Surface-related multiple elimination with deep learning

Siahkoohi, Ali; Verschuur, D.J.; Herrmann, Felix J.

doi:10.1190/segam2019-3216723.1

Cited by 39 publications

(8 citation statements)

References 24 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Siahkoohi et al (2018) utilize CNNs to remove the free-surface multiples and numerical dispersion; Das et al (2019) use CNNs to obtain an elastic subsurface model using recorded normal-incidence seismic data; Wu et al (2019) use CNNs for three-dimensional seismic fault segmentation; Almuteri and Sava (2021) use CNNs to address to ghost removal from seismic data; Kiraz and Snieder (2022) utilize CNNs for one-dimensional (1D) wavefield focusing where the solution of the Marchenko equation is not needed to retrieve the Green's function once the network is trained. Recently, CNNs have been used to tackle free-surface multiples in various ways (Siahkoohi et al, 2018(Siahkoohi et al, , 2019Ovcharenko et al, 2021;Zhang et al, 2021;Liu-Rong et al, 2021). In this paper, we use a trace-by-trace CNN approach because training and prediction take shorter than higher dimensional CNN approaches.…”

Section: Convolutional Neural Network Architecture and Trainingmentioning

confidence: 99%

Attenuating free‐surface multiples and ghost reflection from seismic data using a trace‐by‐trace convolutional neural network approach

Kiraz,

Snieder,

Sheiman

2023

Geophysical Prospecting

View full text Add to dashboard Cite

The presence of the air‐water interface (or free‐surface) creates two major problems in marine seismic data for conventional seismic processing and imaging: free‐surface multiples and ghost reflections. The attenuation of free‐surface multiples remains one of the most challenging noise attenuation problems in seismic data processing. Current solutions suffer from the removal of the primary events along with the multiple events especially when the primary and multiple events overlap (e.g., adaptive subtraction). The effective attenuation of ghost reflections (or deghosting) requires acquisition‐ and/or processing‐related solutions which generally address the source‐side and receiver‐side ghosts separately. Additionally, an essential requirement for a successful implementation of free‐surface multiple attenuation and seismic dehosting is the requirement of dense seismic data acquisition parameters which is not realistic for 2D and/or 3D marine cases. We present a Convolutional Neural Network (CNN) approach for free‐surface multiple attenuation and seismic deghosting. Unlike the existing solutions, our approach operates on a single trace at a time, and neither relies on the dense acquisition parameters nor requires a subtraction process to eliminate free‐surface multiples, and it removes both the source ghost and receiver ghost simultaneously. We train a network using subsets of the Marmousi and Pluto velocity models, and make predictions using subsets of the Sigsbee velocity model. We show that the CNN predictions give a correlation coefficient of 0.97 on average with the numerically modeled data for the synthetic examples. We illustrate the efficacy of our CNN‐based technique using the Mobil AVO Viking Graben field data set. The application of our algorithm demonstrates that our CNN‐based approach removes different orders of free‐surface multiples (e.g., 1 and 2 orders) and recovers the low‐frequency content of the seismic data (which is essential for, for instance, full‐waveform inversion applications and broadband processing) by successfully removing the ghost reflections while preserving and increasing the continuity of the primary reflections.This article is protected by copyright. All rights reserved

show abstract

Section: Convolutional Neural Network Architecture and Trainingmentioning

confidence: 99%

Attenuating free‐surface multiples and ghost reflection from seismic data using a trace‐by‐trace convolutional neural network approach

Kiraz,

Snieder,

Sheiman

2023

Geophysical Prospecting

View full text Add to dashboard Cite

show abstract

“…convolutional layers with migration/demigration operators. This capability has enabled various research projects in our group, including surface-related multiple elimination [25], dispersion attenuation [26], and ML-augmented imaging and uncertainty quantification [27,28]. We show in the following two examples of machine learning for exploration geophysics that take advantage of these high-level abstractions to interface PDE solvers (JUDI, Devito) with deep learning frameworks.…”

Section: Machine Learningmentioning

confidence: 99%

Accelerating innovation with software abstractions for scalable computational geophysics

Louboutin¹,

Witte²,

Siahkoohi³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We present the SLIM open-source software framework for computational geophysics, and more generally, inverse problems based on the wave-equation (e.g., medical ultrasound). We developed a software environment aimed at scalable research and development by designing multiple layers of abstractions. This environment allows the researchers to easily formulate their problem in an abstract fashion, while still being able to exploit the latest developments in high-performance computing. We illustrate and demonstrate the benefits of our software design on many geophysical applications, including seismic inversion and physics-informed machine learning for geophysics(e.g., loop unrolled imaging, uncertainty quantification), all while facilitating the integration of external software.

show abstract

“…Das et al (2019) proposed a wave impedance inversion method based on CNN, which predicted wave impedance from generated samples when the local medium model and source wavelet phase were outside the training set. Wang et al (2022a) proposed a seismic multiple suppression method (Siahkoohi et al, 2019a;Wang et al, 2022b) using the DNN based on data augmentation training, which had a certain ability to work across work areas under transfer learning (Siahkoohi et al, 2019b). Denoising CNNs (DnCNN) and U-Net CNNs are two of the most commonly used network structures in data denoising (Ronneberger et al, 2015;Zhang et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

A multi‐data training method for a deep neural network to improve the separation effect of simultaneous‐source data

Wang

Mao

Song

et al. 2022

Geophysical Prospecting

View full text Add to dashboard Cite

Within the field of seismic data acquisition with active sources, the technique of acquiring simultaneous data, also known as blended data, offers operational advantages. The preferred processing of blended data starts with a step of deblending, that is separation of the data acquired by the different sources, to produce data that mimic data from a conventional seismic acquisition and can be effectively processed by standard methods. Recently, deep learning methods based on the deep neural network have been applied to the deblending task with promising results, in particular using an iterative approach. We propose an enhancement to deblending with an iterative deep neural network, whereby we modify the training stage of the deep neural network in order to achieve better performance through the iterations. We refer to the method that only uses the blended data as the input data as the general training method. Our new multi-data training method allows the deep neural network to be trained by the data set with the input patches composed of blended data, noisy data with low amplitude crosstalk noise, and unblended data, which can improve the ability of the deep neural network to remove crosstalk noise and protect weak signal. Based on such an extended training data set, the multi-data training method embedded in the iterative separation framework can result in different outputs at different iterations and converge to the best result in a shorter iteration number. Transfer learning can further improve the generalization and separation efficacy of our proposed method to deblend the simultaneous-source data. Our proposed method is tested on two synthetic data and two field data to prove the effectiveness and superiority in the deblending of the simultaneous-source data compared with the general training method, generic noise attenuation network and low-rank matrix factorization methods.

show abstract

Surface-related multiple elimination with deep learning

Cited by 39 publications

References 24 publications

Attenuating free‐surface multiples and ghost reflection from seismic data using a trace‐by‐trace convolutional neural network approach

Attenuating free‐surface multiples and ghost reflection from seismic data using a trace‐by‐trace convolutional neural network approach

Accelerating innovation with software abstractions for scalable computational geophysics

A multi‐data training method for a deep neural network to improve the separation effect of simultaneous‐source data

Contact Info

Product

Resources

About