Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning
Grain size has a significant effect on the mechanical properties of metals. It is very important to accurately rate the grain size number of steels. This paper presents a model for automatic detection and quantitative analysis of the grain size of ferrite–pearlite two-phase microstructure to segment...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Materials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1944/16/5/1974 |
| _version_ | 1797614990308409344 |
|---|---|
| author | Xiaolin Zhu Yuhong Zhu Cairong Kang Mingqi Liu Qiang Yao Pingze Zhang Guanxi Huang Linning Qian Zhitao Zhang Zhengjun Yao |
| author_facet | Xiaolin Zhu Yuhong Zhu Cairong Kang Mingqi Liu Qiang Yao Pingze Zhang Guanxi Huang Linning Qian Zhitao Zhang Zhengjun Yao |
| author_sort | Xiaolin Zhu |
| collection | DOAJ |
| description | Grain size has a significant effect on the mechanical properties of metals. It is very important to accurately rate the grain size number of steels. This paper presents a model for automatic detection and quantitative analysis of the grain size of ferrite–pearlite two-phase microstructure to segment ferrite grain boundaries. In view of the challenging problem of hidden grain boundaries in pearlite microstructure, the number of hidden grain boundaries is inferred by detecting them with the confidence of average grain size. The grain size number is then rated using the three-circle intercept procedure. The results show that grain boundaries can be accurately segmented by using this procedure. According to the rating results of grain size number of four types of ferrite–pearlite two-phase microstructure samples, the accuracy of this procedure is greater than 90%. The grain size rating results deviate from those calculated by experts using the manual intercept procedure by less than Grade 0.5—the allowable detection error specified in the standard. In addition, the detection time is shortened from 30 min of the manual intercept procedure to 2 s. The procedure presented in this paper allows automatic rating of grain size number of ferrite–pearlite microstructure, thereby effectively improving the detection efficiency and reducing the labor intensity. |
| first_indexed | 2024-03-11T07:20:07Z |
| format | Article |
| id | doaj.art-f03db63dcfdd424b8d4ff4a0b3ce8bbf |
| institution | Directory Open Access Journal |
| issn | 1996-1944 |
| language | English |
| last_indexed | 2024-03-11T07:20:07Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Materials |
| spelling | doaj.art-f03db63dcfdd424b8d4ff4a0b3ce8bbf2023-11-17T08:05:28ZengMDPI AGMaterials1996-19442023-02-01165197410.3390/ma16051974Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep LearningXiaolin Zhu0Yuhong Zhu1Cairong Kang2Mingqi Liu3Qiang Yao4Pingze Zhang5Guanxi Huang6Linning Qian7Zhitao Zhang8Zhengjun Yao9College of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaCollege of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaJiangsu Zhongxin Pipe Sci-Tec Co., Ltd., Nanjing 211100, ChinaDongying Industrial Product Inspection & Metrology Verification Center, Dongying 257000, ChinaJiangsu Product Quality Testing & Inspection Institute, Nanjing 210007, ChinaCollege of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaJiangsu Product Quality Testing & Inspection Institute, Nanjing 210007, ChinaJiangsu Product Quality Testing & Inspection Institute, Nanjing 210007, ChinaDongying Industrial Product Inspection & Metrology Verification Center, Dongying 257000, ChinaCollege of Material Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaGrain size has a significant effect on the mechanical properties of metals. It is very important to accurately rate the grain size number of steels. This paper presents a model for automatic detection and quantitative analysis of the grain size of ferrite–pearlite two-phase microstructure to segment ferrite grain boundaries. In view of the challenging problem of hidden grain boundaries in pearlite microstructure, the number of hidden grain boundaries is inferred by detecting them with the confidence of average grain size. The grain size number is then rated using the three-circle intercept procedure. The results show that grain boundaries can be accurately segmented by using this procedure. According to the rating results of grain size number of four types of ferrite–pearlite two-phase microstructure samples, the accuracy of this procedure is greater than 90%. The grain size rating results deviate from those calculated by experts using the manual intercept procedure by less than Grade 0.5—the allowable detection error specified in the standard. In addition, the detection time is shortened from 30 min of the manual intercept procedure to 2 s. The procedure presented in this paper allows automatic rating of grain size number of ferrite–pearlite microstructure, thereby effectively improving the detection efficiency and reducing the labor intensity.https://www.mdpi.com/1996-1944/16/5/1974grain sizegrain boundarysegmentationclassificationintelligent ratingdeep learning |
| spellingShingle | Xiaolin Zhu Yuhong Zhu Cairong Kang Mingqi Liu Qiang Yao Pingze Zhang Guanxi Huang Linning Qian Zhitao Zhang Zhengjun Yao Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning Materials grain size grain boundary segmentation classification intelligent rating deep learning |
| title | Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning |
| title_full | Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning |
| title_fullStr | Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning |
| title_full_unstemmed | Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning |
| title_short | Research on Automatic Identification and Rating of Ferrite–Pearlite Grain Boundaries Based on Deep Learning |
| title_sort | research on automatic identification and rating of ferrite pearlite grain boundaries based on deep learning |
| topic | grain size grain boundary segmentation classification intelligent rating deep learning |
| url | https://www.mdpi.com/1996-1944/16/5/1974 |
| work_keys_str_mv | AT xiaolinzhu researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT yuhongzhu researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT cairongkang researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT mingqiliu researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT qiangyao researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT pingzezhang researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT guanxihuang researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT linningqian researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT zhitaozhang researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning AT zhengjunyao researchonautomaticidentificationandratingofferritepearlitegrainboundariesbasedondeeplearning |