About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Computational Thermodynamics and Kinetics
|
Presentation Title |
Applying Computer Vision to Electron
Micrography in AI-Supported Alloy Synthesis and Solidification |
Author(s) |
Kaelin Mittel, Taylor Sparks |
On-Site Speaker (Planned) |
Taylor Sparks |
Abstract Scope |
This work seeks to explore how computer vision can be applied to scanning
electron micrography of alloy surfaces. The use of machine learning has enabled a great
increase in the current rate of materials discovery; however, computer vision has seen little
application in the world of materials discovery. During the course of this work, computer
vision will be used in the analysis of scanning electron micrographs of thin film alloys
of two highly immiscible metals, germanium and tin. The goal is to generate an image
segmentation model to identify and characterize nucleation features found in scanning
electron micrographs for discovering methods for synthesizing glassy alloys of these two
elements with broader implications for the practice of alloy synthesis. The expected outcome is the discovery of the best solidification techniques and parameters for creating
a perfectly disordered alloy of tin and germanium through use of computer vision and
tabular regression. |
Proceedings Inclusion? |
Planned: |
Keywords |
Solidification, Machine Learning, Characterization |