Research on Intelligent Identification and Grading of Nonmetallic Inclusions in Steels Based on Deep Learning
Micromachines · 19.02.2023 · 14 (2) · 482
DOI: 10.3390/mi14020482 ↗Kurz gefasst
Fachartikel in der Zeitschrift Micromachines, online erschienen am 19.02.2023. Verfasst von Xiaolin Zhu und 10 weiteren Autor:innen. Beteiligte Einrichtung: College of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Program of Jiangsu Provincial Administration for Market Regulation.
Abstract (Originalsprache)
Non-metallic inclusions are unavoidable defects in steel, and their type, quantity, size, and distribution have a great impact on the quality of steel. At present, non-metallic inclusions are mainly detected manually, which features high work intensity, low efficiency, proneness to misjudgment, and low consistency of results. In this paper, based on deep neural network algorithm, a small number of manually labeled, low-resolution metallographic images collected by optical microscopes are used as the dataset for intelligent boundary extraction, classification, and rating of non-metallic inclusions. The training datasets are cropped into those containing only a single non-metallic inclusion to reduce the interference of background information and improve the accuracy. To deal with the unbalanced distribution of each category of inclusions, the reweighting cross entropy loss and focal loss are respectively used as the category prediction loss and boundary prediction loss of the DeepLabv3+ semantic segmentation model. Finally, the length and width of the minimum enclosing rectangle of the segmented inclusions are measured to calculate the grade of inclusions. The resulting accuracy is 90.34% in segmentation and 90.35% in classification. As is verified, the model-based rating results are consistent with those of manual labeling. For a single sample, the detection time is reduced from 30 min to 15 s, significantly improving the detection efficiency.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China.
- Jiangsu Product Quality Testing & Inspection Institute, Nanjing 210007, China.
- Jiangsu Changbao Precision Steel Tube Co., Ltd., Changzhou 213200, China.
- Jiangsu Zhongxin Pipe Sci-Tec Co., Ltd., Nanjing 211100, China.
Förderung
- Science and Technology Program of Jiangsu Provincial Administration for Market Regulation
Zitieren
APA
Zhu, X., Wan, W., Qian, L., Cai, Y., Chen, X., Zhang, P., Huang, G., Liu, B., Yao, Q., Li, S., & Yao, Z. (2023). Research on Intelligent Identification and Grading of Nonmetallic Inclusions in Steels Based on Deep Learning. Micromachines, 14(2), 482. https://doi.org/10.3390/mi14020482
Vancouver
Zhu X, Wan W, Qian L, Cai Y, Chen X, Zhang P, et al. Research on Intelligent Identification and Grading of Nonmetallic Inclusions in Steels Based on Deep Learning. Micromachines. 2023;14(2):482. doi:10.3390/mi14020482
BibTeX
@article{Zhu2023,
author = {Zhu, Xiaolin and Wan, Wenhai and Qian, Ling and Cai, Yu and Chen, Xiang and Zhang, Pingze and Huang, Guanxi and Liu, Bo and Yao, Qiang and Li, Shaoyuan and Yao, Zhengjun},
title = {Research on Intelligent Identification and Grading of Nonmetallic Inclusions in Steels Based on Deep Learning},
journal = {Micromachines},
year = {2023},
volume = {14},
number = {2},
pages = {482},
doi = {10.3390/mi14020482}
}
EUVIMED