A physics-constrained sparse basis learning method for mixed noise suppression

Abstract

Suppressing complex mixed noise in seismic data poses a significant challenge for conventional methods, which often cause signal damage or leave residual noise. While sparse basis learning is a promising approach for this task, traditional data-driven learning methods are often insensitive to the physical properties of seismic signals, leading to incomplete noise removal and compromised signal fidelity. To address this limitation, we propose a physics-constrained sparse basis learning method for mixed noise suppression. Our method integrates local dip attributes—estimated and iteratively refined by a plane-wave destructor filter—as a physical constraint within the dictionary learning framework. This constraint guides the learning process to achieve high-fidelity signal reconstruction while effectively suppressing multiple noise types. Tests on complex synthetic and real data demonstrate that the proposed method outperforms conventional techniques and industry-standard workflows in attenuating mixed noise, including strong anomalous amplitudes, ground roll, and random and coherent components, thereby significantly enhancing the signal-to-noise ratio and imaging quality.

Keywords

Multiple-type noise suppression / Dictionary learning / Physical constraint / Plane-wave destructor filter

Funding

This work was supported by the China National Petroleum Corporation Science and Technology Special Project “Research on Risk Exploration Targets and Engineering Technology Breakthroughs in the Qaidam Basin, Including Field Trials” (2023YQX10108).

References

[1] Zhang H Chen, X, Li H, et al. 3D seismic data denoising approach based on curvelet transform. Oil Geophys Prospect. 2017; 52(2):226-232.

[2] Liu Y, Fomel S, Liu C, Wang D, Liu G, Feng X. High-order seislet transform and its application of random noise attenuation. Chin J Geophys. 2009; 52(8):2142-2151. doi:10.3969/j.issn.0001-5733.2009.08.024

[3] Zhang H, Chen X, Yang H. Optimistic wavelet basis selection in seismic signal noise elimination. Oil Geophys Prospect. 2011; 46(1):70-75.

[4] Wrinch D, Jeffreys H. On certain fundamental principles of scientific inquiry. Philos Mag. 1921; 42(250):369-390. doi:10.1080/14786442108633773

[5] Häuser M, Ma J. Interpolating seismic data via the POCS method based on shearlet transform. J Geophys Eng. 2012; 9(3):852-861.

[6] Herrmann FJ. Curvelet-based seismic data processing: A multiscale and nonlinear approach. Geophysics. 2006; 71(4):E11-E25.

[7] Herrmann FJ, Hennenfent G, Moghaddam PP. Compressive seismic data acquisition: An overview. Geophysics. 2008; 73(1):R23-R33.

[8] Huang X. A simulation of acquisition design and data processing for offshore compressive sensing seismic. Oil Geophys Prospect. 2020; 55(2):248-256.

[9] Starck JL, Candès EJ, Donoho DL. The curvelet transform for image denoising. IEEE Trans Image Process. 2002; 11(6):670-684. doi:10.1109/tip.2002.1014998

[10] Aharon M, Elad M, Bruckstein A. K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans Signal Process. 2006; 54(11):4311-4322. doi:10.1109/TSP.2006.881199

[11] Engan K, Aase SO, Husoy JH. Frame Based Signal Compression using Method of Optimal Directions (MOD). In: 1999 IEEE International Symposium on Circuits and Systems, ISCAS’99. Vol. 4. United States: IEEE; 1999. p. 244-247. doi:10.1109/ISCAS.1999.779928

[12] Liu L, Liu Y, Liu C, Zheng Z. Iterative seismic random noise suppression method based on compressive sensing. Chin J Geophys. 2021; 64(12):4629-4643. doi:10.6038/cjg2021P0045

[13] Zhang K, Zuo W, Chen Y, Meng D, Lei L. Beyond a Gaussian denoiser: Residual learning of deep CNN for image denoising. IEEE Trans Image Process. 2017; 26(7):3142-3155. doi:10.1109/TIP.2017.2662206

[14] Kaur R, Waldeland AU, Alaei R, Solberg AHS. Self-supervised learning for seismic denoising. IEEE Trans Geosci Remote Sens. 2022; 60:1-12.

[15] Song C, Alkhalifah T, Riyanti F. A physics-informed deep learning approach for denoising seismic data. Geophysics. 2022; 87(5):V239-V250.

[16] Sacchi MD, Ulrych TJ, Walker CJ. Interpolation and extrapolation using a high-resolution discrete Fourier transform. IEEE Trans Signal Process. 1998; 46(1):31-38. doi:10.1109/78.651165

[17] Yao Z, Sun C, Li H, Yang A. Time-variant wavelet extraction and seismic reflectivity inversion based on basis pursuit. Oil Geophys Prospect. 2019; 54(1):137-144.

[18] Candès EJ, Donoho DL. Curvelets-a new multiresolution representation for visual information. In: Wavelet Applications in Signal and Image Processing VIII. Vol. 4119. United States:SPIE; 2000. p. 248-262.

[19] Chen Y, Ma J, Fomel S. Seismic noise attenuation using a deep-learning approach with a local-correlation-based structural constraint. IEEE Trans Geosci Remote Sens. 2020; 58(7):4668-4679.

[20] Zhang L, AlRegib G. SSIM-based quality control for seismic processing. Geophysics. 2021; 86(3):V235-V248.

[21] Amini H, Javaherian A. A review of seismic attributes. J Appl Geophys. 2022; 205:104764.