Two approaches to inpainting microstructure with deep convolutional generative adversarial networks

TL;DR

This paper presents two deep learning-based methods using generative adversarial networks for microstructure inpainting, aiming to repair defects in material images for accurate analysis, with a user-friendly interface.

Contribution

Introduces two novel GAN-based microstructure inpainting techniques and a no-code GUI for practical application in materials imaging.

Findings

One method offers high speed and simplicity.

The other provides smoother inpainting boundaries.

Both methods effectively generate microstructure inpaintings.

Abstract

Imaging is critical to the characterisation of materials. However, even with careful sample preparation and microscope calibration, imaging techniques are often prone to defects and unwanted artefacts. This is particularly problematic for applications where the micrograph is to be used for simulation or feature analysis, as defects are likely to lead to inaccurate results. Microstructural inpainting is a method to alleviate this problem by replacing occluded regions with synthetic microstructure with matching boundaries. In this paper we introduce two methods that use generative adversarial networks to generate contiguous inpainted regions of arbitrary shape and size by learning the microstructural distribution from the unoccluded data. We find that one benefits from high speed and simplicity, whilst the other gives smoother boundaries at the inpainting border. We also outline the development of a graphical user interface that allows users to utilise these machine learning methods in a 'no-code' environment.