Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 11, 2026Computational GeosciencesOpen Access

PCP-GAN: Property-Constrained Pore-scale image reconstruction via conditional Generative Adversarial Networks

View Full Paper
Ask AI
Bookmark
Share

Authors

ASAli SadeghkhaniBBBrandon BennettMBMasoud Babaei

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates enhanced image accuracy in subsurface characterization, suggesting improved modeling capabilities.

Key Points

  • This study aims to improve pore-scale image accuracy by leveraging a multi-conditional Generative Adversarial Network to better match bulk formation properties in subsurface characterization.
  • Used a multi-conditional Generative Adversarial Network framework for generating pore-scale images.
  • Trained on thin section samples from multiple depths in a carbonate formation while conditioning on porosity and depth.
  • Validated generated images against core sample properties with morphological and statistical analyses.
  • Achieved strong porosity control (R² = 0.95) with mean absolute errors of 0.0099–0.0197.
  • Generated images demonstrated dual-constraint errors of 1.9–12.4% compared to 37.5–713.6% for real sub-images.
  • Preserved key pore network characteristics and spatial continuity, indicating geological authenticity.

Cite This Study

Sadeghkhani et al. (2026) studied this question.

synapsesocial.com/papers/6a51dc79c18d7f28ca4ffb02https://doi.org/10.1007/s10596-026-10465-y
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A scalable framework for synthesizing conditional porous media images using diffusion and GAN models2026
  2. 2Generation of pore-space images using improved pyramid Wasserstein generative adversarial networks2024 · 15 citations
  3. 3A Segmentation-Independent Workflow for Digital Rock Property Prediction Using Pore-Structure-Preserving Generative Adversarial Networks and Synthetic Data: Porosity Case Study2025
  4. 4CAGAN3D: A generative adversarial network framework for generating 3D structures from 2D pore images in selective laser melting2026
  5. 5A Novel Super-Resolution-Based Multi-Scale Pore Network Modeling Method for Characterizing Complex Rock Inner Structures2025