General Resolution Enhancement Method in Atomic Force Microscopy (AFM) Using Deep Learning

TL;DR

This paper introduces a deep learning-based method to enhance the resolution of AFM images, enabling better post-processing and analysis of topography measurements across various materials.

Contribution

A novel deep convolutional neural network approach for improving AFM image resolution in post-processing, applicable to diverse materials.

Findings

High-resolution images comparable to experimental data

Effective across various material types

Potential as a general AFM image enhancement tool

Abstract

This paper develops a resolution enhancement method for post-processing the images from Atomic Force Microscopy (AFM). This method is based on deep learning neural networks in the AFM topography measurements. In this study, a very deep convolution neural network is developed to derive the high-resolution topography image from the low-resolution topography image. The AFM measured images from various materials are tested in this study. The derived high-resolution AFM images are comparable with the experimental measured high-resolution images measured at the same locations. The results suggest that this method can be developed as a general post-processing method for AFM image analysis.