Showing papers in "Geophysics in 2020"
TL;DR: The proposed algorithm to attenuate random noise based on a deep-denoising autoencoder (DDAE) succeeds in attenuating the random noise in an effective manner and is compared with several benchmark algorithms.
Abstract: Attenuation of seismic random noise is considered an important processing step to enhance the signal-to-noise ratio of seismic data. A new approach is proposed to attenuate random noise bas...
155 citations
TL;DR: A theory-designed recurrent neural network (RNN) that allows single- and multidimensional scalar acoustic seismic forward-modeling problems to be set up in terms of its forward propagation is developed, finding that training such a network amounts to a solution of the seismic inverse problem and is equivalent to gradient-based seismic full-waveform inversion (FWI).
Abstract: Deep-learning techniques appear to be poised to play very important roles in our processing flows for inversion and interpretation of seismic data. The most successful seismic applications ...
117 citations
TL;DR: A workflow to automatically build diverse structure models with realistic folding and faulting features and simulates realistic and generalized structure models from which the CNNs effectively learn to recognize real structures in field images is developed.
Abstract: Seismic structural interpretation involves highlighting and extracting faults and horizons that are apparent as geometric features in a seismic image. Although seismic image processing meth...
109 citations
TL;DR: A convolutional neural network is developed to estimate stacking velocities directly from the semblance to predict a consistent velocity model and adopts transfer learning to update the trained model with a small portion of the target data to improve the accuracy of the predicted velocity model.
Abstract: Velocity analysis can be a time-consuming task when performed manually. Methods have been proposed to automate the process of velocity analysis, which, however, typically requires significa...
86 citations
TL;DR: The lack of low-frequency information and a good initial model can seriously affect the success of full waveform inversion (FWI) due to the inherent cycle skipping problem as discussed by the authors.
Abstract: The lack of low-frequency information and a good initial model can seriously affect the success of full-waveform inversion (FWI), due to the inherent cycle skipping problem. Computational l...
84 citations
TL;DR: In this article, the mapping of seismic and lithologic facies from 3D reflection seismic data plays a key role in depositional environment analysis and reservoir characterization during hydrocarbon exploration.
Abstract: Mapping of seismic and lithologic facies from 3D reflection seismic data plays a key role in depositional environment analysis and reservoir characterization during hydrocarbon exploration ...
75 citations
TL;DR: This work has performed seismic trace interpolation by using the convolutional autoencoder (CAE) to solve the problem of rare complete shot gathers in field data applications, and the trained network on synthetic data is used as an initialization of the network training on field data, called the transfer learning strategy.
Abstract: Seismic trace interpolation is an important technique because irregular or insufficient sampling data along the spatial direction may lead to inevitable errors in multiple suppression, imag...
70 citations
TL;DR: Logging data experiments show that MLSTM has better robustness and accuracy in depth sequence prediction, and the porosity value at the depth inflection point can be better predicted when the trend of the depth sequence was predicted.
Abstract: The cost of obtaining a complete porosity value using traditional coring methods is relatively high, and as the drilling depth increases, the difficulty of obtaining the porosity value also...
65 citations
TL;DR: In this paper, an interpolation method based on the denoising convolutional neural network (CNN) for seismic data was developed for a simple and efficient way to break through the problem of seismic data interpolation.
Abstract: We have developed an interpolation method based on the denoising convolutional neural network (CNN) for seismic data. It provides a simple and efficient way to break through the problem of ...
64 citations
TL;DR: A deep neural network is developed, whose design was inspired by the information flow found in semblance analysis, that can estimate velocity from seismic data and that good performance can be achieved on real data even if the training is based on synthetics.
Abstract: Applying deep learning to 3D velocity model building remains a challenge due to the sheer volume of data required to train large-scale artificial neural networks. Moreover, little is known ...
62 citations
TL;DR: A gradient boosting decision tree (GBDT) algorithm combining synthetic minority oversampling technique (SMOTE) to realize fast and automatic lithology identification and experimental results indicate that the proposed approach improves the lithology Identification performance compared with other machine-learning approaches.
Abstract: Lithology identification based on conventional well-logging data is of great importance for geologic features characterization and reservoir quality evaluation in the exploration and produc...
TL;DR: A data-driven deep-learning-based method for fast and efficient seismic deblending using a convolutional neural network designed according to the special characteristics of seismic data and performsdeblending with results comparable to those obtained with conventional industry debl lending algorithms.
Abstract: For economic and efficiency reasons, blended acquisition of seismic data is becoming increasingly commonplace. Seismic deblending methods are computationally demanding and normally consist ...
TL;DR: This work has formulated horizon picking as a multiclass segmentation problem and solved it by supervised training of a 3D convolutional neural network by designing an efficient architecture to analyze the data over multiple scales while keeping memory and computational needs to a practical level.
Abstract: Extracting horizon surfaces from key reflections in a seismic image is an important step of the interpretation process. Interpreting a reflection surface in a geologically complex area is a...
TL;DR: This work investigates a method based on a deep neural network that can separate simultaneous source data efficiently and embeds the trained network into an iterative framework that can further improve the deblending.
Abstract: Simultaneous source technology can accelerate data acquisition and improve subsurface illumination. But those advantages are compromised due to dense interference. To address the intense in...
TL;DR: This study presents the first demonstration of the transferability of a convolutional neural network trained to detect microseismic events in one fiber-optic distributed acoustic sensing (DAS) data set to other data sets, using synthetically generated waveforms with real noise superimposed.
Abstract: This study presents the first demonstration of the transferability of a convolutional neural network (CNN) trained to detect microseismic events in one fiber-optic distributed acoustic sens...
TL;DR: Among the large variety of mathematical and computational methods for estimating reservoir properties such as facies and petrophysical variables from geophysical data, deep machine-learning methods have been proposed.
Abstract: Among the large variety of mathematical and computational methods for estimating reservoir properties such as facies and petrophysical variables from geophysical data, deep machine-learning...
TL;DR: In this paper, the authors propose to invert reservoir porosity from post-stack seismic data using an innovative approach based on deep learning methods, circumventing the requir... and develop an unsupervised approach to circumvent the requi...
Abstract: We propose to invert reservoir porosity from poststack seismic data using an innovative approach based on deep-learning methods. We develop an unsupervised approach to circumvent the requir...
TL;DR: In this article, a relative geologic time (RGT) image from a seismic image is constructed for seismic structural and stratigraphic interpretation, and the image is then used to construct a RGT image from the seismic image.
Abstract: Constructing a relative geologic time (RGT) image from a seismic image is crucial for seismic structural and stratigraphic interpretation. In conventional methods, automatic RGT estimation ...
TL;DR: In this paper, the authors used CNNs to obtain petrophysical properties in the depth domain from prestack seismic data in the time domain and compared end-to-end workflows.
Abstract: We have built convolutional neural networks (CNNs) to obtain petrophysical properties in the depth domain from prestack seismic data in the time domain. We compare two workflows — end-to-en...
TL;DR: The method is examined on sonic log prediction and can produce an accurate prediction of sonic logs from gamma-ray, density, and neutron porosity logs and outputs the uncertainties facilitated by dropout layers and Monte Carlo sampling at inference time.
Abstract: Reservoir characterization involves integration of different types of data to understand the subsurface rock properties. To incorporate multiple well log types into reservoir studies, estim...
TL;DR: A semisupervised workflow for efficient seismic stratigraphy interpretation by using the state-of-the-art deep convolutional neural networks (CNNs), which is designed to be applicable to three scenarios, trace-wise, paintbrushing, and full-sectional annotation.
Abstract: Depicting geologic sequences from 3D seismic surveying is of significant value to subsurface reservoir exploration, but it is usually time- and labor-intensive for manual interpretation by ...
TL;DR: In this article, a roadside section of the Stanford DAS-2 array can record seismic signals from various sources, such as the Earth's magnetic field and the magnetic field of the Earth.
Abstract: Due to the broadband nature of distributed acoustic sensing (DAS) measurement, a roadside section of the Stanford DAS-2 array can record seismic signals from various sources. For example, ...
TL;DR: This article estimated migrated images with meaningful amplitudes matching least-squares migrated images by approximating the inverse Hessian using generative adversarial networks (GANs) in a co...
Abstract: We have estimated migrated images with meaningful amplitudes matching least-squares migrated images by approximating the inverse Hessian using generative adversarial networks (GANs) in a co...
TL;DR: An artificial neural network is developed to estimate P-wave velocity models directly from prestack common-source gathers and indicates encouraging results for predicting velocity models from Prestack seismic data that are acquired with the same geometry as in the training set.
Abstract: We have developed an artificial neural network to estimate P-wave velocity models directly from prestack common-source gathers. Our network is composed of a fully connected layer set and a ...
TL;DR: In 2017, distributed acoustic sensing (DAS) technology was deployed in a horizontal well to conduct a vertical seismic profiling survey before and after each of 78 hydraulically fracturing stag... as mentioned in this paper.
Abstract: In 2017, distributed acoustic sensing (DAS) technology was deployed in a horizontal well to conduct a vertical seismic profiling survey before and after each of 78 hydraulic fracturing stag...
TL;DR: A data-driven low-frequency recovery method based on deep learning from high-frequency signals to reconstruct the missing low- frequency signals more accurately and effectively is developed.
Abstract: The lack of low-frequency signals in seismic data makes the full-waveform inversion (FWI) procedure easily fall into local minima leading to unreliable results. To reconstruct the missing l...
TL;DR: An unsupervised approach, waveform embedding, based on a deep convolutional autoencoder network to learn to transform seismic waveform samples to a latent space in which any waveform can be represented as an embedded vector.
Abstract: Picking horizons from seismic images is a fundamental step that could critically impact seismic interpretation quality. We have developed an unsupervised approach, waveform embedding, based...
TL;DR: A generative adversarial network (GAN) to attenuate ground roll in seismic data is developed based on a large training data set that includes pairs of data with and without ground roll and obtains better results in the attenuation of groundroll and in the preservation of signals compared to the three other methods.
Abstract: Ground roll is a persistent problem in land seismic data. This type of coherent noise often contaminates seismic signals and severely reduces the signal-to-noise ratio of seismic data. A va...
TL;DR: In this paper, the authors present a deterministic and stochastic algorithm for reservoir characterization, which is an essential component of oil and gas production, as well as exploration and exploration.
Abstract: Reservoir characterization is an essential component of oil and gas production, as well as exploration. Classic reservoir characterization algorithms, deterministic and stochastic, are typi...
TL;DR: This work has adopted a novel antinoise classifier for waveform classification and arrival picking by combining the continuous wavelet transform (CWT) and the convolutional neural network (CNN).
Abstract: Microseismic data have a low signal-to-noise ratio (S/N). Existing waveform classification and arrival-picking methods are not effective enough for noisy microseismic data with low S/N. We ...