Steel surface defect segmentation with SME-DeeplabV3+
PLOS One · 14.08.2025 · 20 (8) · e0329628
DOI: 10.1371/journal.pone.0329628 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 14.08.2025. Verfasst von Haiyan Zhang und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, China. Hauptthemen laut MEDLINE: Steel. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Accurate segmentation of steel surface defects is crucial for ensuring steel quality. This paper presents a steel surface defect segmentation method based on SME-DeepLabV3+ to improve the accuracy and efficiency of segmentation. First, StarNet is adopted as the backbone network, whose unique star operation can achieve efficient transformation from low-dimensional space to high-dimensional features, enhancing the model’s ability to capture steel defect features and accurately distinguish between normal and defective areas. Second, the ELA module is introduced, which is based on an efficient local attention mechanism and uses different sizes of convolution kernels for multiscale analysis of feature maps. During training, it adaptively initializes the weights of convolutional layers and introduces a dynamic threshold adjustment mechanism to adjust thresholds in real time according to the defect conditions of training batches, reducing missed detections and false positives. Finally, we integrate the MSAA module from CM-UNet, whose multiscale attention mechanism can dynamically adjust attention allocation based on defect size, avoiding detection omissions or misjudgements caused by size differences. The experimental results show that the improved model performs excellently in steel surface defect segmentation tasks, significantly improving accuracy and efficiency compared with traditional methods. The mIoU, precision, and MPA evaluation metrics increased by 1.65%, 2.19%, and 0.36%, respectively, providing more effective technical support for steel quality inspection. The combination of StarNet with the MSAA and ELA modules effectively enhances the performance of semantic segmentation models in steel defect detection while reducing computational resource requirements. The code is available at https://github.com/Eric-863/SME-main/tree/main .
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- College of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, China.
- Huai'an Power Supply Branch, State Grid Jiangsu Electric Power Co., Ltd., Huaian, China.
ORCID-Profile: Haiyan Zhang
Zitieren
APA
Zhang, H., Zhao, Z., Liu, Y., Liu, J., Ma, T., Wu, K., Zhuang, Z., & Wang, J. (2025). Steel surface defect segmentation with SME-DeeplabV3+. PLOS One, 20(8), e0329628. https://doi.org/10.1371/journal.pone.0329628
Vancouver
Zhang H, Zhao Z, Liu Y, Liu J, Ma T, Wu K, et al. Steel surface defect segmentation with SME-DeeplabV3+. PLOS One. 2025;20(8):e0329628. doi:10.1371/journal.pone.0329628
BibTeX
@article{Zhang2025,
author = {Zhang, Haiyan and Zhao, Zining and Liu, Yilin and Liu, Jiange and Ma, Tingmei and Wu, Kexiao and Zhuang, Zhiwen and Wang, Jiajun},
title = {Steel surface defect segmentation with SME-DeeplabV3+},
journal = {PLOS One},
year = {2025},
volume = {20},
number = {8},
pages = {e0329628},
doi = {10.1371/journal.pone.0329628}
}
EUVIMED