InversionNet: An Efficient and Accurate Data-Driven Full Waveform Inversion

Wu, Yue; Lin, Youzuo

doi:10.1109/tci.2019.2956866

Cited by 130 publications

(78 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The numerical experiments compared with FWI showed that the complicated nonlinear mapping from seismic data to velocity model can be well approximated by the convolutional neural network (CNN). Similar works were described in [40,41,42,43], which share the same purpose by using the universal approximation of different CNN architecture. Most of the aforementioned methods depend on supervised learning to learn the mapping between the paired input and ground-truth.…”

Section: Supervised Learning Approaches For Waveform Inversionmentioning

confidence: 92%

Revisit Geophysical Imaging in A New View of Physics-informed Generative Adversarial Learning

Yang,

2021

Preprint

View full text Add to dashboard Cite

Seismic full waveform inversion (FWI) is a powerful geophysical imaging technique that produces high-resolution subsurface models by iteratively minimizing the misfit between the simulated and observed seismograms. Unfortunately, conventional FWI with least-squares function suffers from many drawbacks such as the local-minima problem and computation of explicit gradient. It is particularly challenging with the contaminated measurements or poor starting models. Recent works relying on partial differential equations and neural networks show promising performance for two-dimensional FWI. Inspired by the competitive learning of generative adversarial networks, we proposed an unsupervised learning paradigm that integrates wave equation with a discriminate network to accurately estimate the physically consistent models in a distribution sense. Our framework needs no labelled training data nor pretraining of the network, is flexible to achieve multi-parameters inversion with minimal user interaction. The proposed method faithfully recovers the well-known synthetic models that outperforms the classical algorithms. Furthermore, our work paves the way to sidestep the local-minima issue via reducing the sensitivity to initial models and noise. Keywords Seismic full waveform inversionRecently, deep-learning (DL) techniques [7], in particular, convolutional neural network (CNN) [8], have achieved the state-of-the-art performance in a variety of applications. They surpass the conventional approaches in many research fields of inverse problem, for instance, image reconstruction [9, 10], x-ray computed tomography [11], and optical diffraction tomography [12,13]. To some extent, this development results from the wide availability of domain-specific languages (DSL) for DL, such as Tensorflow [14] or PyTorch [15], used in academia and industry.

show abstract

Section: Supervised Learning Approaches For Waveform Inversionmentioning

confidence: 92%

Revisit Geophysical Imaging in A New View of Physics-informed Generative Adversarial Learning

Yang,

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The CNN structure has also been utilized for seismic inversion in geophysics. Y. Wu and Lin (2020) proposed the InversionNet for full waveform inversion, which employed an encoder‐decoder structure of CNN. S. Li et al.…”

Section: Introductionmentioning

confidence: 99%

Deep‐Learning‐Based Inverse Modeling Approaches: A Subsurface Flow Example

2021

View full text Add to dashboard Cite

Inverse modeling aims to infer uncertain parameters of a system with noisy observations of the system response, which has been widely utilized in various scientific and engineering practices, such as seismic inversion (Bunks et al., 1995), petroleum reservoir history matching (Oliver et al., 2008), aquifer parameter estimation (Carrera & Neuman, 1986a, 1986b, 1986c), and medical imaging (Arridge, 1999). Under certain conditions, inverse problems can be viewed as optimization problems, in which the model parameters are modified, such that the predictions from forward models can match the measurements. In addition, prior knowledge can be used to regularize the objective function and to obtain the maximum a posteriori estimate of the model parameters. The gradient-based method is a straightforward and common way to perform inverse modeling tasks. Anterion et al. (1989) presented a rigorous analytical method to calculate the gradients of observations with respect to reservoir characteristics, which can then be used to assist the process of parameter adjustment for history matching. For calculating the required gradients in inverse modeling, the adjoint method is a frequently utilized technique (

show abstract

“…Data-driven approaches seek to limit the need for expert knowledge by implicitly deriving the concept of wave phenomena from a corpus of relevant seismic data. Therefore, they can potentially deliver good starting models for FWI by replacing tomography (Araya-Polo et al, 2018;Kazei et al, 2019bKazei et al, , 2020Zwartjes, 2020) or even gradually reproducing the FWI resolution for small models (Lin and Zhang, 2019;Wu and Lin, 2019;Yuan et al, 2020;Araya-Polo et al, 2020;Siahkoohi et al, 2020). Supervised learning approaches for the task of initial velocity model estimation require large datasets of realistic subsurface models.…”

Section: Introductionmentioning

confidence: 99%

Extrapolating low-frequency prestack land data with deep learning

Ovcharenko

Kazei

Plotnitskiy

et al. 2020

SEG Technical Program Expanded Abstracts 2020

View full text Add to dashboard Cite

Missing low-frequency content in seismic data is a common challenge for seismic inversion. Long wavelengths are necessary to reveal large structures in the subsurface and to build an acceptable starting point for later iterations of full-waveform inversion (FWI). High-frequency land seismic data are particularly challenging due to the elastic nature of the Earth contrasting with acoustic air at the typically rugged free surface, which makes the use of low frequencies even more vital to the inversion. We propose a supervised deep learning framework for bandwidth extrapolation of prestack elastic data in the time domain. We utilize a Convolutional Neural Network (CNN) with a UNet-inspired architecture to convert portions of band-limited shot gathers from 5-15 Hz to 0-5 Hz band. In the synthetic experiment, we train the network on 192x192 patches of wavefields simulated for different cross-sections of the elastic SEAM Arid model with free-surface. Then, we test the network on unseen shot gathers from the same model to demonstrate the viability of the approach. The results show promise for future field data applications.

show abstract

InversionNet: An Efficient and Accurate Data-Driven Full Waveform Inversion

Cited by 130 publications

References 48 publications

Revisit Geophysical Imaging in A New View of Physics-informed Generative Adversarial Learning

Revisit Geophysical Imaging in A New View of Physics-informed Generative Adversarial Learning

Deep‐Learning‐Based Inverse Modeling Approaches: A Subsurface Flow Example

Extrapolating low-frequency prestack land data with deep learning

Contact Info

Product

Resources

About