| About this Abstract |
| Meeting |
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Symposium
|
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Presentation Title |
Microstructure Classification and Quantification Method for Regular SEM Images of Complex Steel Microstructures Combining EBSD Labeling and Deep Learning |
| Author(s) |
Chunguang Shen, Chenchong Wang, Wei Xu |
| On-Site Speaker (Planned) |
Wei Xu |
| Abstract Scope |
Present work develops an EBSD-trained deep learning (DL) method to integrate the advantages of traditional material characterization information and artificial intelligence strategy for classification and quantification of complex microstructures only using regular SEM images. In this method, EBSD analysis is applied to produce accurate ground truth for guiding DL model training and U-Net architecture is used to establish the correlation between SEM input image and EBSD ground truth using small sample experimental datasets. The proposed method is successfully applied to two engineering steels with complex microstructures, i.e., dual-phase steel and quenching and partitioning steel, to segment different phases and quantify phase content and grain size. Also, this method contributes to accelerate EBSD analysis because EBSD maps can be rapidly produced via present models inputting regular SEM images. The good generality of trained models is well validated using new DP and Q&P steels not belonging to training and testing set. |
| Proceedings Inclusion? |
Definite: Other |