Determining the density and spatial descriptors of atomic scale defects of 2H–WSe2 with ensemble deep learning
We have demonstrated atomic-scale defect characterization in scanning tunneling microscopy images of single crystal tungsten diselenide using an ensemble of U-Net-like convolutional neural networks. Coordinates, counts, densities, and spatial extents were determined from almost 16 000 defect detections, leading to the rapid identification of defect types and their densities. Our results show that analysis aided by machine learning can be used to rapidly determine the quality of transition metal dichalcogenides and provide much needed quantitative input, which may improve the synthesis process.
Authors
- Luke N. Holtzman (ORCID: https://orcid.org/0009-0007-9898-7075)
- James Hone (ORCID: https://orcid.org/0000-0002-8084-3301)
- Madisen Holbrook (ORCID: https://orcid.org/0000-0002-2019-3562)
- Masa Ishigami (ORCID: https://orcid.org/0000-0001-6014-2199)
- Darian Smalley (ORCID: https://orcid.org/0000-0002-8980-0180)
- Stephanie D. Lough (ORCID: https://orcid.org/0000-0002-4634-0946)
- Katayun Barmak (ORCID: https://orcid.org/0000-0003-0070-158X)
Institutions
- University of Central Florida (US)
- Columbia University (US)
Publication Details
- Journal
- APL Machine Learning
- Published
- 2024-08-05
- DOI
- https://doi.org/10.1063/5.0195116
- Citations
- 1
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 0.12
Funders
- National Science Foundation