Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

4D-PreNet: A Unified Preprocessing Framework for 4D-STEM Data Analysis

Created by
  • Haebom

Author

Mingyu Liu (Global Institute of Future Technology of Shanghai Jiao Tong University), Zian Mao (Global Institute of Future Technology of Shanghai Jiao Tong University, University of Michigan Shanghai Jiao Tong University Joint Institute), Zhu Liu (Global Institute of Future Technology of Shanghai Jiao Tong University, School of Chemistry and Chemical Engineering of Shanghai Jiao Tong University), Haoran Zhang (Global Institute of Future Technology of Shanghai Jiao Tong University, University of Michigan Shanghai Jiao Tong University) University Joint Institute), Jintao Guo (Global Institute of Future Technology of Shanghai Jiao Tong University), Xiaoya He (Global Institute of Future Technology of Shanghai Jiao Tong University, University of Michigan Shanghai Jiao Tong University Joint Institute), Xi Huang (Global Institute of Future Technology of Shanghai Jiao Tong University), Shufen Chu (Global Institute of Future Technology of Shanghai Jiao Tong University), Chun Cheng (Global Institute of Future Technology of Shanghai Jiao Tong University), Jun Ding (Center for Alloy Innovation and Design State Key Laboratory for Mechanical Behavior of Materials of Xian Jiaotong University), Yujun Xie (Global Institute of Future Technology of Shanghai Jiao Tong University)

Outline

This paper presents 4D-PreNet, a deep learning-based end-to-end pipeline to address bottlenecks in 4D-STEM (four-dimensional scanning transmission electron microscopy) data preprocessing. 4D-PreNet integrates attention-enhanced U-Net and ResNet architectures to simultaneously perform noise removal, centroid correction, and elliptical distortion correction. Trained on a large-scale simulation dataset with varying noise levels, drift magnitudes, and distortion types, 4D-PreNet effectively generalizes to experimental data obtained under diverse conditions. Quantitative evaluation results demonstrate up to 50% MSE reduction in noise removal and subpixel centroid location with a mean error of less than 0.04 pixels. Compared to existing algorithms, 4D-PreNet demonstrates improved noise suppression and diffraction pattern recovery, enabling high-throughput, reliable real-time 4D-STEM analysis for automated feature analysis.

Takeaways, Limitations

•
Takeaways:
◦
Improving the speed and accuracy of 4D-STEM data preprocessing.
◦
Overcoming the limitations of existing material-specific and difficult-to-generalize algorithms.
◦
High-throughput 4D-STEM real-time analysis and automated characterization are possible.
◦
Improved noise removal and diffraction pattern restoration performance.
◦
Subpixel-level accurate center position detection.
•
Limitations:
◦
Learning based on simulated data may not fully reflect the diversity of actual experimental data.
◦
The generalization performance of 4D-PreNet may vary depending on the diversity of experimental data.
◦
Additional optimization may be required for specific materials or microscope settings.
👍
Made with Slashpage