Enhancing prediction of fluid-saturated fracture characteristics using deep learning super resolution DOI Creative Commons
Manju Pharkavi Murugesu,

Vignesh Krishnan,

Anthony R. Kovscek

et al.

Applied Computing and Geosciences, Journal Year: 2024, Volume and Issue: unknown, P. 100208 - 100208

Published: Nov. 1, 2024

Language: Английский

ESM data downscaling: a comparison of super-resolution deep learning models DOI Creative Commons
Nikhil M. Pawar, Ramin Soltanmohammadi,

Seyed Kourosh Mahjour

et al.

Earth Science Informatics, Journal Year: 2024, Volume and Issue: unknown

Published: June 12, 2024

Abstract Climate projections at fine spatial resolutions are required to conduct accurate risk assessment for critical infrastructure and design adaptation planning. Generating these using advanced Earth system models (ESM) requires significant computational resources. To address this issue, various statistical downscaling techniques have been introduced generate fine-resolution data from coarse-resolution simulations. In study, we evaluate compare five deep learning-based techniques, namely, super-resolution convolutional neural networks, fast network ESM, efficient sub-pixel network, enhanced residual (EDRN), generative adversarial (SRGAN). These applied a dataset generated by the Energy Exascale System Model (E3SM), focusing on key surface variables such as temperature, shortwave heat flux, longwave flux. Models trained validated paired (0.25 $$^{\circ }$$ ∘ ) (1 monthly obtained 9-year simulation. Next, blind testing is performed two different years outside of training validation set. efficiency each technique, metrics used, including mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), learned perceptual image patch (LPIPS). The results show that EDRN outperforms other algorithms in terms PSNR, SSIM, MSE, but struggles capture fine-scale features data. contrast, SRGAN, model uses loss, excels capturing details boundaries internal structures, resulting lower LPIPS than methods.

Language: Английский

Citations

1
Super-resolution reconstruction of hydrate-bearing CT images for microscopic detection of pore DOI Creative Commons
Wangquan Ye, Yu Chen, Liang Chen

et al.

Intelligent Marine Technology and Systems, Journal Year: 2024, Volume and Issue: 2(1)

Published: Sept. 9, 2024

Language: Английский

Citations

1
Advanced machine learning artificial neural network classifier for lithology identification using Bayesian optimization DOI Creative Commons
Saâd Soulaimani, Ayoub Soulaimani, Kamal Abdelrahman

et al.

Frontiers in Earth Science, Journal Year: 2024, Volume and Issue: 12

Published: Nov. 20, 2024

Identifying lithology is crucial for geological exploration, and the adoption of artificial intelligence progressively becoming a refined approach to automate this process. A key feature strategy leveraging population search algorithms fine-tune hyperparameters, thus boosting prediction accuracy. Notably, Bayesian optimization has been applied first time select most effective learning parameters neural network classifiers used identification. This technique utilizes capability utilize past classification outcomes enhance models performance based on physical calculated from well log data. In comparison architectures, Bayesian-optimized (BOANN) demonstrably achieved superior accuracy in validation significantly outperformed non-optimized wide, bilayer, tri-layer configurations, indicating that incorporating can advance lithofacies recognition, offering more accurate intelligent solution identifying lithology.

Language: Английский

Citations

1
Deformation of the void space of pores and fractures of carbonates: comprehensive analysis of core and field data DOI Creative Commons
Dmitriy A. Martyushev, Inna N. Ponomareva, Shadfar Davoodi

et al.

Energy Geoscience, Journal Year: 2024, Volume and Issue: unknown, P. 100364 - 100364

Published: Dec. 1, 2024

Language: Английский

Citations

1
Enhancing experimental image quality in two-phase bubbly systems with super-resolution using generative adversarial networks DOI Creative Commons
Manoel Neves,

João Filgueiras,

Zafeiris Kokkinogenis

et al.

International Journal of Multiphase Flow, Journal Year: 2024, Volume and Issue: 180, P. 104952 - 104952

Published: Aug. 5, 2024

Fluid Dynamics is a key scientific field to multitudes of engineering applications. Experimental work in this requires careful set-up and expensive image-capturing equipment, particularly when considering the finer details complex phenomena. In work, we study application super-resolution Generative Adversarial Networks (GANs) achieve high-resolution results by upscaling lower-resolution experimental images. We train GANs proposed for natural images on bubbly flow dataset compare common evaluation metrics domain expert assessments upscaled find that these models promising results, as evaluated experts, transfer learning from translates better performance overall. Attention mechanisms are found be useful recreating sharper details. On other hand, traditional align poorly with perception quality, signaling need systematic methodologies domain.

Language: Английский

Citations

0
Resolution enhancement and segmentation of polar bubbly ice micro CT scans via 3D convolutional neural network DOI Creative Commons
Faramarz Bagherzadeh, Johannes Freitag, Udo Frese

et al.

Applied Computing and Geosciences, Journal Year: 2024, Volume and Issue: 23, P. 100193 - 100193

Published: Sept. 1, 2024

Language: Английский

Citations

0
Geostatistics and artificial intelligence coupling: advanced machine learning neural network regressor for experimental variogram modelling using Bayesian optimization DOI Creative Commons
Saâd Soulaimani, Ayoub Soulaimani, Kamal Abdelrahman

et al.

Frontiers in Earth Science, Journal Year: 2024, Volume and Issue: 12

Published: Dec. 12, 2024

Experimental variogram modelling is an essential process in geostatistics. The use of artificial intelligence (AI) a new and advanced way automating experimental modelling. One part this AI approach the population search algorithms to fine-tune hyperparameters for better prediction performing. We Bayesian optimization first time find optimal learning parameters more precise neural network regressor goal leverage capability consider previous regression results improve output using three variograms as inputs one training, calculated from ore grades four orebodies, characterised by same genetic aspect. In comparison architectures, Bayesian-optimized demonstrably achieved superior Coefficient determination validation 78.36%. This significantly outperformed non-optimized wide, bilayer, tri-layer configurations, which yielded 32.94%, 14.00%, −46.03% determination, respectively. improved reliability demonstrates its superiority over traditional, regressors, indicating that incorporating can advance modelling, thus offering accurate intelligent solution, combining geostatistics specifically machine

Language: Английский

Citations

0
Deep Super Resolution Techniques for Remote Sensing Big Data: A Comparative Study DOI
Rishabh Jain, Ranga Raju Vatsavai

2021 IEEE International Conference on Big Data (Big Data), Journal Year: 2024, Volume and Issue: unknown, P. 6011 - 6020

Published: Dec. 15, 2024

Language: Английский

Citations

0
Geosystems Risk and Uncertainty: The Application of Chatgpt with Targeted Prompting DOI

Seyed Kourosh Mahjour,

Ramin Soltanmohammadi, Ehsan Heidaryan

et al.

Published: Jan. 1, 2024

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Language: Английский

Citations

0
Enhancing Experimental Image Quality in Two-Phase Bubbly Systems with Super-Resolution Using Generative Adversarial Networks DOI
Miguel M. Neves,

João Filgueiras,

Zafeiris Kokkinogenis

et al.

Published: Jan. 1, 2024

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Language: Английский

Citations

0