computational complexity, and frequent missed detections of small targets. To overcome these challenges, a novel two-stage detection framework is proposed in this article. In the first stage, a Lightweight U-Net integrated with a Convolutional Block Attention Module (CBAM) is employed to achieve precise container surface segmentation and effective background stripping. By incorporating depthwise separable convolutions in shallow layers and CBAM-enhanced skip connections, the proposed model reduces parameters by 28.4% (to 20.21M) while maintaining high segmentation accuracy (99.03% mIoU) and real-time inference speed (48 FPS). In the second stage, a Vision Transformer (ViT) equipped with an adaptive dual-threshold mechanism is used to classify the segmented container surface as damaged or undamaged. Leveraging ViT’s global attention capability, classification sensitivity is dynamically adjusted through a confidence threshold ($\tau$_s) in conjunction with an optimized classification threshold ($\tau$_c), effectively reducing false positives. Experimental results demonstrate an F1-score of 94.7%, a false positive rate of 5.0%, and a manual review rate of only 10.0%, significantly outperforming You Only Look Once (YOLO)-based detectors, conventional convolutional neural networks (CNNs), and standalone ViT models. Overall, the proposed framework achieves a favorable balance between accuracy, efficiency, and deployability on edge devices, providing a robust “machine-dominant, human-assisted” solution for automated container inspection in port environments.

Förderung

  • National Natural Science Foundation of China
  • Central Guidance on Local Science and Technology Development Fund of Shandong Province
  • Technology Innovation Program of Yantai City
  • Shandong Provincial Natural Science Foundation

Zitieren

APA

Yali, L., Wang, J., Fang, Y., Heting, S., Hao, J., & Zhao, F. (2026). Container damage detection based on lightweight U-Net and vision transformer. PeerJ Computer Science, 12, e3770. https://doi.org/10.7717/peerj-cs.3770

Vancouver

Yali L, Wang J, Fang Y, Heting S, Hao J, Zhao F. Container damage detection based on lightweight U-Net and vision transformer. PeerJ Computer Science. 2026;12:e3770. doi:10.7717/peerj-cs.3770

BibTeX
@article{Yali2026,
  author = {Yali, Li and Wang, Jintao and Fang, Ye and Heting, Sun and Hao, Jinyu and Zhao, Feng},
  title = {Container damage detection based on lightweight U-Net and vision transformer},
  journal = {PeerJ Computer Science},
  year = {2026},
  volume = {12},
  pages = {e3770},
  doi = {10.7717/peerj-cs.3770}
}
Kostenlos starten →