A Mixture of Experts Foundation Model for Scanning Electron Microscopy Image Analysis
This work addresses scalability issues in SEM imaging for materials science, accelerating materials discovery by enabling adaptable models.
The paper tackles the problem of task-specific models and labor-intensive acquisition in scanning electron microscopy (SEM) by introducing the first foundation model for SEM images, which generalizes across material systems and imaging conditions and outperforms state-of-the-art techniques in defocus-to-focus image translation.
Scanning Electron Microscopy (SEM) is indispensable in modern materials science, enabling high-resolution imaging across a wide range of structural, chemical, and functional investigations. However, SEM imaging remains constrained by task-specific models and labor-intensive acquisition processes that limit its scalability across diverse applications. Here, we introduce the first foundation model for SEM images, pretrained on a large corpus of multi-instrument, multi-condition scientific micrographs, enabling generalization across diverse material systems and imaging conditions. Leveraging a self-supervised transformer architecture, our model learns rich and transferable representations that can be fine-tuned or adapted to a wide range of downstream tasks. As a compelling demonstration, we focus on defocus-to-focus image translation-an essential yet underexplored challenge in automated microscopy pipelines. Our method not only restores focused detail from defocused inputs without paired supervision but also outperforms state-of-the-art techniques across multiple evaluation metrics. This work lays the groundwork for a new class of adaptable SEM models, accelerating materials discovery by bridging foundational representation learning with real-world imaging needs.