# Detecting Atomic Scale Surface Defects in STM of TMDs with Ensemble Deep Learning

arXiv: https://arxiv.org/abs/2312.05160 · HTML: https://mlsanity.com/paper/2312.05160/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 22 references that are in the map.
- First posted on arXiv 8 Dec 2023.

## Builds on (most cited first, 7 of 7)

- [AtomAI: A Deep Learning Framework for Analysis of Image and Spectroscopy Data in (Scanning) Transmission Electron Microscopy and Beyond](https://mlsanity.com/paper/2105.07485/) (2021)
- [Improving the Segmentation of Scanning Probe Microscope Images using Convolutional Neural Networks](https://mlsanity.com/paper/2008.12371/) (2020)
- [Robust Feature Disentanglement in Imaging Data via Joint Invariant Variational Autoencoders: from Cards to Atoms](https://mlsanity.com/paper/2104.10180/) (2021)
- [Discovery of structure-property relations for molecules via hypothesis-driven active learning over the chemical space](https://mlsanity.com/paper/2301.02665/) (2023)
- [Combining Variational Autoencoders and Physical Bias for Improved Microscopy Data Analysis](https://mlsanity.com/paper/2302.04216/) (2023)
- [Ensemble learning and iterative training (ELIT) machine learning: applications towards uncertainty quantification and automated experiment in atom-resolved microscopy](https://mlsanity.com/paper/2101.08449/) (2021)
- [Decoding the shift-invariant data: applications for band-excitation scanning probe microscopy](https://mlsanity.com/paper/2104.10207/) (2021)

## Abstract (arXiv)

Atomic-scale defect detection is shown in scanning tunneling microscopy images of single crystal WSe2 using an ensemble of U-Net-like convolutional neural networks. Standard deep learning test metrics indicated good detection performance with an average F1 score of 0.66 and demonstrated ensemble generalization to C-AFM images of WSe2 and STM images of MoSe2. Defect coordinates were automatically extracted from defect detections maps showing that STM image analysis enhanced by machine learning can be used to dramatically increase sample characterization throughput.