EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

VQGNet: An Unsupervised Defect Detection Approach for Complex Textured Steel Surfaces

Ronghao Yu iD, Yun Liu iD, Rui Yang, Yingna Wu

Sensors · 27.09.2024 · 24 (19) · 6252

DOI: 10.3390/s24196252 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 27.09.2024. Verfasst von Ronghao Yu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Center for Adaptive System Engineering, ShanghaiTech University, Shanghai 201210, China. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Defect detection on steel surfaces with complex textures is a critical and challenging task in the industry. The limited number of defect samples and the complexity of the annotation process pose significant challenges. Moreover, performing defect segmentation based on accurate identification further increases the task’s difficulty. To address this issue, we propose VQGNet, an unsupervised algorithm that can precisely recognize and segment defects simultaneously. A feature fusion method based on aggregated attention and a classification-aided module is proposed to segment defects by integrating different features in the original images and the anomaly maps, which direct the attention to the anomalous information instead of the irregular complex texture. The anomaly maps are generated more confidently using strategies for multi-scale feature fusion and neighbor feature aggregation. Moreover, an anomaly generation method suitable for grayscale images is introduced to facilitate the model’s learning on the anomalous samples. The refined anomaly maps and fused features are both input into the classification-aided module for the final classification and segmentation. VQGNet achieves state-of-the-art (SOTA) performance on the industrial steel dataset, with an I-AUROC of 99.6%, I-F1 of 98.8%, P-AUROC of 97.0%, and P-F1 of 80.3%. Additionally, ViT-Query demonstrates robust generalization capabilities in generating anomaly maps based on the Kolektor Surface-Defect Dataset 2.

Schlagwörter

Stichwörter der Autor:innen

VQGNetdefect detectionfeature fusionunsupervised anomaly algorithm

Autor:innen und Institutionen

  • Center for Adaptive System Engineering, ShanghaiTech University, Shanghai 201210, China.

ORCID-Profile: Ronghao Yu, Yun Liu

Zitieren

APA

Yu, R., Liu, Y., Yang, R., & Wu, Y. (2024). VQGNet: An Unsupervised Defect Detection Approach for Complex Textured Steel Surfaces. Sensors, 24(19), 6252. https://doi.org/10.3390/s24196252

Vancouver

Yu R, Liu Y, Yang R, Wu Y. VQGNet: An Unsupervised Defect Detection Approach for Complex Textured Steel Surfaces. Sensors. 2024;24(19):6252. doi:10.3390/s24196252

BibTeX
@article{Yu2024,
  author = {Yu, Ronghao and Liu, Yun and Yang, Rui and Wu, Yingna},
  title = {VQGNet: An Unsupervised Defect Detection Approach for Complex Textured Steel Surfaces},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {19},
  pages = {6252},
  doi = {10.3390/s24196252}
}
Kostenlos starten →