Generating realistic 3D geological structures from 2D images with AI
Generating Heterogeneous 3D Geological Microstructures from 2D Images via a Stable Diffusion-Adversarial Model
Artificial Intelligence
Summary
It can be hard and costly to get detailed 3D pictures of materials like clay or cement, which are important in many industries. The authors present a new AI method that creates 3D models from simpler 2D images, making this process more accessible. Their approach combines two types of AI techniques to capture complex features better than previous methods. Tests show the generated 3D structures look realistic and match real material properties closely.
What this means in practice
- •For materials engineers: Create detailed 3D models of complex materials from 2D images to aid property simulation without costly 3D imaging equipment.
- •For geological waste planners: Generate accurate 3D models of geological microstructures to better assess material behavior in waste disposal projects.
Authors
Ali Aouf, Eric Laloy, Bart Rogiers, Christophe De Vleeschouwer
Abstract
Characterizing the physical properties of clay and cementitious materials matters across many fields, from materials science to geological waste disposal. Property simulation typically calls for 3D imaging, which is expensive, not always accessible, and technically limited for certain materials. Recent progress in deep generative models offers a way around this, reconstructing 3D volumes from the more easily acquired 2D images. Among GAN-based methods for 3D microstructure generation, SliceGAN has shown strong results for homogeneous isotropic and anisotropic systems. It struggles, however, to capture the finer detail of more complex heterogeneous microstructures, which motivates alternative generative frameworks. We introduce a hybrid approach that draws on the stability and generation quality of denoising diffusion models. Since no 3D ground truth is available, we replace the standard denoising loss with an adversarial loss, which yields a stable training process in our experiments. We show that the resulting model generates microstructures of varying complexity with minimal slice artefacts and close agreement with ground-truth phase fractions and structural descriptors.