EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

EAS-DETR: An Enhanced Real-Time Transformer with Sparse Attention and Global Context for PCB Defect Inspection

Yuxin Yan, Ruize Wu, Jia Ren

Electronics · 15.04.2026 · 15 (8) · 1662

DOI: 10.3390/electronics15081662 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 15.04.2026. Verfasst von Yuxin Yan und 2 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang Provincial Natural Science Foundation, Science and Technology Department of Zhejiang Province, China National Textile and Apparel Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Printed circuit board (PCB) defect inspection is critical for ensuring product reliability, yet it remains challenging due to the microscopic scale of defects and complex background patterns. To improve the localization of fine anomalies, this paper proposes EAS-DETR, an efficient and highly sensitive real-time end-to-end detector. First, we reconstruct the feature extraction backbone by introducing a novel C2f-EC module, which jointly models local textures and global structural dependencies. Second, an Adaptive Sparse Attention-based Intra-scale Feature Interaction (ASAFI) module is proposed to suppress background noise and focus the network’s attention on sparse defect regions. Finally, an optimized feature pyramid network, SGO-FPN, is designed to mitigate cross-scale feature misalignment and preserve high-resolution spatial details for small object localization. Experiments demonstrate that EAS-DETR achieves an mAP@0.5 of 93.0% and a 91.9% recall on a multi-source PCB dataset. The model outperforms mainstream YOLO variants and baseline RT-DETR models while maintaining a moderate parameter count of 14.6M and achieving a real-time inference speed of over 70 FPS. Furthermore, cross-domain validations on public benchmarks confirm its robust generalization capability for complex tiny object detection tasks.

Förderung

  • Zhejiang Provincial Natural Science Foundation
  • Science and Technology Department of Zhejiang Province
  • China National Textile and Apparel Council

Zitieren

APA

Yan, Y., Wu, R., & Ren, J. (2026). EAS-DETR: An Enhanced Real-Time Transformer with Sparse Attention and Global Context for PCB Defect Inspection. Electronics, 15(8), 1662. https://doi.org/10.3390/electronics15081662

Vancouver

Yan Y, Wu R, Ren J. EAS-DETR: An Enhanced Real-Time Transformer with Sparse Attention and Global Context for PCB Defect Inspection. Electronics. 2026;15(8):1662. doi:10.3390/electronics15081662

BibTeX
@article{Yan2026,
  author = {Yan, Yuxin and Wu, Ruize and Ren, Jia},
  title = {EAS-DETR: An Enhanced Real-Time Transformer with Sparse Attention and Global Context for PCB Defect Inspection},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {8},
  pages = {1662},
  doi = {10.3390/electronics15081662}
}
Kostenlos starten →