Ground-roll attenuation using generative adversarial networks

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. WA255-WA267 ◽  
Author(s):  
Yijun Yuan ◽  
Xu Si ◽  
Yue Zheng
Keyword(s):  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Coherent Noise ◽  
Traditional Methods ◽  
Generative Adversarial Network ◽  
Data Set ◽  
Persistent Problem ◽  
Ground Roll

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 variety of methods for addressing ground-roll attenuation have been developed. However, existing methods are limited, especially when using real land seismic data. For example, when ground roll and reflections overlap in the time or frequency domains, traditional methods cannot completely separate them and they often distort the signals during the suppression process. We have developed a generative adversarial network (GAN) to attenuate ground roll in seismic data. Unlike traditional methods for ground-roll attenuation dependent on various filters, the GAN method is based on a large training data set that includes pairs of data with and without ground roll. After training the neural network with the training data, the network can identify and filter out any noise in the data. To fulfill this purpose, the proposed method uses a generator and a discriminator. Through network training, the generator learns to create the data that can fool the discriminator, and the discriminator can then distinguish between the data produced by the generator and the training data. As a result of the competition between generators and discriminators, generators produce better images whereas discriminators accurately recognize targets. Tests on synthetic and real land seismic data show that the proposed method effectively reveals reflections masked by the ground roll and obtains better results in the attenuation of ground roll and in the preservation of signals compared to the three other methods.

Deep learning-driven velocity model building workflow

2019 ◽  
Vol 38 (11) ◽  
pp. 872a1-872a9 ◽  
Author(s):  
Mauricio Araya-Polo ◽  
Stuart Farris ◽  
Manuel Florez
Keyword(s):  
Deep Learning ◽  
Seismic Data ◽  
Model Building ◽  
Ground Truth ◽  
Velocity Model ◽  
Training Data ◽  
Quality Data ◽  
Generative Adversarial Network ◽  
Data Set ◽  
Velocity Models

Exploration seismic data are heavily manipulated before human interpreters are able to extract meaningful information regarding subsurface structures. This manipulation adds modeling and human biases and is limited by methodological shortcomings. Alternatively, using seismic data directly is becoming possible thanks to deep learning (DL) techniques. A DL-based workflow is introduced that uses analog velocity models and realistic raw seismic waveforms as input and produces subsurface velocity models as output. When insufficient data are used for training, DL algorithms tend to overfit or fail. Gathering large amounts of labeled and standardized seismic data sets is not straightforward. This shortage of quality data is addressed by building a generative adversarial network (GAN) to augment the original training data set, which is then used by DL-driven seismic tomography as input. The DL tomographic operator predicts velocity models with high statistical and structural accuracy after being trained with GAN-generated velocity models. Beyond the field of exploration geophysics, the use of machine learning in earth science is challenged by the lack of labeled data or properly interpreted ground truth, since we seldom know what truly exists beneath the earth's surface. The unsupervised approach (using GANs to generate labeled data)illustrates a way to mitigate this problem and opens geology, geophysics, and planetary sciences to more DL applications.

scholarly journals 3D Design Using Generative Adversarial Networks and Physics-Based Validation

2019 ◽  
Vol 142 (7) ◽  
Author(s):  
Dule Shu ◽  
James Cunningham ◽  
Gary Stump ◽  
Simon W. Miller ◽  
Michael A. Yukish ◽  
...  
Keyword(s):  
Evaluation Process ◽  
3D Models ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
3D Design ◽  
Generative Adversarial Network ◽  
Data Set ◽  
High Performing ◽  
Adversarial Network

Abstract The authors present a generative adversarial network (GAN) model that demonstrates how to generate 3D models in their native format so that they can be either evaluated using complex simulation environments or realized using methods such as additive manufacturing. Once initially trained, the GAN can create additional training data itself by generating new designs, evaluating them in a physics-based virtual environment, and adding the high performing ones to the training set. A case study involving a GAN model that is initially trained on 4045 3D aircraft models is used for demonstration, where a training data set that has been updated with GAN-generated and evaluated designs results in enhanced model generation, in both the geometric feasibility and performance of the designs. Z-tests on the performance scores of the generated aircraft models indicate a statistically significant improvement in the functionality of the generated models after three iterations of the training-evaluation process. In the case study, a number of techniques are explored to structure the generate-evaluate process in order to balance the need to generate feasible designs with the need for innovative designs.

GROUND ROLL ATTENUATION USING ADAPTIVE SINGULAR VALUE DECOMPOSITION FILTERING IN THE F-K DOMAIN

2016 ◽  
Vol 33 (3) ◽  
Author(s):  
Danilo S. Cruz ◽  
Milton J. Porsani
Keyword(s):  
Singular Value Decomposition ◽  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Singular Value ◽  
Signal To Noise ◽  
Coherent Noise ◽  
Ground Roll ◽  
Noise Ratio ◽  
Value Decomposition

ABSTRACT. The land seismic data often have low signal-to-noise ratio due, among other factors, the presence of ground roll. It is a coherent noise present in seismograms that appears as linear events... RESUMO. Os dados sísmicos terrestres geralmente apresentam baixa razão sinal-ruído devido, entre outros fatores, à presença do ground roll . Trata-se de um ruído dominado por altas amplitudes...

Coherent noise attenuation in the radial trace domain

Geophysics ◽  
2003 ◽  
Vol 68 (4) ◽  
pp. 1408-1416 ◽  
Author(s):  
David C. Henley
Keyword(s):  
Seismic Data ◽  
Noisy Data ◽  
Data Mapping ◽  
Noise Attenuation ◽  
Practical Application ◽  
Coherent Noise ◽  
Data Set ◽  
Persistent Problem ◽  
Mapping Algorithm ◽  
Better Than

Coherent noise is a persistent problem in seismic imaging, and a number of techniques have been developed to attenuate it. The radial trace (RT) transform, a simple seismic data mapping algorithm, can be used as the basis for a particularly flexible and effective method for attenuating coherent noise on both prestack and poststack seismic data. Described here are the principles and some practical application details for attenuating coherent noise in the RT domain. A comparison between frequency–wavenumber (f–k) and RT domain filtering on a synthetic model is presented, and some of the differences and advantages of RT methods are identified. Next, RT coherent noise attenuation is demonstrated using a set of good‐quality field data; it is then applied to a very noisy data set. The results obtained with this last set prove to be as good as, or better than, those produced using f–k filtering.

scholarly journals GROUND ROLL ATTENUATION USING SHAPING FILTERS AND BAND LIMITED SWEEP SIGNALS

2012 ◽  
Vol 30 (4) ◽  
pp. 545 ◽  
Author(s):  
Quézia C. dos Santos ◽  
Milton José Porsani
Keyword(s):  
Data Processing ◽  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Direct Methods ◽  
Data Set ◽  
Low Frequencies ◽  
Shaping Filter ◽  
Ground Roll ◽  
Band Limited ◽  
Receiver Arrays

Os dados sísmicos terrestres, geralmente, apresentam baixa razão sinal-ruído devido, entre outros fatores, à presença deground roll, um ruído caracterizado por eventos coerentes e lineares, com altas amplitudes, baixas frequências temporais e baixas velocidades e, na maioria dos casos, dispersivos, que se sobrepõem àsreflexões, prejudicando o processamento e a interpretação dos dados. Quando a tentativa de atenuar oground roll durante a aquisição dos dados (utilizando arranjos de fontes e receptores) falha, diversos métodos podem ser empregados no processamento. Neste trabalho, discute-se um método de filtragem baseado no filtrode forma de Wiener, sua implementação e seus principais parâmetros. Também é apresentada uma variante do método, baseada no algoritmo de deconvolução direta. Os resultados da aplicação da filtragem direta em dados sísmicos reais são bastante satisfatórios, quando comparados com aqueles obtidos com os métodos convencionais FK e corta-baixas. ABSTRACT: Onshore seismic data often have low signal to noise ratio due to, among other factors, the presence of ground-roll, a noise characterized by coherent,linear and dispersive events with high amplitudes, low frequencies and velocities. This noise overlaps with reflections, hindering the data processing and interpretation.When the attempts to reduce the ground-roll during data acquisition (using source and receiver arrays) fail, several methods can be used in seismic processing. Herewe discuss a filtering method based on Wiener shaping filter, its implementation and its main parameters. We also present a different approach based on the directdeconvolution algorithm. The results of the application of direct methods to a real seismic data set are quite satisfactory when compared with those obtained withconventional FK and low-cut filters.Keywords: ground roll, shaping filters, seismic data processing.

Linear electromagnetic inverse scattering via generative adversarial networks

2021 ◽  
pp. 1-9
Author(s):  
Huilin Zhou ◽  
Huimin Zheng ◽  
Qiegen Liu ◽  
Jian Liu ◽  
Yuhao Wang
Keyword(s):  
Inverse Scattering ◽  
Optimization Methods ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Scattering Problems ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Adversarial Networks ◽  
Highly Nonlinear ◽  
Electromagnetic Inverse Scattering

Abstract Electromagnetic inverse-scattering problems (ISPs) are concerned with determining the properties of an unknown object using measured scattered fields. ISPs are often highly nonlinear, causing the problem to be very difficult to address. In addition, the reconstruction images of different optimization methods are distorted which leads to inaccurate reconstruction results. To alleviate these issues, we propose a new linear model solution of generative adversarial network-based (LM-GAN) inspired by generative adversarial networks (GAN). Two sub-networks are trained alternately in the adversarial framework. A linear deep iterative network as a generative network captures the spatial distribution of the data, and a discriminative network estimates the probability of a sample from the training data. Numerical results validate that LM-GAN has admirable fidelity and accuracy when reconstructing complex scatterers.

Underwater Acoustic Target Recognition Based on Generative Adversarial Network Data Augmentation

2021 ◽  
Vol 263 (2) ◽  
pp. 4558-4564
Author(s):  
Minghong Zhang ◽  
Xinwei Luo
Keyword(s):  
Data Augmentation ◽  
Target Recognition ◽  
Training Data ◽  
Small Samples ◽  
Generative Adversarial Network ◽  
Data Set ◽  
Underwater Acoustic ◽  
Adversarial Network ◽  
Acoustic Target ◽  
The Impact

Underwater acoustic target recognition is an important aspect of underwater acoustic research. In recent years, machine learning has been developed continuously, which is widely and effectively applied in underwater acoustic target recognition. In order to acquire good recognition results and reduce the problem of overfitting, Adequate data sets are essential. However, underwater acoustic samples are relatively rare, which has a certain impact on recognition accuracy. In this paper, in addition of the traditional audio data augmentation method, a new method of data augmentation using generative adversarial network is proposed, which uses generator and discriminator to learn the characteristics of underwater acoustic samples, so as to generate reliable underwater acoustic signals to expand the training data set. The expanded data set is input into the deep neural network, and the transfer learning method is applied to further reduce the impact caused by small samples by fixing part of the pre-trained parameters. The experimental results show that the recognition result of this method is better than the general underwater acoustic recognition method, and the effectiveness of this method is verified.

scholarly journals INFRASTRUCTURE DEGRADATION AND POST-DISASTER DAMAGE DETECTION USING ANOMALY DETECTING GENERATIVE ADVERSARIAL NETWORKS

2020 ◽  
Vol V-2-2020 ◽  
pp. 573-582 ◽  
Author(s):  
S. M. Tilon ◽  
F. Nex ◽  
D. Duarte ◽  
N. Kerle ◽  
G. Vosselman
Keyword(s):  
Deep Learning ◽  
Damage Detection ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Generative Adversarial Network ◽  
Essential Information ◽  
Urban Scenes ◽  
Adversarial Network ◽  
Collapsed Buildings ◽  
Post Disaster

Abstract. Degradation and damage detection provides essential information to maintenance workers in routine monitoring and to first responders in post-disaster scenarios. Despite advance in Earth Observation (EO), image analysis and deep learning techniques, the quality and quantity of training data for deep learning is still limited. As a result, no robust method has been found yet that can transfer and generalize well over a variety of geographic locations and typologies of damages. Since damages can be seen as anomalies, occurring sparingly over time and space, we propose to use an anomaly detecting Generative Adversarial Network (GAN) to detect damages. The main advantages of using GANs are that only healthy unannotated images are needed, and that a variety of damages, including the never before seen damage, can be detected. In this study we aimed to investigate 1) the ability of anomaly detecting GANs to detect degradation (potholes and cracks) in asphalt road infrastructures using Mobile Mapper imagery and building damage (collapsed buildings, rubble piles) using post-disaster aerial imagery, and 2) the sensitivity of this method against various types of pre-processing. Our results show that we can detect damages in urban scenes at satisfying levels but not on asphalt roads. Future work will investigate how to further classify the found damages and how to improve damage detection for asphalt roads.

ROBUST SVD FILTERING FOR LOW SIGNAL-TO-NOISE RATIO SEISMIC DATA

Geophysics ◽  
2021 ◽  
pp. 1-51
Author(s):  
Chao Wang ◽  
Yun Wang
Keyword(s):  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Singular Values ◽  
New Method ◽  
Signal To Noise ◽  
Coherent Noise ◽  
Edge Preserving ◽  
Singular Vectors ◽  
Reduced Rank ◽  
Noise Ratio

Reduced-rank filtering is a common method for attenuating noise in seismic data. As conventional reduced-rank filtering distinguishes signals from noises only according to singular values, it performs poorly when the signal-to-noise ratio is very low, or when data contain high levels of isolate or coherent noise. Therefore, we developed a novel and robust reduced-rank filtering based on the singular value decomposition in the time-space domain. In this method, noise is recognized and attenuated according to the characteristics of both singular values and singular vectors. The left and right singular vectors corresponding to large singular values are selected firstly. Then, the right singular vectors are classified into different categories according to their curve characteristics, such as jump, pulse, and smooth. Each kind of right singular vector is related to a type of noise or seismic event, and is corrected by using a different filtering technology, such as mean filtering, edge-preserving smoothing or edge-preserving median filtering. The left singular vectors are also corrected by using the filtering methods based on frequency attributes like main-frequency and frequency bandwidth. To process seismic data containing a variety of events, local data are extracted along the local dip of event. The optimal local dip is identified according to the singular values and singular vectors of the data matrices that are extracted along different trial directions. This new filtering method has been applied to synthetic and field seismic data, and its performance is compared with that of several conventional filtering methods. The results indicate that the new method is more robust for data with a low signal-to-noise ratio, strong isolate noise, or coherent noise. The new method also overcomes the difficulties associated with selecting an optimal rank.

Sparse spectral attributes to detect hydrocarbons: A case study from deep dolomite reservoirs

2019 ◽  
Vol 7 (3) ◽  
pp. T701-T711
Author(s):  
Jianhu Gao ◽  
Bingyang Liu ◽  
Shengjun Li ◽  
Hongqiu Wang
Keyword(s):  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Low Frequency ◽  
Spectral Amplitude ◽  
Data Set ◽  
Frequency Spectra ◽  
Hydrocarbon Production ◽  
Hydrocarbon Detection ◽  
Time Frequency ◽  
Gas Bearing

Hydrocarbon detection is always one of the most critical sections in geophysical exploration, which plays an important role in subsequent hydrocarbon production. However, due to the low signal-to-noise ratio and weak reflection amplitude of deep seismic data, some conventional methods do not always provide favorable hydrocarbon prediction results. The interesting dolomite reservoirs in Central Sichuan are buried over an average depth of 4500 m, and the dolomite rocks have a low porosity below approximately 4%, which is measured by well-logging data. Furthermore, the dominant system of pores and fractures as well as strong heterogeneity along the lateral and vertical directions lead to some difficulties in describing the reservoir distribution. Spectral decomposition (SD) has become successful in illuminating subsurface features and can also be used to identify potential hydrocarbon reservoirs by detecting low-frequency shadows. However, the current applications for hydrocarbon detection always suffer from low resolution for thin reservoirs, probably due to the influence of the window function and without a prior constraint. To address this issue, we developed sparse inverse SD (SISD) based on the wavelet transform, which involves a sparse constraint of time-frequency spectra. We focus on investigating the applications of sparse spectral attributes derived from SISD to deep marine dolomite hydrocarbon detection from a 3D real seismic data set with an area of approximately [Formula: see text]. We predict and evaluate gas-bearing zones in two target reservoir segments by analyzing and comparing the spectral amplitude responses of relatively high- and low-frequency components. The predicted results indicate that most favorable gas-bearing areas are located near the northeast fault zone in the upper reservoir segment and at the relatively high structural positions in the lower reservoir segment, which are in good agreement with the gas-testing results of three wells in the study area.

Sign in / Sign up

Export Citation Format

Share Document