Intelligent Deblending of Seismic Data Based on U-Net and Transfer Learning

Wang, Benfeng; Li, Jiakuo; Luo, Jungang; Wang, Yingying; Geng, Jianhua

doi:10.1109/tgrs.2020.3048746

Cited by 44 publications

(16 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The classical dictionary learning based deblending method [15] and the end-to-end deblending CNN are used for comparisons. For the latter comparative method, we follow the same training strategy in [19], which utilizes the pseudo-deblended data and original unblended data as "noisy-clean" training pairs, and specially use the blended CSGs for the network training. The network architecture is the widely used DnCNN [26].…”

Section: Methodsmentioning

confidence: 99%

“…Inversion methods work by explicitly reversing the blending process, which is an ill-posed inverse problem that needs regularization. Within these two categories, deblending methods may build upon different methodologies: median filtering [6]- [8], rank-reduction [9], [10], sparse representation [11]- [16], and deep learning [17]- [19].…”

mentioning

confidence: 99%

“…In recent years, owing to the ability of convolutional neural networks (CNNs) to learn representative features from large amount of training data, end-to-end deblending methods trained on pairs of pseudo-deblended data and unblended data have been proposed [17] and further integrated into deblending iterative strategies [18], [19]. However, the performance of these methods largely depends on the training data, and collecting proper training data to avoid generalization issues is a challenging task.…”

mentioning

confidence: 99%

“…The decoder branch of the trained AE is subsequently used as a nonlinear preconditioner for the deblending inverse problem which is cast as a nonlinear optimization in the latent space of the decoder, and can be effectively solved by second-order nonlinear solvers. Synthetic and field datasets are used to test the performance of our presented method against well-known techniques belonging to different families: a dictionary learning based deblending method [15]; an end-to-end deblending CNN [19].…”

mentioning

confidence: 99%

See 3 more Smart Citations

Intelligent Seismic Deblending Through Deep Preconditioner

Lipari

Bestagini

et al. 2022

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

Seismic deblending is an ill-posed inverse problem that involves counteracting the effect of a blending matrix derived from the shots position and firing time. In this letter, we propose a seismic deblending method based on so-called deep preconditioners. A convolutional Autoencoder (AE) is first trained in a patch-wise fashion to learn an effective sparse representation of the common receiver gathers (CRGs) we aim to reconstruct. Then, the decoder branch of the trained AE is used as a nonlinear preconditioner for the deblending problem. Particularly, to avoid the explicit creation of a training dataset, we suggest to use the common shot gathers (CSGs) of the blended dataset itself to train the AE network, as they are not affected by incoherent blending noise. Numerical examples on synthetic and field datasets demonstrate the effectiveness of the proposed method in comparison with significantly comparable techniques: a dictionary-learning based deblending method; an end-to-end deblending convolution neutral network (CNN).

show abstract

Section: Methodsmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Intelligent Seismic Deblending Through Deep Preconditioner

Lipari

Bestagini

et al. 2022

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

show abstract

“…2a), which is composed primarily of three parts, namely, the encoder, decoder and channel concatenation, and more information can be found in Wang et al . (2021). Furthermore, we employ the nonlinear LeakyReLU (He et al ., 2015) activation function to approach seismic data more accurately.…”

Section: Theorymentioning

confidence: 99%

Consistent convolution kernel design for missing shots interpolation using an improved U‐net

Han

Wang

2022

Geophysical Prospecting

Self Cite

View full text Add to dashboard Cite

During seismic data acquisition, receivers with finer trace intervals can be distributed to record seismic signals. The distance between neighbouring shots, on the other hand, can be significant, resulting in regularly missing shots in the common receiver gathers, which may cause significant spatial aliasing and reduce the precision of later seismic imaging. Traditional interpolation algorithms have several issues and limitations for seismic data with spatial aliasing due to prior assumptions and human–computer interactions. As a result, we present a new deep learning method that includes the generation of adaptive training data and the design of a consistent convolution kernel. Deep learning has the ability to characterize seismic data in a nonlinear manner that is ideal for accurate seismic interpolation. Because the common shot gathers and common receiver gathers have similar features due to the spatial reciprocity of Green's function of the wave equation if all shots have the same physical signatures, the common shot gathers with a refined trace interval are adaptively extracted as the training dataset to guarantee the interpolation performance in the common receiver gathers. With regularly subsampled data as input an improved U‐Net is created and trained to match the desired output, that is, the completed data. The trained network can be deployed to the test common receiver gathers to rebuild regularly missing shots. We develop a novel consistent convolution kernel to ensure high accuracy of missing shot reconstruction while accounting for differences between prestack unmigrated seismic data and images. Using numerically subsampled synthetic and field data, the effectiveness and validity of the developed consistent convolution kernel and the upgraded U‐Net with adaptive training data for missing shot interpolation are demonstrated.

show abstract

Deep Learning Study on Seismic Data Interpretation Method

He,

Yu,

et al. 2024

Springer Series in Geomechanics and Geoengineering

View full text Add to dashboard Cite

Intelligent Deblending of Seismic Data Based on U-Net and Transfer Learning

Cited by 44 publications

References 38 publications

Intelligent Seismic Deblending Through Deep Preconditioner

Intelligent Seismic Deblending Through Deep Preconditioner

Consistent convolution kernel design for missing shots interpolation using an improved U‐net

Deep Learning Study on Seismic Data Interpretation Method

Contact Info

Product

Resources

About