A Digital Simulation Toolkit for Physics-Based Generation of Realistic Experimental Scanning Tunneling Microscopy Images

2026-09-28

AI Summary

This study addresses the limitation of supervised learning for scanning tunneling microscopy (STM) image denoising caused by the scarcity of paired training data by constructing a physics-driven data generation framework. Clean images are obtained via quantum mechanical simulations, and a novel physics-prior-based noise modeling approach is introduced to superimpose multimodal realistic noise sources, including Gaussian electronic noise, 1/f flicker noise, and mechanical drift. This enables the batch synthesis of high-fidelity, AI-ready datasets that preserve atomic, defect, and electron wave information. Validation on Cu(111) surfaces demonstrates that supervised models trained on these synthetic datasets significantly outperform unsupervised baselines in denoising performance, effectively facilitating scientific discovery of quantum interference patterns.

Abstract

Scanning Tunneling Microscopy (STM) is a widely used tool for characterizing surfaces of materials at the atomic scale, playing a crucial role in discoveries across condensed matter physics and materials science. Despite its extreme spatial resolution, STM is one of the most sensitive microscopy techniques and is highly prone to noise. While existing unsupervised denoising methods are very cheap to train, these are primarily focused on removing the noise with minimal recovery of key physical information. While supervised methods can offer superior performance, the major bottleneck is that a large amount of paired clean-noisy experimental images is required which are impractical to obtain. Thus, we developed a low-cost physics-driven digital toolkit to rapidly generate large volume of realistic STM images. Firstly, we simulate clean images from a chosen material system. Then, with prior knowledge of the physical characteristics of the artifacts and noise present in STM experiments, we formulate several artifact-noise functions such as Gaussian electronic noise, 1/f flicker noise, scan-line noise, background tilt and mechanical drift. These physically informed noise components are then added to the simulated clean images to generate realistic STM images. We demonstrated the capability of the proposed digital toolkit to generate AI-ready data for denoising images of the (111) surfaces of copper and lead, while preserving atoms, defects, and electron waves. We also validated the quality of the downstream image analysis of learning electron wave patterns induced by quantum interference from Cu(111) images. Results show that the supervised models trained on digitally generated AI-ready data can more effectively denoise and learn electron wave patterns on Cu(111) images than benchmarked unsupervised approaches, indicating that the proposed toolkit facilitate scientific discovery.

Problem

• Scanning Tunneling Microscopy
• Image Denoising
• Supervised Learning
• Data Generation
• Physics-Based Simulation

Innovation

• Scanning Tunneling Microscopy
• Physics-based simulation
• Supervised denoising
• Synthetic data generation
• Artifact-noise modeling