Super-Resolution of 3D Micro-CT Images Using Generative Adversarial Networks: Enhancing Resolution and Segmentation Accuracy

TL;DR

This paper presents a novel 3D GAN-based method that enhances micro-CT image resolution eightfold, significantly improving segmentation accuracy for rock minerals and pore spaces, thereby advancing digital rock physics.

Contribution

The paper introduces a 3D Wasserstein GAN with Gradient Penalty for super-resolution of micro-CT images, trained on unpaired 2D high-res images, to improve resolution and segmentation accuracy.

Findings

Achieved 8x resolution enhancement in 3D micro-CT images.

Produced high-quality images with 0.4375 micro-m/voxel resolution.

Enabled more accurate mineral and pore segmentation.

Abstract

We develop a procedure for substantially improving the quality of segmented 3D micro-Computed Tomography (micro-CT) images of rocks with a Machine Learning (ML) Generative Model. The proposed model enhances the resolution eightfold (8x) and addresses segmentation inaccuracies due to the overlapping X-ray attenuation in micro-CT measurement for different rock minerals and phases. The proposed generative model is a 3D Deep Convolutional Wasserstein Generative Adversarial Network with Gradient Penalty (3D DC WGAN-GP). The algorithm is trained on segmented 3D low-resolution micro-CT images and segmented unpaired complementary 2D high-resolution Laser Scanning Microscope (LSM) images. The algorithm was demonstrated on multiple samples of Berea sandstones. We achieved high-quality super-resolved 3D images with a resolution of 0.4375 micro-m/voxel and accurate segmentation for constituting minerals and pore space. The described procedure can significantly expand the modern capabilities of digital rock physics.