Segmentation and analysis of cement particles in cement paste with deep learning
Scanning electron microscopy (SEM) is a widely used method for the analysis of concrete micro structure. To quantitatively analyze the SEM images with high efficiency and accuracy, an automatic segmentation framework is proposed in this paper. The deep segmentation algorithm is purposely optimized f...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/164657 |
| _version_ | 1824456310575333376 |
|---|---|
| author | Qian, Hanjie Li, Ye Yang, Jianfei Xie, Lihua Tan, Kang Hai |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Qian, Hanjie Li, Ye Yang, Jianfei Xie, Lihua Tan, Kang Hai |
| author_sort | Qian, Hanjie |
| collection | NTU |
| description | Scanning electron microscopy (SEM) is a widely used method for the analysis of concrete micro structure. To quantitatively analyze the SEM images with high efficiency and accuracy, an automatic segmentation framework is proposed in this paper. The deep segmentation algorithm is purposely optimized from PointRend based on the characteristic of SEM images to improve prediction accuracy, especially the performance around boundaries. Moreover, the SEM images can be segmented without additional treatment. Cement paste samples with 0.2 and 0.4 water-to-cement ratios are prepared and cured for 1, 3, 7, 14, and 28 days. Totally SEM images with 2267 labeled cement particles are included to build the dataset. From the results of intersection over union and pixel accuracy, the proposed algorithm outperforms the trainable waikato environment for knowledge analysis (WEKA) segmentation, Fully Convolutional Networks (FCN), and the original PointRend method. The segmentation results are used to calculate the hydration degree of two cement paste samples. Good agreement is obtained with the hydration degree calculated by using nonevaporable water in the samples for the 5 curing durations. At last, the shape of the cement particles is analyzed. Irregularity and roundness of the cement particles do not change significantly with an increase in curing duration. |
| first_indexed | 2025-02-19T03:52:04Z |
| format | Journal Article |
| id | ntu-10356/164657 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:52:04Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1646572023-02-07T08:48:52Z Segmentation and analysis of cement particles in cement paste with deep learning Qian, Hanjie Li, Ye Yang, Jianfei Xie, Lihua Tan, Kang Hai School of Electrical and Electronic Engineering School of Civil and Environmental Engineering Engineering::Civil engineering Segmentation Machine Learning Scanning electron microscopy (SEM) is a widely used method for the analysis of concrete micro structure. To quantitatively analyze the SEM images with high efficiency and accuracy, an automatic segmentation framework is proposed in this paper. The deep segmentation algorithm is purposely optimized from PointRend based on the characteristic of SEM images to improve prediction accuracy, especially the performance around boundaries. Moreover, the SEM images can be segmented without additional treatment. Cement paste samples with 0.2 and 0.4 water-to-cement ratios are prepared and cured for 1, 3, 7, 14, and 28 days. Totally SEM images with 2267 labeled cement particles are included to build the dataset. From the results of intersection over union and pixel accuracy, the proposed algorithm outperforms the trainable waikato environment for knowledge analysis (WEKA) segmentation, Fully Convolutional Networks (FCN), and the original PointRend method. The segmentation results are used to calculate the hydration degree of two cement paste samples. Good agreement is obtained with the hydration degree calculated by using nonevaporable water in the samples for the 5 curing durations. At last, the shape of the cement particles is analyzed. Irregularity and roundness of the cement particles do not change significantly with an increase in curing duration. Ministry of National Development (MND) National Research Foundation (NRF) This work was partially supported by the National Key Research and Development Program of China (Grant No. 2021YFF0500801), the National Research Foundation, Singapore, and Ministry of National Development, Singapore, under its Cities of Tomorrow R&D Programme (CoT Award No. COT-V2-2019-1), Shenzhen Science and Technology Program, China (Grant No. RCYX20200714114525013) and Open Funding of State Key Laboratory of High Performance Civil Engineering Materials (Grant No. 2021CEM006). 2023-02-07T08:48:52Z 2023-02-07T08:48:52Z 2023 Journal Article Qian, H., Li, Y., Yang, J., Xie, L. & Tan, K. H. (2023). Segmentation and analysis of cement particles in cement paste with deep learning. Cement and Concrete Composites, 136, 104819-. https://dx.doi.org/10.1016/j.cemconcomp.2022.104819 0958-9465 https://hdl.handle.net/10356/164657 10.1016/j.cemconcomp.2022.104819 2-s2.0-85144418681 136 104819 en COT-V2-2019-1 Cement and Concrete Composites © 2022 Elsevier Ltd. All rights reserved. |
| spellingShingle | Engineering::Civil engineering Segmentation Machine Learning Qian, Hanjie Li, Ye Yang, Jianfei Xie, Lihua Tan, Kang Hai Segmentation and analysis of cement particles in cement paste with deep learning |
| title | Segmentation and analysis of cement particles in cement paste with deep learning |
| title_full | Segmentation and analysis of cement particles in cement paste with deep learning |
| title_fullStr | Segmentation and analysis of cement particles in cement paste with deep learning |
| title_full_unstemmed | Segmentation and analysis of cement particles in cement paste with deep learning |
| title_short | Segmentation and analysis of cement particles in cement paste with deep learning |
| title_sort | segmentation and analysis of cement particles in cement paste with deep learning |
| topic | Engineering::Civil engineering Segmentation Machine Learning |
| url | https://hdl.handle.net/10356/164657 |
| work_keys_str_mv | AT qianhanjie segmentationandanalysisofcementparticlesincementpastewithdeeplearning AT liye segmentationandanalysisofcementparticlesincementpastewithdeeplearning AT yangjianfei segmentationandanalysisofcementparticlesincementpastewithdeeplearning AT xielihua segmentationandanalysisofcementparticlesincementpastewithdeeplearning AT tankanghai segmentationandanalysisofcementparticlesincementpastewithdeeplearning |