Deep Learning Segmentation of Polycrystalline Superconductors with Different Compositions
Image analysis to identify the phases from microstructural images is an important issue for understanding the mechanism associated with the microstructures of functional polycrystalline materials. In this study, the segmentation ability of the deep learning model and the effect of data augmentation were investigated when applied to ceramic superconducting materials with different compositions than the trained materials.
Authors
- Takahiro Hosokawa
- Akiyasu Yamamoto (ORCID: https://orcid.org/0000-0001-6346-3422)
- Yusuke Shimada (ORCID: https://orcid.org/0000-0003-0348-2765)
- Shinnosuke Tokuta (ORCID: https://orcid.org/0000-0001-9558-6670)
- Yoshiki Nishiya
- Haruka Iga
- Yu Hirabayashi
Institutions
- Kyushu University (JP)
- Tokyo University of Agriculture and Technology (JP)
Publication Details
- Journal
- IEEJ Transactions on Fundamentals and Materials
- Published
- 2024-08-31
- DOI
- https://doi.org/10.1541/ieejfms.144.373
- Citations
- 1
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 0.12