Seismic Random Noise Removal Based on a Multiscale Convolution and Densely Connected Network for Noise Level Evaluation

Luo, Renze; Guo, Liang; Li, Xingyu; Zhou, Yang; Juanjuan, Tuo; Lei, CanRu

doi:10.1109/access.2022.3147242

Cited by 6 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, we adjust the dilation rate of the dilated convolution in the module. The purpose of seismic data denoising is to remove noise and recover effective signals as much as possible, according to the characteristics of random noise in seismic data [48]. Therefore, we propose MSDC-Unet model to fully extract noise features by using residual learning and a multi-scale convolution module.…”

Section: Parameters Selection In Multi-scale Convolution Modulementioning

confidence: 99%

A U-Net Based Multi-Scale Deformable Convolution Network for Seismic Random Noise Suppression

Zhao,

Zhou,

Bai

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Seismic data processing plays a key role in the field of geophysics. The collected seismic data are inevitably contaminated by various types of noise, which makes the effective signals difficult to be accurately discriminated. A fundamental issue is how to improve the signal-to-noise ratio of seismic data. Due to the complex characteristics of noise and signals, it is a challenge for the denoising model to suppress noise and recover weak signals. To suppress random noise in seismic data, we propose a multi-scale deformable convolution neural network denoising model based on U-Net, named MSDC-Unet. The MSDC-Unet mainly contains modules of deformable convolution and dilated convolution. The deformable convolution can change the shape of the convolution kernel to adjust the shape of seismic signals to fit different features, while the dilated convolution with different dilation rates is used to extract feature information at different scales. Furthermore, we combine Charbonnier loss and structure similarity index measure (SSIM) to better characterize geological structures of seismic data. Several examples of synthetic and field seismic data demonstrate that the proposed method is effective in the comprehensive results in terms of quantitative metrics and visual effect of denoising, compared with two traditional denoising methods and two deep convolutional neural network denoising models.

show abstract

Section: Parameters Selection In Multi-scale Convolution Modulementioning

confidence: 99%

A U-Net Based Multi-Scale Deformable Convolution Network for Seismic Random Noise Suppression

Zhao,

Zhou,

Bai

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In recent years, with the advancement of mathematical theory, computer hardware and technology, various artificial intelligence-based algorithms have been widely used in various fields [19]- [21]. In this field, the spectrum analysis has been re-focused.…”

Section: Introductionmentioning

confidence: 99%

Seismic Attribute Extraction and Application Based on the Gabor Wavelet Transform

Xiong,

Huang,

Guo

et al. 2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

The most common method for interpreting strata with seismic data is to relate the peaks and troughs of adjacent traces based on the seismic waveform characteristics. This can be captured by machine learning and deep learning methods to stratigraphic segmentation as many people investigated in industry. However, the spatial variability and instability of the peaks and troughs of seismic signals increases the difficulty of applying this technology. In addition, the nonlinear relationship and complicated subsurface geological setting make it more difficult. Thus, we propose a new seismic attribute extraction method based on the Gabor wavelet transform and linear dimensionality reduction. This method does not use the peaks or troughs of the seismic signal and instead focuses on the energy change in the seismic signal at the strata interface. It uses the characteristics of the energy change to identify strata. A sliding window Fourier transforms (STFTs) pretreatment is applied to convert the seismic signal to a spectrum energy form. On this basis, the local texture information of the spectrum can be processed by the Gabor wavelet transform to obtain the Gabor attribute of the seismic signal. The seismic Gabor attributes extracted using the above method contain time, frequency, and energy features, solving the problem of single seismic amplitude data features. Finally, the validity of the extracted seismic attributes is verified by a field data. In this process, the seismic amplitude, spectrum data and Gabor attributes are used as sample data for the support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost) models and deep residual shrinkage network (DRSN) for comparison. The results show that when the seismic Gabor attributes are used, the accuracy and root mean square error (RMSE) of the stratigraphic identification with the SVM, RF and XGBoost models are significantly better than those of the seismic amplitude and spectrum data only.

show abstract

“…D UE to the constraints on the acquisition condition, the cost limitations, and the dead traces, the received seismic data is often sampled [1], [2]. Sampled seismic data reconstruction is one of the key and tough tasks in seismic data processing [3], [4], which benefits further seismic data processing and interpretation, e.g., coherent and incoherent noise attenuation [5]- [9], geological structure characterization [10]- [12], attribute analysis [13]- [16], fault and horizon interpretation [17]- [19], lithology recognition [20]- [22], etc. The sampled seismic data can be mainly divided into randomly and successively sampled cases [23].…”

Section: Introductionmentioning

confidence: 99%

Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data

Gao

Gui

et al. 2023

IEEE Access

View full text Add to dashboard Cite

One of the major hot topics in seismic data processing is the reconstruction of successively sampled seismic data. There are numerous traditional methods proposed for addressing this issue; however, they still have unavoidable drawbacks, such as high computational cost and sensitive tuning parameters. In this study, we suggest a deep learning model for reconstructing successively sampled seismic data, termed fully connected U-Net (FCU-Net). FCU-Net maintains the high-resolution representations by connecting the parallel different-resolution representations and repeating multi-scale fusion. Such a structure allows FCU-Net to successfully extract multi-scale information, which is beneficial for accurate seismic data reconstruction. Additionally, the extending subnetwork of FCU-Net contains a large number of feature channels and sufficient information interaction between different resolution representations via the composite cascades, which contributes to locating successively sampled traces with big gaps and then performing the seismic interpolation. To verify the effectiveness of FCU-Net, we compare it with state-of-the-art networks, i.e., U-Net and HRNet, using synthetic and field examples. The results show that FCU-Net performs best when interpolating successively sampled seismic data, proving its superiority and availability.

show abstract

Seismic Random Noise Removal Based on a Multiscale Convolution and Densely Connected Network for Noise Level Evaluation

Cited by 6 publications

References 34 publications

A U-Net Based Multi-Scale Deformable Convolution Network for Seismic Random Noise Suppression

A U-Net Based Multi-Scale Deformable Convolution Network for Seismic Random Noise Suppression

Seismic Attribute Extraction and Application Based on the Gabor Wavelet Transform

Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data

Contact Info

Product

Resources

About