| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Characterization of Materials through High Resolution Coherent Imaging
|
| Presentation Title |
Enhanced Mineral Characterization With 3D X-Ray CT and AI-Driven Imaging |
| Author(s) |
Parisa Asadi, Matthew Andrew, Andriy Andreyev, Zeyu Zhou |
| On-Site Speaker (Planned) |
Parisa Asadi |
| Abstract Scope |
Accurate characterization of materials is crucial for advancing scientific research and industrial applications. Traditional 2D imaging techniques often fail to capture accurate 3D properties. Our study introduces an advanced 3D X-ray CT imaging framework combined with AI-driven image reconstruction to address these limitations. We employ projection monochromation for material identification and deep learning to enhance signal-to-noise ratios. This workflow produces high-quality images that are automatically classified based on corrected attenuation measurements. Our framework encompasses comprehensive material analysis, including microstructural features, phase distribution, and surface area quantification. Evaluations against traditional 2D SEM-BSE and SEM-EDS analyses demonstrate our method's superior accuracy and consistency. This approach so far has significantly improved mineral identification and structural analysis, offering a transformative potential for high-resolution characterization in materials science. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Characterization, Machine Learning, Extraction and Processing |