Guangyu Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Guangyu Wang has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Xinming Wu
Laboratory of Seismology and Physics of the Earth’s Interior, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, 230026, China
9 shared publications - Xiaodong Ma
University of Science and Technology of China
3 shared publications Xiaoming Sun
State Key Laboratory of Precision Geodesy, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, China
3 shared publicationsHang Gao
School of Earth and Space Sciences, University of Science and Technology of China, Hefei, China
2 shared publicationsHanlin Sheng
School of Earth and Space Sciences, University of Science and Technology of China, Hefei, China
2 shared publications- Hui Gao
Laboratory of Seismology and Physics of the Earth’s Interior, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, 230026, China
2 shared publications
Latest Publications
- Open Access
- 14 July 2026
- Published by Elsevier in International Journal of Rock Mechanics and Mining Sciences
- Vol. 206
Continuous spatial stress variations within a granitic rock mass: Revealed by cross-sectional ellipticity of an array of boreholes
- 14 July 2026
- Published by Elsevier in International Journal of Rock Mechanics and Mining Sciences
- Vol. 206
Open AccessContinuous spatial stress variations within a granitic rock mass: Revealed by cross-sectional ellipticity of an array of boreholes
- 23 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23 (1545598X) , 1-5
A Multiclass Training Dataset and Hybrid Neural Network for Simultaneous Karst and Channel Detection
Seismic karsts and channels interpretation is vital for reservoir characterization. These geobodies often coexist in seismic volumes and exhibit similar reflection features, making them challenging to distinguish using existing methods that typically interpret them separately. We propose a multi-class hybrid neural network for simultaneously detecting karsts and channels, offering more efficient and accurate results than independent detection. To address the lack of multi-class labeled data, we developed a workflow to generate a diverse synthetic training dataset. This involves constructing 3D impedance models characterized by stratigraphic sequences with karsts and channels embedded according to geological and empirical data. To simulate diverse models, key parameters such as velocities, dimensions, and orientations of geobodies are randomly assigned within reasonable ranges based on geological context. Reflectivity models are computed from impedance models and convolved with Ricker wavelets to simulate seismic data, with noise added to enhance realism, yielding diverse seismic volumes with multi-class labels. This dataset is used to train a U-shaped hybrid neural network combining a Swin-Transformer encoder with a residual module decoder. The Swin-transformer provides global context-awareness and captures long-range dependencies, while the decoder ensures detailed feature restoration. The network performs multi-class segmentation, simultaneously detecting and distinguishing karsts and channels. Field applications demonstrate that the trained model can detect these geobodies with high accuracy and integrity. From the detection, 3D meshes of the geobodies are constructed to model and analyze their structural geometries. We have made our multi-class training dataset publicly available (https://zenodo.org/records/10781510) for further research on karst and channel interpretation.
- 23 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23 (1545598X) , 1-5
A Multiclass Training Dataset and Hybrid Neural Network for Simultaneous Karst and Channel Detection
Seismic karsts and channels interpretation is vital for reservoir characterization. These geobodies often coexist in seismic volumes and exhibit similar reflection features, making them challenging to distinguish using existing methods that typically interpret them separately. We propose a multi-class hybrid neural network for simultaneously detecting karsts and channels, offering more efficient and accurate results than independent detection. To address the lack of multi-class labeled data, we developed a workflow to generate a diverse synthetic training dataset. This involves constructing 3D impedance models characterized by stratigraphic sequences with karsts and channels embedded according to geological and empirical data. To simulate diverse models, key parameters such as velocities, dimensions, and orientations of geobodies are randomly assigned within reasonable ranges based on geological context. Reflectivity models are computed from impedance models and convolved with Ricker wavelets to simulate seismic data, with noise added to enhance realism, yielding diverse seismic volumes with multi-class labels. This dataset is used to train a U-shaped hybrid neural network combining a Swin-Transformer encoder with a residual module decoder. The Swin-transformer provides global context-awareness and captures long-range dependencies, while the decoder ensures detailed feature restoration. The network performs multi-class segmentation, simultaneously detecting and distinguishing karsts and channels. Field applications demonstrate that the trained model can detect these geobodies with high accuracy and integrity. From the detection, 3D meshes of the geobodies are constructed to model and analyze their structural geometries. We have made our multi-class training dataset publicly available (https://zenodo.org/records/10781510) for further research on karst and channel interpretation.
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 2152-2166
High-Fidelity Seismic Super-Resolution Using Prior-Informed Deep Learning With 3D Awareness
The limitations of seismic vertical resolution pose significant challenges for the identification of thin beds. Improving the vertical resolution of seismic data using deep learning methods often encounters challenges related to unrealistic outputs and limited generalization. To address these challenges, we propose a novel framework that improves the fidelity and generalization of seismic super-resolution. Our approach begins with the generation of realistic synthetic training data that aligns with the structural and amplitude characteristics of field surveys. We then introduce an enhanced 2D network with 3D awareness, which builds on the 2D Swin-Transformer and 3D convolution blocks to effectively capture 3D spatial features while maintaining computational efficiency. This network addresses the limitations of traditional 2D approaches by reducing stitching artifacts and improving spatial consistency. Finally, we develop a prior-informed fine-tuning strategy using field data without the need for labels, which incorporates a self-supervised data consistency loss and a spectral matching loss based on prior knowledge. This strategy ensures that the super-resolution results preserve the original low frequency information while yielding a spectral distribution as expected. Experiments on multiple field datasets demonstrate the robustness and generalization capability of our method, making it a practical solution for seismic resolution enhancement in diverse field datasets.
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 2152-2166
High-Fidelity Seismic Super-Resolution Using Prior-Informed Deep Learning With 3D Awareness
The limitations of seismic vertical resolution pose significant challenges for the identification of thin beds. Improving the vertical resolution of seismic data using deep learning methods often encounters challenges related to unrealistic outputs and limited generalization. To address these challenges, we propose a novel framework that improves the fidelity and generalization of seismic super-resolution. Our approach begins with the generation of realistic synthetic training data that aligns with the structural and amplitude characteristics of field surveys. We then introduce an enhanced 2D network with 3D awareness, which builds on the 2D Swin-Transformer and 3D convolution blocks to effectively capture 3D spatial features while maintaining computational efficiency. This network addresses the limitations of traditional 2D approaches by reducing stitching artifacts and improving spatial consistency. Finally, we develop a prior-informed fine-tuning strategy using field data without the need for labels, which incorporates a self-supervised data consistency loss and a spectral matching loss based on prior knowledge. This strategy ensures that the super-resolution results preserve the original low frequency information while yielding a spectral distribution as expected. Experiments on multiple field datasets demonstrate the robustness and generalization capability of our method, making it a practical solution for seismic resolution enhancement in diverse field datasets.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64, 4507721
Breakout-Picker: Reducing False Positives in Deep Learning-Based Borehole Breakout Characterization From Acoustic Image Logs
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64, 4507721
Open AccessBreakout-Picker: Reducing False Positives in Deep Learning-Based Borehole Breakout Characterization From Acoustic Image Logs
- Open Access
- 2 August 2025
- Published by Wiley
Continuous Spatial Stress Variations within a Granitic Rock Mass: Revealed by Cross-sectional Ellipticity of an Array of Boreholes
Understanding the spatial variations of in-situ stress is crucial for optimizing subsurface resource development and mitigating geological hazards. However, due to the spatial sparsity of conventional stress measurement methods (e.g., hydraulic fracturing and overcoring), critical stress variations might be overlooked. To enhance the continuity of stress profiling, we utilize borehole cross-sectional ellipticity resulting from stress-induced borehole deformation to obtain continuous and abundant stress information. We developed a workflow to extract borehole cross-sectional ellipticity from the acoustic televiewer (ATV) logs, which provide high-resolution measurements of borehole cross-sectional geometry. Based on a borehole array in the Bedretto Underground Laboratory in Switzerland, we characterized the stress variations within a hectometer-scale granitic rock mass using borehole cross-sectional ellipticity. We found that each borehole exhibited significant variations in cross-sectional ellipticity along its depth, indicating strong heterogeneity of the stress field. Subsequently, we employed a grid search algorithm to invert the continuous stress variations along each borehole based on its cross-sectional ellipticity. The results indicate that the rock mass is generally in a normal faulting stress regime, but the stress orientations and relative stress magnitudes vary significantly. The causes of such stress variations could be related to local stress concentrations caused by fractures or stress perturbations resulting from fault slip. Our work provides a quantitative characterization of continuous stress variations within rock masses without direct stress measurements, which is beneficial to various geoscientific and subsurface engineering applications.
- 2 August 2025
- Published by Wiley
Open AccessContinuous Spatial Stress Variations within a Granitic Rock Mass: Revealed by Cross-sectional Ellipticity of an Array of Boreholes
Understanding the spatial variations of in-situ stress is crucial for optimizing subsurface resource development and mitigating geological hazards. However, due to the spatial sparsity of conventional stress measurement methods (e.g., hydraulic fracturing and overcoring), critical stress variations might be overlooked. To enhance the continuity of stress profiling, we utilize borehole cross-sectional ellipticity resulting from stress-induced borehole deformation to obtain continuous and abundant stress information. We developed a workflow to extract borehole cross-sectional ellipticity from the acoustic televiewer (ATV) logs, which provide high-resolution measurements of borehole cross-sectional geometry. Based on a borehole array in the Bedretto Underground Laboratory in Switzerland, we characterized the stress variations within a hectometer-scale granitic rock mass using borehole cross-sectional ellipticity. We found that each borehole exhibited significant variations in cross-sectional ellipticity along its depth, indicating strong heterogeneity of the stress field. Subsequently, we employed a grid search algorithm to invert the continuous stress variations along each borehole based on its cross-sectional ellipticity. The results indicate that the rock mass is generally in a normal faulting stress regime, but the stress orientations and relative stress magnitudes vary significantly. The causes of such stress variations could be related to local stress concentrations caused by fractures or stress perturbations resulting from fault slip. Our work provides a quantitative characterization of continuous stress variations within rock masses without direct stress measurements, which is beneficial to various geoscientific and subsurface engineering applications.
- Open Access
- 21 July 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (7) , 3447-3471
cigChannel: a large-scale 3D seismic dataset with labeled paleochannels for advancing deep learning in seismic interpretation
Identifying paleochannels in 3D seismic volumes (seismic paleochannel interpretation) is essential for georesource development and offering insights into paleoclimate conditions. However, it remains a labor-intensive and time-consuming task. Deep learning has shown great promise in automating seismic paleochannel interpretation with high efficiency and accuracy, as demonstrated in similar image segmentation tasks in computer vision (CV). Yet, unlike the CV domain, seismic exploration lacks a comprehensive labeled dataset for paleochannels, significantly hindering the development, application, and evaluation of deep learning methods in this field. Manual labeling of paleochannels in 3D seismic volumes is tedious and subjective, potentially leading to mislabeling that degrades deep learning model's performance. To address this, we propose a workflow to generate a synthetic seismic dataset, cigChannel, consisting of 1600 seismic volumes with over 10 000 labeled paleochannels. Each volume has a size of samples. This is the largest dataset to date for seismic paleochannel interpretation, featuring geologically reasonable seismic volumes with accurately labeled meandering channels, tributary channel networks, and submarine canyons. A convolutional neural network (simplified from U-Net) trained on this dataset achieves F1 scores of 0.52, 0.73, and 0.63 in identifying meandering channels, tributary channel networks, and submarine canyons in three field seismic volumes, respectively. However, the synthetic seismic volumes in the cigChannel dataset still lack the variability and realism of field seismic data, potentially affecting the deep learning model's generalizability. To facilitate further research, we publicly release the dataset ( Wang et al., 2024 , https://doi.org/10.5281/zenodo.10791151), data generation codes, and trained U-Net model ( Wang, 2024 , https://github.com/wanggy-1/cigChannel, last access: 5 July 2025), aiming to advance deep learning approaches for seismic paleochannel interpretation.- 21 July 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (7) , 3447-3471
Open AccesscigChannel: a large-scale 3D seismic dataset with labeled paleochannels for advancing deep learning in seismic interpretation
Identifying paleochannels in 3D seismic volumes (seismic paleochannel interpretation) is essential for georesource development and offering insights into paleoclimate conditions. However, it remains a labor-intensive and time-consuming task. Deep learning has shown great promise in automating seismic paleochannel interpretation with high efficiency and accuracy, as demonstrated in similar image segmentation tasks in computer vision (CV). Yet, unlike the CV domain, seismic exploration lacks a comprehensive labeled dataset for paleochannels, significantly hindering the development, application, and evaluation of deep learning methods in this field. Manual labeling of paleochannels in 3D seismic volumes is tedious and subjective, potentially leading to mislabeling that degrades deep learning model's performance. To address this, we propose a workflow to generate a synthetic seismic dataset, cigChannel, consisting of 1600 seismic volumes with over 10 000 labeled paleochannels. Each volume has a size of samples. This is the largest dataset to date for seismic paleochannel interpretation, featuring geologically reasonable seismic volumes with accurately labeled meandering channels, tributary channel networks, and submarine canyons. A convolutional neural network (simplified from U-Net) trained on this dataset achieves F1 scores of 0.52, 0.73, and 0.63 in identifying meandering channels, tributary channel networks, and submarine canyons in three field seismic volumes, respectively. However, the synthetic seismic volumes in the cigChannel dataset still lack the variability and realism of field seismic data, potentially affecting the deep learning model's generalizability. To facilitate further research, we publicly release the dataset ( Wang et al., 2024 , https://doi.org/10.5281/zenodo.10791151), data generation codes, and trained U-Net model ( Wang, 2024 , https://github.com/wanggy-1/cigChannel, last access: 5 July 2025), aiming to advance deep learning approaches for seismic paleochannel interpretation. - Open Access
- 10 February 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (2) , 595-609
cigFacies: a massive-scale benchmark dataset of seismic facies and its application
Seismic facies classification is crucial for seismic stratigraphic interpretation and hydrocarbon reservoir characterization but remains a tedious and time-consuming task that requires significant manual effort. Data-driven deep-learning approaches are highly promising for automating the seismic facies classification with high efficiency and accuracy, as they have already achieved significant success in similar image classification tasks within the field of computer vision (CV). However, unlike the CV domain, the field of seismic exploration lacks a comprehensive benchmark dataset for seismic facies, severely limiting the development, application, and evaluation of deep-learning approaches in seismic facies classification. To address this gap, we propose a comprehensive workflow to construct a massive-scale benchmark dataset of seismic facies and evaluate its effectiveness in training a deep-learning model. Specifically, we first develop a knowledge graph of seismic facies based on geological concepts and seismic reflection configurations. Guided by the graph, we then implement the three strategies of field seismic data curation, knowledge-guided synthesization, and generative adversarial network (GAN)-based generation to construct a benchmark dataset of 8000 diverse samples for five common seismic facies. Finally, we use the benchmark dataset to train a network and then apply it to two 3-D seismic data for automatic seismic facies classification. The predictions are highly consistent with expert interpretation results, demonstrating that the diversity and representativeness of our benchmark dataset are sufficient to train a network that can be generalized well in seismic facies classification across field data. We have made this dataset (https://doi.org/10.5281/zenodo.10777460, Gao et al., 2024a), the trained model, and the associated codes (https://doi.org/10.5281/zenodo.13150879, Gao et al., 2024b) publicly available for further research and validation of intelligent seismic facies classification.
- 10 February 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (2) , 595-609
Open AccesscigFacies: a massive-scale benchmark dataset of seismic facies and its application
Seismic facies classification is crucial for seismic stratigraphic interpretation and hydrocarbon reservoir characterization but remains a tedious and time-consuming task that requires significant manual effort. Data-driven deep-learning approaches are highly promising for automating the seismic facies classification with high efficiency and accuracy, as they have already achieved significant success in similar image classification tasks within the field of computer vision (CV). However, unlike the CV domain, the field of seismic exploration lacks a comprehensive benchmark dataset for seismic facies, severely limiting the development, application, and evaluation of deep-learning approaches in seismic facies classification. To address this gap, we propose a comprehensive workflow to construct a massive-scale benchmark dataset of seismic facies and evaluate its effectiveness in training a deep-learning model. Specifically, we first develop a knowledge graph of seismic facies based on geological concepts and seismic reflection configurations. Guided by the graph, we then implement the three strategies of field seismic data curation, knowledge-guided synthesization, and generative adversarial network (GAN)-based generation to construct a benchmark dataset of 8000 diverse samples for five common seismic facies. Finally, we use the benchmark dataset to train a network and then apply it to two 3-D seismic data for automatic seismic facies classification. The predictions are highly consistent with expert interpretation results, demonstrating that the diversity and representativeness of our benchmark dataset are sufficient to train a network that can be generalized well in seismic facies classification across field data. We have made this dataset (https://doi.org/10.5281/zenodo.10777460, Gao et al., 2024a), the trained model, and the associated codes (https://doi.org/10.5281/zenodo.13150879, Gao et al., 2024b) publicly available for further research and validation of intelligent seismic facies classification.
- Open Access
- 7 October 2024
- Published by Copernicus GmbH in EGUsphere
- Vol. 2024, 1-19
cigFacies: a massive-scale benchmark dataset of seismic facies and its application
Seismic facies classification is crucial for seismic stratigraphic interpretation and hydrocarbon reservoir characterization but remains a tedious and time-consuming task that requires significant manual effort. The data-driven deep learning approaches are highly promising to automate the seismic facies classification with high efficiency and accuracy, as they have already achieved significant success in similar image classification tasks within the field of computer vision (CV). However, unlike the CV domain, the field of seismic exploration lacks a comprehensive benchmark dataset for seismic facies, severely limiting the development, application, and evaluation of deep learning approaches in seismic facies classification. To address this gap, we propose a comprehensive workflow to construct a massive-scale benchmark dataset of seismic facies and evaluate its effectiveness in training a deep learning model. Specifically, we first develop a knowledge graph of seismic facies based on the geological concepts and seismic reflection configurations. Guided by the graph, we then implement three strategies of field seismic data curation, knowledge-guided synthesization, and GAN-based generation to construct a benchmark dataset of 8000 diverse samples for five common seismic facies. Finally, we use the benchmark dataset to train a network and then apply it on two 3-D seismic data for automatic seismic facies classification. The predictions are highly consistent with expert interpretation results, demonstrating the diversity and representativeness of our benchmark dataset is sufficient to train a network that can generalize well in seismic facies classification across field data. We have made this dataset, the trained model and associated codes publicly available for further research and validation of intelligent seismic facies classification.
- 7 October 2024
- Published by Copernicus GmbH in EGUsphere
- Vol. 2024, 1-19
Open AccesscigFacies: a massive-scale benchmark dataset of seismic facies and its application
Seismic facies classification is crucial for seismic stratigraphic interpretation and hydrocarbon reservoir characterization but remains a tedious and time-consuming task that requires significant manual effort. The data-driven deep learning approaches are highly promising to automate the seismic facies classification with high efficiency and accuracy, as they have already achieved significant success in similar image classification tasks within the field of computer vision (CV). However, unlike the CV domain, the field of seismic exploration lacks a comprehensive benchmark dataset for seismic facies, severely limiting the development, application, and evaluation of deep learning approaches in seismic facies classification. To address this gap, we propose a comprehensive workflow to construct a massive-scale benchmark dataset of seismic facies and evaluate its effectiveness in training a deep learning model. Specifically, we first develop a knowledge graph of seismic facies based on the geological concepts and seismic reflection configurations. Guided by the graph, we then implement three strategies of field seismic data curation, knowledge-guided synthesization, and GAN-based generation to construct a benchmark dataset of 8000 diverse samples for five common seismic facies. Finally, we use the benchmark dataset to train a network and then apply it on two 3-D seismic data for automatic seismic facies classification. The predictions are highly consistent with expert interpretation results, demonstrating the diversity and representativeness of our benchmark dataset is sufficient to train a network that can generalize well in seismic facies classification across field data. We have made this dataset, the trained model and associated codes publicly available for further research and validation of intelligent seismic facies classification.
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-15
Enhancing Deep Learning for Seismic Interpretation With Constraints From Seismic Attributes: A Fault Detection Case Study
Deep learning (DL) models can exhibit powerful feature extraction ability once trained on massive, diverse labeled datasets. However, such datasets are typically lacking in geophysics, limiting the performance of DL models on solving geophysical problems. The high complexity and diversity of field data in the inference step further constrains the generalizability of DL models trained on only limited data. To address these challenges, incorporating geophysical knowledge as constraints has proven effective. In seismic interpretation, seismic attributes provide quantified geophysical insights into subsurface situation and can be used as constraints to enhance the performance of DL models. We consider the geophysically meaningful attributes as feature channels and propose optimal ways to select and incorporate suitable seismic attributes as constraints on DL models, with a focus on seismic fault interpretation. Specifically, we utilize U-Net as the DL architecture and explore three different ways of incorporating seismic attributes at the input, encoder, and decoder stages. Extensive field data applications show that seismic attributes, as features of seismic data carrying prior information, can compensate for the limitations of DL models caused by the lack of training datasets and the diversity of inference data in geophysics. This is reflected in improved accuracy, generalizability, and geological consistency of predictions by DL models with constraints. It is also discovered that simplified DL models with seismic attribute constraints outperform the complex DL models without such constraints. This further demonstrates that incorporating seismic attributes as constraints allows for the simplification of DL models while improving the prediction accuracy.
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-15
Enhancing Deep Learning for Seismic Interpretation With Constraints From Seismic Attributes: A Fault Detection Case Study
Deep learning (DL) models can exhibit powerful feature extraction ability once trained on massive, diverse labeled datasets. However, such datasets are typically lacking in geophysics, limiting the performance of DL models on solving geophysical problems. The high complexity and diversity of field data in the inference step further constrains the generalizability of DL models trained on only limited data. To address these challenges, incorporating geophysical knowledge as constraints has proven effective. In seismic interpretation, seismic attributes provide quantified geophysical insights into subsurface situation and can be used as constraints to enhance the performance of DL models. We consider the geophysically meaningful attributes as feature channels and propose optimal ways to select and incorporate suitable seismic attributes as constraints on DL models, with a focus on seismic fault interpretation. Specifically, we utilize U-Net as the DL architecture and explore three different ways of incorporating seismic attributes at the input, encoder, and decoder stages. Extensive field data applications show that seismic attributes, as features of seismic data carrying prior information, can compensate for the limitations of DL models caused by the lack of training datasets and the diversity of inference data in geophysics. This is reflected in improved accuracy, generalizability, and geological consistency of predictions by DL models with constraints. It is also discovered that simplified DL models with seismic attribute constraints outperform the complex DL models without such constraints. This further demonstrates that incorporating seismic attributes as constraints allows for the simplification of DL models while improving the prediction accuracy.
- 27 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-13
Completing Any Borehole Images
Borehole images contain the physical information and chemical properties of geological formations, which are crucial for high-resolution interpretation of subsurface stratigraphic and structural features and geological modeling of subsurface. However, due to the special design of borehole tools and variations in borehole diameter, all kinds of borehole images (FMI, Earth-Imager, OMRI, and OBMI) obtained from scanning the borehole walls exhibit varying degrees of data missing, with OBMI data missing up to 70%. We propose a deep-learning approach with a hybrid CNN and Transformer architecture to fill in the gaps in borehole images, addressing the challenges of missing training labels and filling large-scale gaps. To solve the challenge of missing labels of complete borehole images, our deep learning model is pre-trained on a vast collection of complete natural and seismic images, then fine-tuned with a partial loss function on incomplete borehole images. A multistage completion strategy is further introduced into the inference stage to enhance the continuity and textural features of the completed areas. In addition, by incorporating the circular consistency constraint between the left and right sides of the borehole image, our method is able to reasonably complete the gaps with highly consistent features on both sides of the image. During the tests on borehole images from multiple wells in different work areas with various geological features, our model is capable of completing any type of borehole images with masks of any size, ultimately yielding complete images free of any artifacts, while also possessing richer and more reasonable textures and semantic information. We have opensourced the code and the fine-tuned models, which are available at https://github.com/zgyustc/LogMAT/tree/master.
- 27 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-13
Completing Any Borehole Images
Borehole images contain the physical information and chemical properties of geological formations, which are crucial for high-resolution interpretation of subsurface stratigraphic and structural features and geological modeling of subsurface. However, due to the special design of borehole tools and variations in borehole diameter, all kinds of borehole images (FMI, Earth-Imager, OMRI, and OBMI) obtained from scanning the borehole walls exhibit varying degrees of data missing, with OBMI data missing up to 70%. We propose a deep-learning approach with a hybrid CNN and Transformer architecture to fill in the gaps in borehole images, addressing the challenges of missing training labels and filling large-scale gaps. To solve the challenge of missing labels of complete borehole images, our deep learning model is pre-trained on a vast collection of complete natural and seismic images, then fine-tuned with a partial loss function on incomplete borehole images. A multistage completion strategy is further introduced into the inference stage to enhance the continuity and textural features of the completed areas. In addition, by incorporating the circular consistency constraint between the left and right sides of the borehole image, our method is able to reasonably complete the gaps with highly consistent features on both sides of the image. During the tests on borehole images from multiple wells in different work areas with various geological features, our model is capable of completing any type of borehole images with masks of any size, ultimately yielding complete images free of any artifacts, while also possessing richer and more reasonable textures and semantic information. We have opensourced the code and the fine-tuned models, which are available at https://github.com/zgyustc/LogMAT/tree/master.