Reconstruction of 3D Porous Media From 2D Slices

Volkhonskiy, Denis; Muravleva, Ekaterina; Судаков, Олег Валериевич; Orlov, Denis; Belozerov, Boris; Burnaev, Evgeny; Koroteev, Dmitry

doi:10.48550/arxiv.1901.10233

Cited by 8 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mosser et al 13 proposed to use average grain size and chord length as the minimum training image size. However, a representative elementary size (RES) analysis should be carried out for heterogeneous and complex samples to find an adequate training image size 34 . RES analysis is a methodology to determine the smallest size of a system that is large enough to capture the system's heterogeneity as a whole 51 .…”

Section: Reconstruction Of Representative Microstructuresmentioning

confidence: 99%

See 1 more Smart Citation

Quantifying microstructures of Earth materials using higher-order spatial correlations and deep generative adversarial networks

Amiri

Vasconcelos

Jiao

et al. 2022

Preprint

View full text Add to dashboard Cite

Key to most subsurface processes is to determine how structural and topological features at small length scales, i.e., the microstructure, control the effective and macroscopic properties of earth materials. Recent progress in imaging technology has enabled us to visualise and characterise microstructures at different length scales and dimensions. However, one limitation of these technologies is the trade-off between resolution and sample size (or representativeness). A promising approach to this problem is image reconstruction which aims to generate statistically equivalent microstructure but at a larger scale and/or additional dimension. In this work, a generative adversarial network (GAN), trained with Wasserstein-loss and gradient penalty, is used to reconstruct two-dimensional images of two hydrothermally rock samples with varying degrees of complexity. Moreover, we employ multi-point spatial correlation functions - known as statistical microstructural descriptors (SMDs) to evaluate the reconstruction performance of our GAN and show that it can reconstruct higher-order, spatially-correlated patterns of complex earth materials, capturing underlying structural and morphological properties. Comparing our results with a stochastic reconstruction method based on a two-point correlation function, we show the importance of coupling training/assessment of GANs with higher-order SMDs, especially in the case of complex microstructures.

show abstract

Section: Reconstruction Of Representative Microstructuresmentioning

confidence: 99%

“…Several studies have shown the successful application of these techniques for 2D and 3D 13,29,30 . Notably, a growing body of literature has recently investigated 2D to 3D image reconstructions intending to infer 3D morphological and structural properties using features extracted from 2D images in specific orientations [31][32][33][34] .…”

Section: Introductionmentioning

confidence: 99%

Quantifying microstructures of Earth materials using higher-order spatial correlations and deep generative adversarial networks

Amiri

Vasconcelos

Jiao

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The benefit of such loss is that it can be easily applied to 3D textures. Previous works [27,39] on synthesizing 3D porous material used GANs-based methods with 3D convolutional layers inside a generator and a discriminator. However, they trained separate models for each texture.…”

Section: Related Workmentioning

confidence: 99%

“…We apply our method to 3D texture-like porous media structures which is a real-world problem from Digital Rock Physics. Synthesis of porous structures plays an important role [39] because an assessment of the variability in the inherent material properties is often experimentally not feasible. Moreover, usually it is necessary to acquire a number of representative samples of the void-solid structure.…”

Section: Introductionmentioning

confidence: 99%

User-controllable Multi-texture Synthesis with Generative Adversarial Networks

Alanov

Kochurov

Volkhonskiy

et al. 2020

Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

Self Cite

View full text Add to dashboard Cite

Figure 1: One can take 1) New Guinea 3264 × 4928 landscape photo, learn 2) a manifold of 2D texture embeddings for this photo, visualize 3) texture map for the image and perform 4) texture detection for a patch using distances between learned embeddings. AbstractWe propose a novel multi-texture synthesis model based on generative adversarial networks (GANs) with a usercontrollable mechanism. The user control ability allows to explicitly specify the texture which should be generated by the model. This property follows from using an encoder part which learns a latent representation for each texture from the dataset. To ensure a dataset coverage, we use an adversarial loss function that penalizes for incorrect reproductions of a given texture. In experiments, we show that our model can learn descriptive texture manifolds for large datasets and from raw data such as a collection of high-resolution photos. Moreover, we apply our method to produce 3D textures and show that it outperforms existing baselines.

show abstract

“…Mosser et al (2017) firstly used DCGANs to reconstruct 3D porous structures of sandstone and carbonate. After their ground-breaking work, DCGANs rapidly swept the community and became an important approach for digital rock reconstruction (Mosser et al, 2018;Liu et al, 2019;Volkhonskiy et al, 2019;Valsecchi et al, 2020). Inspired by this, various GAN models have also been used, including but not limited to conditional GAN (Feng et al, 2019;Volkhonskiy et al, 2019), style GAN (Fokina et al, 2020), progressively growing GAN (Zheng and Zhang, 2020), Bicycle GAN (Feng, et al, 2020), as well as some hybrid models combining variational autoencoder and DCGAN (Shams et al, 2020;.…”

Section: Introductionmentioning

confidence: 99%

Digital Rock Reconstruction Using Wasserstein GANs with Gradient Penalty

Zhu

et al. 2022

Day 1 Mon, February 21, 2022

View full text Add to dashboard Cite

Due to the scarcity and vulnerability of physical rock samples, digital rock reconstruction plays an important role in the numerical study of reservoir rock properties and fluid flow behaviors. With the rapid development of deep learning technologies, generative adversarial networks (GANs) have become a promising alternative to reconstruct complex pore structures. Numerous GAN models have been applied in this field, but many of them suffer the unstable training issue. In this study, we apply the Wasserstin GAN with gradient penalty, also known as the WGAN-GP network, to reconstruct Berea sandstone and Ketton limestone. Unlike many other GANs using the Jesnen-Shannon divergence, the WGAN-GP network exhibits a stable training performance by using the Wasserstin distance to measure the difference between generated and real data distributions. Moreover, the generated image quality correlates with the discriminator loss. This provides us an indicator of the training state instead of frequently subjective assessments in the training of deep convolutional GAN (DCGAN) based models. An integrated framework is presented to automate the entire workflow, including training set generation, network setup, model training and synthetic rock validation. Numerical results show that the WGAN-GP network accurately reconstructs both Berea sandstone and Ketton limestone in terms of two-point correlation and morphological properties.

show abstract

Reconstruction of 3D Porous Media From 2D Slices

Cited by 8 publications

References 18 publications

Quantifying microstructures of Earth materials using higher-order spatial correlations and deep generative adversarial networks

Quantifying microstructures of Earth materials using higher-order spatial correlations and deep generative adversarial networks

User-controllable Multi-texture Synthesis with Generative Adversarial Networks

Digital Rock Reconstruction Using Wasserstein GANs with Gradient Penalty

Contact Info

Product

Resources

About