1887
Volume 54, Issue 6
  • ISSN: 0812-3985
  • E-ISSN: 1834-7533

Abstract

Random noise, which has a significant impact on subsequent processing and interpretation, easily interferes with seismic data. Current convolutional neural networks (CNN) use a single-stage technique to boost network capacity by exploiting the complicated network structure, but the performance of the network becomes saturated and prone to overfitting at a certain stage. Hence, we propose a two-stage U-Net denoising network with a supervised attention module (UNet-SAM). In this supervised algorithm, the first stage obtains the pre-denoising results, while the second stage achieves more accurate data. The supervised attention module (SAM) block is inserted in the first stage, extracting features with supervised attention to utilise as a priori information and guide the fine denoising in the second stage. The combination of the attention mechanism and two-stage strategy provides prior information that helps to train a network with better denoising performance. Experiments on synthetic and field data illustrate that the proposed UNet-SAM not only has a superior denoising effect but also retains more of the original effective signal.

© 2023 Australian Society of Exploration Geophysicists
Loading

Article metrics loading...

/content/journals/10.1080/08123985.2023.2218870
2023-11-02
2026-01-13

  • From This Site

    /content/journals/10.1080/08123985.2023.2218870
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Anvari, R., A.R.Kahoo, M.S.Monfared, M.Mohammadi, R.M.D.Omer, and A.H.Mohammed. 2021. Random noise attenuation in seismic data using Hankel sparse low-rank approximation. Computers & Geosciences153: 104802.
     [Google Scholar]
  2. Chen, Y.2015. Deblending using a space-varying median filter. Exploration Geophysics46, no. 4: 332–341.
     [Google Scholar]
  3. Chen, L., X.Lu, J.Zhang, X.Chu, and C.Chen. 2021. HINet: Half instance normalization network for image restoration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 182–192.
     [Google Scholar]
  4. Cui, M., L.Fu, and W.Fang. 2022. Seismic noise attenuation via convolution neural network with learningnoise prior. Exploration Geophysics53, no. 1: 38–51.
     [Google Scholar]
  5. Duncan, G., and G.Beresford. 1995. Median filter behaviour with seismic data. Geophysical Prospecting43, no. 3: 329–345.
     [Google Scholar]
  6. Galbraith, J.N., and R.A.Wiggins. 1968. Characteristics of optimum multichannel stacking filters. Geophysics33, no. 1: 36–48.
     [Google Scholar]
  7. Gómez, J.L., and D.R.Velis. 2016. A simple method inspired by empirical mode decomposition for denoising seismic data. Geophysics81, no. 6: V403–V413.
     [Google Scholar]
  8. Guo, L.Y., C.Liu, Y.Liu, Z.-S.Zheng, and Q.-H.Wang. 2020. Seismic random noise attenuation based on streaming prediction filter in the f-x domain. Chinese Journal of Geophysics (in Chinese)63, no. 1: 329–338.
     [Google Scholar]
  9. Han, J., and M.van der Baan. 2015. Microseismic and seismic denoising via ensemble empirical mode decomposition and adaptive thresholding. Geophysics80, no. 6: KS69–KS80.
     [Google Scholar]
  10. Hao, S., J.Ma, and H.Yang. 2009. Comparisons of wavelets, contourlets and curvelets in seismic denoising. Journal of Applied Geophysics69, no. 2: 103–115.
     [Google Scholar]
  11. Hennenfent, G., L.Fenelon, and F.J.Herrmann. 2010. Nonequispaced curvelet transform for seismic data reconstruction: A sparsity-promoting approach. Geophysics75, no. 6: WB203–WB210.
     [Google Scholar]
  12. Jia, N., H.Ma, X.Dong, and Y.Li. 2020. Background noise suppression using trainable nonlinear reaction diffusion assisted by robust principal component analysis. Exploration Geophysics51, no. 6: 642–651.
     [Google Scholar]
  13. Jiang, X., S.Liu, X.Dai, G.Hu, X.Huang, Y.Yao, G.Xie, and L.Shao. 2022. Deep metric learning based on meta-mining strategy with semiglobal information. IEEE Transactions on Neural Networks and Learning Systems. doi:10.1109/TNNLS.2022.3202571.
     https://doi.org/10.1109/TNNLS.2022.3202571 [Google Scholar]
  14. Li, J., Y.Li, Y.Li, and Z.Qian. 2018. Downhole microseismic signal denoising via empirical wavelet transform and adaptive thresholding. Journal of Geophysics and Engineering15, no. 6: 2469–2480.
     [Google Scholar]
  15. Liu, X., X.Chen, J.Li, and Y.Chen. 2021. Nonlocal weighted robust principal component analysis for seismic noise attenuation. IEEE Transactions on Geoscience and Remote Sensing59, no. 2: 1745–1756.
     [Google Scholar]
  16. Luo, H., G.Lin, F.Shen, X.Huang, Y.Yao, and H.Shen. 2023. Robust-EQA: Robust learning for embodied question answering with noisy labels. IEEE Transactions on Neural Networks and Learning Systems. doi:10.1109/TNNLS.2023.3251984.
     https://doi.org/10.1109/TNNLS.2023.3251984 [Google Scholar]
  17. Ma, J.2018. Deep learning for attenuation random and coherence noise simultaneously. European Association of Geoscientists & Engineers,no.2: 1–5.
     [Google Scholar]
  18. Ma, J.T., J.H.Wang, and G.C.Liu. 2016. Seismic data noise attenuation and interpolation using singular value decomposition in frequency domain. Geophysical Prospecting for Petroleum55, no. 2: 205–213.
     [Google Scholar]
  19. Mafakheri, J., A.R.Kahoo, R.Anvari, M.Mohammadi, M.Radad, and M.S.Monfared. 2022. Expand dimensional of seismic data and random noise attenuation using low-rank estimation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing15: 2773–2781.
     [Google Scholar]
  20. Robinson, E.A., and S.Treitel. 1967. Principles of digital Wiener filtering. Geophysical Prospecting15, no. 3: 311–332.
     [Google Scholar]
  21. Saad, O.M., and Y.Chen. 2020. Deep denoising autoencoder for seismic random noise attenuation. Geophysics85, no. 4: V367–V376.
     [Google Scholar]
  22. Shen, H.Y., and Q.C.Li. 2010. SVD (singular value decomposition) seismic wave field noise elimination in frequency domain. Oil Geophysical Prospecting45, no. 2: 185–189.
     [Google Scholar]
  23. Wang, C., X.Huang, Y.Li, and K.Jensen. 2023. Removing multiple types of noise of distributed acoustic sensing seismic data using attention-guided denoising convolutional neural network. Frontiers in Earth Science. doi:10.3389/feart.2022.986470
     https://doi.org/10.3389/feart.2022.986470 [Google Scholar]
  24. Wang, J.J., Q.Li, X.H.Xu, and L.Cao. 2018. Random noise attenuation method of seismic signal based on the fractional order wavelet domain GSM model. Chinese Journal of Geophysics (in Chinese)61, no. 7: 2989–2997.
     [Google Scholar]
  25. Wang, Y.Q., W.K.Lu, J.L.Liu, M.Zhang, and Y.-K.Miao. 2019. Random seismic noise attenuation based on data augmentation and CNN. Chinese Journal of Geophysics (in Chinese)62, no. 1: 421–433.
     [Google Scholar]
  26. Yu, S., J.Ma, and W.Wang. 2019. Deep learning for denoising. Geophysics84, no. 6: V333–V350.
     [Google Scholar]
  27. Zamir, S.W., A.Arora, S.Khan S, M.Hayat, F.S.Khan, M.-H.Yang, and L.Shao. 2021. Multi-stage progressive image restoration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 14821–14831.
     [Google Scholar]
  28. Zhai, M.Y.2014. Seismic data denoising based on the fractional Fourier transformation. Journal of Applied Geophysics109: 62–70.
     [Google Scholar]
  29. Zhang, K., W.Zuo, Y.Chen, D.Meng, and L.Zhang. 2017. Beyond a Gaussian denoiser: Residual learning of deep CNN for image denoising. IEEE Transactions on Image Processing26, no. 7: 3142–3155.
     [Google Scholar]
  30. Zhang, K., W.Zuo, and L.Zhang. 2018. FFDnet: Toward a fast and flexible solution for CNN-based image denoising. IEEE Transactions on Image Processing27, no. 9: 4608–4622.
     [Google Scholar]
  31. Zhao, Y., Y.Li, X.Dong, and B.Yang. 2019. Low-frequency noise suppression method based on improved DnCNN in desert seismic data. IEEE Geoscience and Remote Sensing Letters16, no. 5: 811–815.
     [Google Scholar]
  32. Zhao, Y., Y.Li, and N.Wu. 2022. Coupled noise reduction in distributed acoustic sensing seismic data based on convolutional neural network. IEEE Geoscience and Remote Sensing Letters19: 1–5.
     [Google Scholar]
/content/journals/10.1080/08123985.2023.2218870
Loading
Seismic random noise attenuation via a two-stage U-net with supervised attention
Exploration Geophysics 54, 636 (2023); https://doi.org/10.1080/08123985.2023.2218870
/content/journals/10.1080/08123985.2023.2218870
/content/journals/10.1080/08123985.2023.2218870
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): neural networks; Noise attenuation; SAM; two-stage network

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error