| About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Advancements in Lightweight Composites, Materials & Alloys
|
| Presentation Title |
Automated Image Segmentation for High-Throughput Microstructure Analysis to Aid Aluminum Alloy Development |
| Author(s) |
Michael Andrew Tershakovec, Jon-Erik Mogonye, Taylor Wilson Cain |
| On-Site Speaker (Planned) |
Michael Andrew Tershakovec |
| Abstract Scope |
Historically, microstructural characterization of multiphase alloys has been a tedious and qualitative process. With the emergence of different open-source tools, primarily multi-class segmentation, quantitative processes can be expedited for large data sets to extract information more accurately and with less human input. In this talk, I will discuss how these automated image segmentation tools can be used to help complement high-throughput mechanical testing of materials for aluminum alloy development. With automated image segmentation and appropriate workflows, analysis of tested materials can be made in a fraction of the time. Properly segmented micrographs enable statistics from large datasets on features such as phase fractions and characteristic particle dimensions. These statistics, supplemented with local chemical composition data measured using EDS, can then be used to directly correlate with the mechanical behavior observed during testing to aid further development of said alloy. |