Seismic Facies Segmentation Using Deep Learning; #42286 (2018)

Daniel Chevitarese, Daniela Szwarcman, Reinaldo Mozart D. Silva, Emilio Vital Brazil

Search and Discovery.com

... the input image into tiles of size 80×120, as the model would increase significantly in the number of parameters and in training time. Results...

Abstract: Color Correction for Gabor Deconvolution and Nonstationary Phase Rotation; #90171 (2013)

Peng Cheng and Gary F. Margrave

Search and Discovery.com

... deconvolution is based on a nonstationary convolution model of the seismic trace. Margrave (1998) presented a nonstationary convolutional model, which...

Abstract: Color Correction for Gabor Deconvolution and Nonstationary Phase Rotation; #90171 (2013)

Peng Cheng and Gary F. Margrave

Search and Discovery.com

... deconvolution is based on a nonstationary convolution model of the seismic trace. Margrave (1998) presented a nonstationary convolutional model, which...

Fast self-supervised learning for reconstruction of 3D seismic data

Yinshuo Li, Wenkai Lu, We Cao

International Meeting for Applied Geoscience and Energy (IMAGE)

...% missing traces. The training time of DL-based methods is limited. (a) Fragment of 3D prestack data from Sandia/SEG Salt Model 45 shot subset, which...

Convolution Neural Networks … If They can Identify an Oncoming Car, can They Identify Lithofacies in Core?; #42312 (2018)

Rafael Pires de Lima, Fnu Suriamin, Kurt Marfurt, Matthew Pranter, Gerilyn Soreghan

Search and Discovery.com

... drive our cars but also taste our beer. Specifically, recent advances in the architecture of deep-learning convolutional neural networks (CNN) have...

Intelligent Prediction of Shale Oil Fracturing Curves Based on A Sequence-to-Sequence Model

Leyi Zheng, Tianbo Liang, Yunjin Wang, Fujian Zhou, Junlin Wu, Bin Wang, Jiaming Zhang, Maoqin Yang, Gong Chen, Xingyuan Liang

Unconventional Resources Technology Conference (URTEC)

... convolutional structure (a), and the internal structure of the LSTM unit (b) The model constructed in this study establishes a multivariate time-series mapping...

AI to Improve the Reliability and Reproducibility of Descriptive Data: A Case Study Using Convolutional Neural Networks to Recognize Carbonate Facies in Cores

Search and Discovery.com

-- no title --

user1

Search and Discovery.com

... Lithotypes Classification with Convolutional Neural Networks Evgeny E. Baraboshkin1, Evdokiya A. Panchenko2, Andrey E. Demidov1, Ardiansyah...

Seismic inversion with dictionary learning using unsupervised machine learning

Debajeet Barman, Mrinal K. Sen

International Meeting for Applied Geoscience and Energy (IMAGE)

... number of time samples is 200, with a sampling interval of 2ms. We use a synthetic model which is advantageous over a synthetically generated model because...

Enhancing seismic image resolution using Brownian diffusion bridge model

Bingbing Sun, Abdulmoshen M. Ali, Tariq Alkhalifah

International Meeting for Applied Geoscience and Energy (IMAGE)

...Enhancing seismic image resolution using Brownian diffusion bridge model Bingbing Sun, Abdulmoshen M. Ali, Tariq Alkhalifah Enhancing seismic image...

Extrapolated surface-wave dispersion inversion

Hongyu Sun, Laurent Demanet

International Meeting for Applied Geoscience and Energy (IMAGE)

..., T., A. Abubakar, X. Cheng, and L. Fu, 2020, Augment time-domain FWI with iterative deep learning: 90th Annual International Meeting, SEG, Expanded...

Abstract: GAN-Based Multipoint Geostatistical Inversion Method and Application;

Pengfei Xie, Jiagen Hou

Search and Discovery.com

... technology. Multi-point statistics (MPS) generate model realizations by training image (TI) that are consistent with prior information. This method often...

An integrated workflow for deep learning-accelerated seismic modelling of the Groningen gas field, the Netherlands

Haibin Di, Vanessa Simoes, Zhun Li, Cen Li, Anisha Kaul, Aria Abubakar

International Meeting for Applied Geoscience and Energy (IMAGE)

... exploration and development, build accurate subsurface models is a complicated and time-consuming process, which usually requires intradisciplinary...

Deep learning to maximize the value of fast-track 4D seismic processing

Arnab Dhara, Haron Abdel-Raziq, Denis Kiyashchenko, Asiya Kudarova, Janaki Vamaraju, Albena Mateeva, Pandu Devarakota, Kanglin Wang, Jorge Lopez

International Meeting for Applied Geoscience and Energy (IMAGE)

... Shell Global Solutions (US) Inc., 2Shell International Exploration and Production Inc.,3Shell Brasil Petróleo Ltda Summary Time lapse seismic monitoring...

Marchenko focusing using convolutional neural networks

Mert S. R. Kiraz, Roel Snieder

International Meeting for Applied Geoscience and Energy (IMAGE)

...Marchenko focusing using convolutional neural networks Mert S. R. Kiraz, Roel Snieder Marchenko focusing using convolutional neural networks Mert...

Subsurface hydrogen storage: A feasibility study

Dwaipayan Chakraborty, Subhashis Mallick, Mortezaa Dejam

International Meeting for Applied Geoscience and Energy (IMAGE)

.... (a), (b) same as Fig. 2 but in time domain. (c) Angle domain synthetic seismic response before and (d) after H2 injection. (e) Difference between (c...

The impact of the synthetic seismic data generation method on automated AI-based horizon interpretation

F. Vizeu, J. Zambrini, A. Canning

International Meeting for Applied Geoscience and Energy (IMAGE)

... by using the convolutional model with full control of the synthetic wavelet, and add noise to it. To convert the 2D data into 3D we use a technique...

A hybrid machine learning model for improving regression of mineral composition estimation using well logging data

Xiaojun Liu, Kezhen Hu, Stephen E. Grasby, Benjamin Lee

International Meeting for Applied Geoscience and Energy (IMAGE)

... classification problems. This ConvXGB architecture consists of a network with several stacked convolutional layers and XGBoost as the last layer of the model...

Introduction to Special Issue: Geoscience Data Analytics and Machine Learning

Michael J. Pyrcz

AAPG Bulletin

... computational resources and the development of new algorithms, this is an exciting time for data-driven geoscience. For example, convolutional neural networks...

A deep learning-based inverse Hessian for full-waveform inversion

Mustafa Alfarhan, Matteo Ravasi, Tariq Alkhalifah

International Meeting for Applied Geoscience and Energy (IMAGE)

... continuation: Geophysics, 81, no. 3, R89–R98, doi: https://doi.org/10.1190/geo2015-0537.1. AlTheyab, A., X. Wang, and G. T. Schuster, 2013, Time-domain...

An automatic velocity picking method based on object detection

Ce Bian, Weifeng Geng, Ping Yang, Pengyuan Sun, Guiren Xue, Haikun Lin

International Meeting for Applied Geoscience and Energy (IMAGE)

... an automatic velocity spectrum picking method based on object detection, and applies neural network model named FCOS (Fully Convolutional OneStage Object...

Machine-Learning-Assisted Segmentation of FIB-SEM Images with Artifacts for Improved of Pore Space Characterization of Tight Reservoir Rocks

Andrey Kazak, Kirill Simonov, Victor Kulikov

Unconventional Resources Technology Conference (URTEC)

... on a convolutional neural network (CNN) in the DeepUnet configuration. The implementation utilized the Pytorch framework in a Linux environment...

InvMixer An efficient deep neural network for seismic inversion

Tianyi Zhang, Mauricio Araya-Polo, Anshumali Shrivastava

International Meeting for Applied Geoscience and Energy (IMAGE)

... layers in UNet use learnable kernels with of size 3×3 or 5×5 to model the relationships between traces. Since a single convolutional layer has limited...

CO2 Plume Imaging with Accelerated Deep Learning-based Data Assimilation Using Distributed Pressure and Temperature Measurements at the Illinois Basin-Decatur Carbon Sequestration Project

Takuto Sakai, Masahiro Nagao, Chin Hsiang Chan, Akhil Datta-Gupta

Carbon Capture, Utilization and Storage (CCUS)

... the diffusive time of flight (DTOF) map as a representative reservoir image of the flow field. Reservoir model calibration can be implemented by selecting...

A denoising diffusion probabilistic modeling (DDPM) approach for predicting CO2 plume evolution from seismic shot gathers

Alexander Y. Sun, Zi Xian Leong, Tieyuan Zhu

International Meeting for Applied Geoscience and Energy (IMAGE)

... experiments was used. At inference time, our model predicts CO2 saturation distributions using seismic gathers and prediction times as conditions. Unlike...