Abstract:State-of-the-art electron microscopes such as scanning electron microscopes (SEM), scanning transmission electron microscopes (STEM) and transmission electron microscopes (TEM) have become increasingly sophisticated. However, the quality of experimental images is often hampered by stochastic and deterministic distortions arising from the instrument or its environment. These distortions can arise during any stage of the imaging process, including image acquisition, transmission, or visualization. In this paper, we will discuss the main sources of distortion in TEM and S(T)EM images, develop models to describe them and propose a method to correct these distortions using a convolutional neural network. We demonstrate the effectiveness of our approach on a variety of experimental images and show that it can significantly improve the signal-to-noise ratio resulting in an increase in the amount of quantitative structural information that can be extracted from the image. Overall, our findings provide a powerful framework for improving the quality of electron microscopy images and advancing the field of structural analysis and quantification in materials science and biology.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the influence of various random and deterministic distortions on electron microscope images during the imaging process. These distortions will reduce the image quality and further affect the ability to extract quantitative information from samples. Specifically, the paper focuses on how to restore single - shot electron microscope images through Deep Convolutional Neural Networks (DCNNs) to improve the Signal - to - Noise Ratio (SNR) and thus increase the amount of quantitative structural information that can be extracted from the images. The main distortion sources mentioned in the paper include: 1. **Detector noise**: including counting noise, dark - current noise and read - out noise. 2. **X - ray noise**: caused by X - rays passing through the detector. 3. **Dead - pixel noise**: caused by permanently damaged pixels on the sensor. 4. **Non - linear distortion caused by scanning delay**: especially at high doses, the fly - back time of the probe will lead to horizontal and vertical scanning distortion. 5. **Blurring effect caused by non - uniform scanning speed**: especially obvious in high - speed scanning. To address these problems, the author has developed a method based on deep convolutional neural networks, using a combination of Concatenated Grouped Residual Dense Network (CGRDN) and Generative Adversarial Network (GAN), aiming to restore single - shot electron microscope images affected by multiple distortions. This method can not only learn the end - to - end mapping from distorted images to undistorted images, but also train this mapping through an adaptive loss function. Experimental results show that this method can significantly improve the image quality of periodic and non - periodic samples under different types of severe distortion conditions, and reliably determine the peak position and intensity in atomic - resolution images.

Deep convolutional neural networks to restore single-shot electron microscopy images

Deep convolutional neural networks to restore single-shot electron microscopy images

Deep Learning Enhanced Fluorescence Emission Difference Microscopy

Reconstruction of 3-D Atomic Distortions from Electron Microscopy with Deep Learning.

A Unified Deep Learning Framework for Sstem Image Restoration

Deep learning-based noise filtering toward millisecond order imaging by using scanning transmission electron microscopy

Cryo-CARE: Content-Aware Image Restoration for Cryo-Transmission Electron Microscopy Data

Deep Learning Super-Resolution Electron Microscopy Based on Deep Residual Attention Network

Partial Scanning Transmission Electron Microscopy with Deep Learning

Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy

Accurate atomic scanning transmission electron microscopy analysis enabled by deep learning

Deep Learning of Atomically Resolved Scanning Transmission Electron Microscopy Images: Chemical Identification and Tracking Local Transformations

Deep learning-enhanced fluorescence microscopy via confocal physical imaging model.

Super-resolution Single-Molecule Microscopic Reconstruction Based on Improved Deep-Storm

Mixup-Based Neural Network for Image Restoration and Structure Prediction From SEM Images

A Hybrid Frequency-Spatial Domain Model for Sparse Image Reconstruction in Scanning Transmission Electron Microscopy

Distortion Correction and Denoising of Light Sheet Fluorescence Images

DeepContrast: Deep Tissue Contrast Enhancement using Synthetic Data Degradations and OOD Model Predictions

Advancing electron microscopy using deep learning

Deep Learning Microscopy

Deep Learning Supersampled Scanning Transmission Electron Microscopy