EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SAMF-YOLO: A self-supervised, high-precision approach for defect detection in complex industrial environments

Jun Huang iD, Shamsul Arrieya Ariffin iD, Qiang Zhu, Wanting Xu, Qun Yang

PLOS One · 2025

DOI: 10.1371/journal.pone.0327001 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Jun Huang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Intelligent Manufacturing, Wuhu Institute of Technology, Anhui, China. Hauptthemen laut MEDLINE: Industry, Models, Theoretical. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the General Program for the Training of Outstanding Young Teachers of Anhui Province, the Key Nature Science Research Projects of Anhui Province, the Wuhu Institute of Technology Natural Science Research Key Projects.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As object detection models grow in complexity, balancing computational efficiency and feature expressiveness becomes a critical challenge. To address this, we propose SAMF-YOLO, a novel model integrating three key components: SONet, BFAM, and FASFF-Head. The UniRepLKNet backbone, enhanced by the Star Operation, expands the feature space with high efficiency. FASFF-Head performs adaptive multi-scale feature fusion with minimal overhead, and the Bi-temporal Feature Aggregation Module (BFAM) strengthens the detection of small defects. Additionally, the Focaler-IoU loss improves bounding box regression for challenging object scales, and a self-supervised contrastive learning strategy enhances feature representation and model robustness without relying on labeled data. Experimental results demonstrate that SAMF-YOLO surpasses YOLOv11s with a 6.38% improvement in mAP@0.5 and a notable reduction in computational cost, confirming its superiority in accuracy, efficiency, and robustness. The code is released at https://github.com/Missing24ff/SAMF-YOLO.git .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

IndustryModels, TheoreticalAlgorithmsSoftware

Autor:innen und Institutionen

  • Faculty of Intelligent Manufacturing, Wuhu Institute of Technology, Anhui, China.
  • Faculty of Information Technology, City University Malaysia, Kuala Lumpur, Malaysia.
  • Faculty of Computing and Meta Technology, Sultan Idris Education University, Perak, Malaysia.
  • Faculty of Automotive and Aviation, Wuhu Institute of Technology, Anhui, China.

ORCID-Profile: Jun Huang, Shamsul Arrieya Ariffin

Förderung

  • the General Program for the Training of Outstanding Young Teachers of Anhui Province
  • the Key Nature Science Research Projects of Anhui Province
  • the Wuhu Institute of Technology Natural Science Research Key Projects

Zitieren

APA

Huang, J., Arrieya Ariffin, S., Zhu, Q., Xu, W., & Yang, Q. (2025). SAMF-YOLO: A self-supervised, high-precision approach for defect detection in complex industrial environments. PLOS One. https://doi.org/10.1371/journal.pone.0327001

Vancouver

Huang J, Arrieya Ariffin S, Zhu Q, Xu W, Yang Q. SAMF-YOLO: A self-supervised, high-precision approach for defect detection in complex industrial environments. PLOS One. 2025. doi:10.1371/journal.pone.0327001

BibTeX
@article{Huang2025,
  author = {Huang, Jun and Arrieya Ariffin, Shamsul and Zhu, Qiang and Xu, Wanting and Yang, Qun},
  title = {SAMF-YOLO: A self-supervised, high-precision approach for defect detection in complex industrial environments},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0327001}
}
Kostenlos starten →