AI‐Based Digital Rocks Augmentation and Assessment Metrics

Abstract Reliable uncertainty model calculation in subsurface engineering from pore‐ and grain‐scale to field‐scale relies on sufficient data, but subsurface data set acquisition remains a challenge, particularly in domains where data collection is expensive or time‐consuming, such as Computed Topog...

Full description

Saved in:
Bibliographic Details
Main Authors: Lei Liu, Bernard Chang, Maša Prodanović, Michael J. Pyrcz
Format: Article
Language:English
Published: Wiley 2025-05-01
Series:Water Resources Research
Subjects:
Online Access:https://doi.org/10.1029/2024WR037939
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850115889792286720
author Lei Liu
Bernard Chang
Maša Prodanović
Michael J. Pyrcz
author_facet Lei Liu
Bernard Chang
Maša Prodanović
Michael J. Pyrcz
author_sort Lei Liu
collection DOAJ
description Abstract Reliable uncertainty model calculation in subsurface engineering from pore‐ and grain‐scale to field‐scale relies on sufficient data, but subsurface data set acquisition remains a challenge, particularly in domains where data collection is expensive or time‐consuming, such as Computed Topography (CT) imaging for digital rock images. While AI‐based data augmentation may assist the model training, it still requires many training images as well as the quality assessment of generated data. Yet, most data quantitative metrics flatten spatial data into vectors; therefore, removing the essential spatial relationships within the data. We evaluate topology‐based metrics for quality assessment of AI‐based image augmentation, coupled with digital rocks augmentation practice using the Single image Generative Adversarial Network (SinGAN) for binarized (segmented) images. Compared to most traditional dimensionality reduction methods that process images into a flattened vector, we propose topological image analysis for dimensionality reduction while preserving the essential geometric and topological features of the high‐dimensional data. To demonstrate our proposed approach, we evaluate the generated images starting from four distinct digital rock samples, sorted sandstone, synthetic sphere pack, limestone, and poorly sorted sandstone, using Minkowski functionals, image graph network‐based measures, graph Laplacian‐based measures, local trend maps, and a homogeneity‐heterogeneity classifier. Our workflow suggests that AI‐based digital rock augmentation, combined with topological dimensionality reduction offers a powerful tool for enhanced quality assessment and diagnostic of digital rock augmentation and improved interpretation to support decision‐making.
format Article
id doaj-art-0d8e4196228b4d1a816b468f4f59315f
institution OA Journals
issn 0043-1397
1944-7973
language English
publishDate 2025-05-01
publisher Wiley
record_format Article
series Water Resources Research
spelling doaj-art-0d8e4196228b4d1a816b468f4f59315f2025-08-20T02:36:28ZengWileyWater Resources Research0043-13971944-79732025-05-01615n/an/a10.1029/2024WR037939AI‐Based Digital Rocks Augmentation and Assessment MetricsLei Liu0Bernard Chang1Maša Prodanović2Michael J. Pyrcz3Hildebrand Department of Petroleum and Geosystems Engineering The University of Texas at Austin Austin TX USAHildebrand Department of Petroleum and Geosystems Engineering The University of Texas at Austin Austin TX USAHildebrand Department of Petroleum and Geosystems Engineering The University of Texas at Austin Austin TX USAHildebrand Department of Petroleum and Geosystems Engineering The University of Texas at Austin Austin TX USAAbstract Reliable uncertainty model calculation in subsurface engineering from pore‐ and grain‐scale to field‐scale relies on sufficient data, but subsurface data set acquisition remains a challenge, particularly in domains where data collection is expensive or time‐consuming, such as Computed Topography (CT) imaging for digital rock images. While AI‐based data augmentation may assist the model training, it still requires many training images as well as the quality assessment of generated data. Yet, most data quantitative metrics flatten spatial data into vectors; therefore, removing the essential spatial relationships within the data. We evaluate topology‐based metrics for quality assessment of AI‐based image augmentation, coupled with digital rocks augmentation practice using the Single image Generative Adversarial Network (SinGAN) for binarized (segmented) images. Compared to most traditional dimensionality reduction methods that process images into a flattened vector, we propose topological image analysis for dimensionality reduction while preserving the essential geometric and topological features of the high‐dimensional data. To demonstrate our proposed approach, we evaluate the generated images starting from four distinct digital rock samples, sorted sandstone, synthetic sphere pack, limestone, and poorly sorted sandstone, using Minkowski functionals, image graph network‐based measures, graph Laplacian‐based measures, local trend maps, and a homogeneity‐heterogeneity classifier. Our workflow suggests that AI‐based digital rock augmentation, combined with topological dimensionality reduction offers a powerful tool for enhanced quality assessment and diagnostic of digital rock augmentation and improved interpretation to support decision‐making.https://doi.org/10.1029/2024WR037939AIdata augmentationdigital rockassessment metric
spellingShingle Lei Liu
Bernard Chang
Maša Prodanović
Michael J. Pyrcz
AI‐Based Digital Rocks Augmentation and Assessment Metrics
Water Resources Research
AI
data augmentation
digital rock
assessment metric
title AI‐Based Digital Rocks Augmentation and Assessment Metrics
title_full AI‐Based Digital Rocks Augmentation and Assessment Metrics
title_fullStr AI‐Based Digital Rocks Augmentation and Assessment Metrics
title_full_unstemmed AI‐Based Digital Rocks Augmentation and Assessment Metrics
title_short AI‐Based Digital Rocks Augmentation and Assessment Metrics
title_sort ai based digital rocks augmentation and assessment metrics
topic AI
data augmentation
digital rock
assessment metric
url https://doi.org/10.1029/2024WR037939
work_keys_str_mv AT leiliu aibaseddigitalrocksaugmentationandassessmentmetrics
AT bernardchang aibaseddigitalrocksaugmentationandassessmentmetrics
AT masaprodanovic aibaseddigitalrocksaugmentationandassessmentmetrics
AT michaeljpyrcz aibaseddigitalrocksaugmentationandassessmentmetrics