A partial convolution-based deep-learning network for seismic data regularization1

Shulin Pan,Kai Chen,Jingyi Chen,Ziyu Qin,Qinghui Cui,Jing Li

DOI: https://doi.org/10.1016/j.cageo.2020.104609

2020-12-01

Abstract:<p>Spatial undersampling is a common problem in actual seismic data due to limitations in seismic survey environments, which can be satisfactorily solved by data regularization. The convolution-based deep-learning reconstruction methods require fewer assumptions than the conventional reconstruction methods (e.g., Curvelet-domain and F-X domain data regularization methods). However, the traditional convolution methods are not suitable for the large percentages of missing data. In this study, we propose an improved partial convolution-based (PConv-based) deep-learning network to reconstruct the missing data, which is evolved from the conventional convolution-based (CConv-based) method. The U-net is used as deep learning network to analyze both PConv-based method and CConv-based method. The PConv-based method adopts a hierarchical, regional-learning mechanism to dynamically update the constrained convolution results for the sample matrix. Hence, the problem of poor amplitude preservation in the data reconstruction has been addressed when multiple consecutive traces are missing. The influence of data loss ratio on reconstruction algorithm is also discussed in this study. The numerical test demonstrates that the trained network is able to process a sample dataset with 50% data lost and largely eliminate the noises in the frequency-wavenumber domain caused by the missing data. This proposed method is further evaluated by actual data, and the results are better than those obtained from the Curvelet-domain method. Moreover, the dataset reconstructed by the PConv-based deep-learning network has a great agreement with the original dataset in terms of amplitude.</p>

geosciences, multidisciplinary,computer science, interdisciplinary applications

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the incomplete data in seismic data processing due to factors such as acquisition environment, equipment failure and economic cost. Specifically, the paper focuses on how to reconstruct the missing seismic data through data regularization methods, especially in situations where traditional methods are difficult to handle effectively when the proportion of missing data is large or there are continuous multi - trace missing. The paper proposes a deep - learning network based on Partial Convolution (PConv) to improve the deficiencies of traditional convolution methods in dealing with large - scale data missing, especially in maintaining amplitude consistency and reducing frequency - wavenumber domain (F - K domain) noise. This method can handle up to 50% data loss and performs better than the traditional Curvelet - domain method in actual data tests.

A partial convolution-based deep-learning network for seismic data regularization1

Intelligent Reconstruction for Spatially Irregular Seismic Data by Combining Compressed Sensing with Deep Learning

High-precision Seismic Data Reconstruction with Multi-Domain Sparsity Constraints Based on Curvelet and High-Resolution Radon Transforms

Convolutional Neural Network, Res‐Unet++, ‐based Dispersion Curve Picking from Noise Cross‐Correlations

Residual Learning of Deep Convolutional Neural Network for Seismic Random Noise Attenuation

Deep-seismic-prior-based reconstruction of seismic data using convolutional neural networks

Seismic Data Reconstruction via Wavelet-Based Residual Deep Learning

Seismic data recovery using deep targeted denoising priors in an alternating optimization framework

Attention and Hybrid Loss Guided 2-D Network for Seismic Impedance Inversion

Self-Supervised Multitask 3-D Partial Convolutional Neural Network for Random Noise Attenuation and Reconstruction in 3-D Seismic Data

A Deep-Learning-Based Generalized Convolutional Model for Seismic Data and Its Application in Seismic Deconvolution

Unsupervised Deep Learning for Seismic Data Reconstruction

A Convolutional Autoencoder Method for Simultaneous Seismic Data Reconstruction and Denoising

Ensemble Deep Learning for enhanced seismic data reconstruction

Depthwise separable convolution Unet for 3D seismic data interpolation

Reconstruct 3-D Seismic Data With Randomly Missing Traces via Fast Self-Supervised Deep Learning

Robust unsupervised 5D seismic data reconstruction on both regular and irregular grid

A Deep-Learning Neural Network for Postseismic Deformation Reconstruction From InSAR Time Series

Reconstructing missing seismic data using Deep Learning

Adaptive Convolutional Filter for Seismic Noise Attenuation