Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Research Highlight
  • Published:

Microscopy

Autonomous machine vision meets 2D materials science

Nature Reviews Electrical Engineering volume 2, page 714 (2025)Cite this article

Subjects

The characterization of materials at the atomic scale is crucial for advancing the speed and reducing the cost of nanotechnology research. Traditional microscopy methods, while powerful, often require manual inspection and interpretation, which can slow down sample processing. Artificial intelligence (AI) such as ChatGPT, and even those already applied to microscopy automation (particularly, supervised learning algorithms), often demand extensive high-quality training data. Now, writing in ACS Nano, Jingyun Yang et al. present the autonomous technology for optical microscopy & intelligent characterization (ATOMIC), which integrates the foundation model segment anything model (SAM) with large language models (LLMs), unsupervised clustering and topological analysis to enable fully autonomous optical microscopy for data-less characterization of 2D materials when combined with an appropriate text prompt — known as ‘zero-shot’ measurements.

“Our previous work applying machine learning to analyse Raman spectroscopic mapping images demonstrated the feasibility of automated characterization, giving us confidence that foundation models could achieve autonomous characterization capabilities in 2D materials research,” recalls Wang. Now, the results demonstrate the framework’s high accuracy in characterizing 2D materials, achieving 99.7% segmentation accuracy for monolayer MoS2. The system effectively identifies material boundaries and distinguishes between monolayer, bilayer and multilayer regions, despite difficult imaging conditions such as defocus, colour temperature variations and exposure changes. The authors highlight the framework’s ability to detect subtle features, such as grain boundary slits, which are often imperceptible to human observers. This capability is crucial for assessing the quality of 2D material devices.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

References

Original article

  • Yang, J. et al. Zero-shot autonomous microscopy for scalable and intelligent characterization of 2D materials. ACS Nano 19, 35493–35502 (2025)

    Article  Google Scholar 

Related article

  • Wang, H. et al. Frank-van der Merwe growth in bilayer graphene. Matter 4, 3339–3353 (2021)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Nature Reviews Electrical Engineering https://www.nature.com/natrevelectreng/

    Miranda L. Vinay

Authors
  1. Miranda L. Vinay
    View author publications

    Search author on:PubMed Google Scholar

Corresponding author

Correspondence to Miranda L. Vinay.

Additional information

This text on an editorially selected paper was initially drafted using artificial intelligence, then revised and fact-checked by an editor to meet Nature Portfolio publication standards.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Vinay, M.L. Autonomous machine vision meets 2D materials science. Nat Rev Electr Eng 2, 714 (2025). https://doi.org/10.1038/s44287-025-00232-z

Download citation

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s44287-025-00232-z

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing