Wavelet shaping deconvolution based on deep learning
Ni Wenjun1, Liu Shaoyong1, Wang Liping1, Han Bingkai1, Sheng Shen2
1. School of Geophysics and Geomatics, China University of Geosciences, Wuhan, Hubei 430074, China; 2. School of Ocean and Earth Science, Tongji University, Shanghai 200092, China
Abstract：Seismic data migration imaging is one of the important methods for estimating the reflectivity of underground media.However，the imaging results are often affected by the wavelet，with limited wavenumber band distribution.Effectively extending the wavenumber band of the imaging results to improve the spatial resolution is a key objective in broadband reflectivity estimation.To achieve this，we firstly point out that the wavelet and the illumination of the geometry system are two important factors that affect the resolution of imaging results from an inversion imaging perspective.Then，based on convolutional neural networks （CNN），we use broadband wavelets to construct labels and employ conventional imaging results as input features to explore the mapping relationship using CNN.We also develop a corresponding deep learning algorithm，namely the wavelet shaping deconvolution method，and design a solution to the problem of inaccurate initial wavelet estimation in deconvolution by concatenating，iterating，and updating wavelets and reflectivity.Customized broadband wavelets can take into account both low-wavenumber and high-wavenumber information and can better restore broadband reflectivity during network training.Finally，we use a known model for network pre-training，extract effective wavelets based on the target data as the initial wavelets for deconvolution of the target data，carry out wavelet shaping deconvolution processing，and test the correctness and reliability of the method through thin-layer model testing.The filed data processing results indicate that this method has great potential for practical applications.
倪文军, 刘少勇, 王丽萍, 韩冰凯, 盛燊. 基于深度学习的子波整形反褶积方法[J]. 石油地球物理勘探, 2023, 58(6): 1313-1321.
Ni Wenjun, Liu Shaoyong, Wang Liping, Han Bingkai, Sheng Shen. Wavelet shaping deconvolution based on deep learning. Oil Geophysical Prospecting, 2023, 58(6): 1313-1321.
CAO J,WU R.Influence of propagator and acquisition aperture on image amplitude[C].SEG Technical Program Expanded Abstracts,2005,24: 1946-1949.
[9]
CAO J,WU R.Amplitude compensation for one-way wave propagators in inhomogeneous media and its application to seismic imaging[J].Communications in Computational Physics,2008,3(1): 203-221.
[10]
LIU W,ZHOU Y,YONG X,et al.A new true-amplitude imaging condition of reverse time migration for reflected seismic data[J].Chinese Journal of Geophysics,2020,63(9): 3442-3451.
[11]
YILMAZ O.Seismic deta processing[J].Investigation in Geophysics,1987,2: 526.
[12]
OPPENHEIM A,SCHAFER R.Homomorphic analysis of speech[J].IEEE Transactions on Audio and Electroacoustics,1968,16(2): 221-226.
[13]
ROSA A L R,ULRYCH T J.Processing via spectral modeling[J].Geophysics,1991,56(8): 1244-1251.
[14]
WIGGINS R A.Minimum entropy deconvolution[J].Geoexploration,1978,16(1/2): 21-35.
[15]
GRAY W C.Variable Norm Deconvolution[D].Stanford University,California,1979.
[16]
CANADAS G.A mathematical framework for blind deconvolution inverse problem[C].SEG Technical Program Expanded Abstracts,2002,21: 2202-2205.
康治梁,张雪冰.基于L1/2正则化理论的地震稀疏反褶积[J].石油物探,2019,58(6): 855-863.KANG Zhiliang,ZHANG Xuebing.Seismic sparse deconvolution based on L1/2 regularization[J].Geophysical Prospecting for Petroleum,2019,58(6): 855-863.
[19]
WANG L,ZHAO Q,GAO J,et al.Seismic sparse-spike deconvolution via Toeplitz-sparse matrix factorization[J].Geophysics,2016,81(3): V169-V182.
[20]
SUI Y,MA J.A nonstationary sparse spike deconvolution with an elastic attenuation[J].Geophysics,2019,84(2): R221-R234.
[21]
江雨濛,曹思远,陈思远,等.应用时变子波的盲反射系数反演[J].石油地球物理勘探,2021,56(5): 1001-1009.JIANG Yumeng,CAO Siyuan,CHEN Siyuan,et al.A blind deconvolution method based on the time-varying wavelet[J].Oil Geophysical Prospecting,2021,56(5): 1001-1009.
GAO Z,HU S,LI C,et al.A deep-learning-based generalized convolutional model for seismic data and its application in seismic deconvolution[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60: 1-17.
[24]
黄颜茹,高静怀,陈红灵,等.融入先验信息的深度学习地震子波振幅谱估计[J].石油地球物理勘探,2021,56(5): 969-978.HUANG Yanru,GAO Jinghuai,CHEN Hongling,et al.Estimating the amplitude spectrum of seismic wavelet via prior knowledge embedded deep learning[J].Oil Geophysical Prospecting,2021,56(5): 969-978.
[25]
AVILA M R V,OSORIO L N,DE CASTRO VARGAS FERNANDES J,et al.Migration deconvolution via deep learning[J].Pure and Applied Geophysics,2021,178(5): 1677-1695.
[26]
LIU C,SUN M,DAI N,et al.Deep learning-based point-spread function deconvolution for migration image deblurring[J].Geophysics,2022,87(4): S249-S265.
