Iterative scheme inspired network for non-stationary random denoising
ZHANG Wenzheng1, TANG Jie1, LIU Yingchang1, MENG Tao1, CHEN Xueguo2
1. School of Geosciences, China University of Petroleum(East China), Qingdao, Shandong 266580, China; 2. Research Institute of Exploration & Production, SINOPEC Shengli Oilfield, Dongying, Shandong 257015, China
Abstract:Conventional filtering methods often magnify the influence of noise,which in return impedes the improvement of resolution and “smooths” discontinuous information in seismic data.We introduce a non-stationary random noise filtering method based on an iterative scheme-inspired network (ⅡN) which has a simple and tight structure and can be used to smooth non-stationary random noises.The L1 norm is used to optimize the objective function of the alternating directional multiplier algorithm which the ⅡN is derived from.A new auxiliary vari-able is added to transform the extreme value of the objective function into an augmented Lagrange form,and using the L-BFGS algorithm to distinguish and train all the network parameters.Finally an optimal denoising model is obtained.Applications to model and real data show that: ① the trained denoising model can effectively suppress noises while maintaining the characteristics of events according to the features of useful signals; and the simple and tight iterative network can speed up the rate of convergence and rapidly finish denoising and achieve expected results using a smaller database and shorter training time; ② the method proposed has a good adaptability and can suppress non-stationary random noises in conventional seismic data.
钟铁,李月,杨宝俊,等.陆地地震勘探随机噪声统计特性[J].地球物理学报,2017,60(2):655-664.ZHONG Tie,LI Yue,YANG Baojun,et al.Statistical features of the random noise in land seismic prospecting[J].Chinese Journal of Geophysics,2017,60(2):655-664.
[2]
邵婕,孙成禹,唐杰,等.基于字典训练的小波域稀疏表示微地震去噪方法[J].石油地球物理勘探,2016,51(2):254-260.SHAO Jie,SUN Chengyu,TANG Jie,et al.Micro-seismic data denoising based on sparse representations over learned dictionary in the wavelet domain[J].Oil Geophysical Prospecting,2016,51(2):254-260.
[3]
刘国昌,蔡加铭,闫海洋,等.利用稀疏约束非平稳多项式回归去除地震噪声及拾取初至[J].石油地球物理勘探,2020,55(3):548-556.LIU Guochang,CAI Jiaming,YAN Haiyang,et al.Using sparse-constrained nonstationary polynomial regression to remove seismic noises and picking up first arrival[J].Oil Geophysical Prospecting,2020,55(3):548-556.
[4]
李海山,陈德武,吴杰,等.叠前随机噪声深度残差网络压制方法[J].石油地球物理勘探,2020,55(3):493-503.LI Haishan,CHEN Dewu,WU Jie,et al.Pre-stack random noise suppression with deep residual network[J].Oil Geophysical Prospecting,2020,55(3):493-503.
[5]
唐杰,张文征,戚瑞轩,等.基于噪声水平估计的加权核范数最小化噪声压制方法研究[J].石油物探,2019,58(5):734-740.TANG Jie,ZHANG Wenzheng,QI Ruixuan,et al.Seismic data denoising by weighted nuclear norm mi-nimization based on noise estimation[J].Geophysical Prospecting for Petroleum,2019,58(5):734-740.
[6]
王伟,高静怀,陈文超,等.基于结构自适应中值滤波器的随机噪声衰减方法[J].地球物理学报,2012,55(5):1732-1741.WANG Wei,GAO Jinghuai,CHEN Wenchao,et al.Random seismic noise suppression via structure-adaptive median filter[J].Chinese Journal of Geophysics,2012,55(5):1732-1741.
[7]
张岩,任伟建,唐国维.利用多道相似组稀疏表示方法压制随机噪声[J].石油地球物理勘探,2017,52(3):442-450.ZHANG Yan,REN Weijian,TANG Guowei.Random noise suppression based on sparse representation of multi-trace similarity group[J].Oil Geophysical Prospecting,2017,52(3):442-450.
[8]
Nodes T,Gallagher N.Median filters:some modifications and their properties[J].IEEE Transactions Acoustics Speech Signal Processing,1982,30(5):739-746.
[9]
林红波,李月,徐学纯.压制地震勘探随机噪声的分段时频峰值滤波方法[J].地球物理学报,2011,54(5):1358-1366.LIN Hongbo,LI Yue,XU Xuechun.Segmenting time-frequency peak filtering method to attenuation of seismic random noise[J].Chinese Journal of Geophysics,2011,54(5):1358-1366.
[10]
唐杰,张文征,梁雨薇,等.自适应数据驱动的紧框架微地震数据随机噪声压制[J].石油地球物理勘探,2019,54(5):954-961.TANG Jie,ZHANG Wenzheng,LIANG Yuwei,et al.A random-noise suppression approach with self-adaptive data-driven tight frame for micro-seismic data[J].Oil Geophysical Prospecting,2019,54(5):954-961.
[11]
Deng X Y,Yang D,Liu T,et al.Study of least squares support vector regression filtering technology with a new 2D Ricker wavelet kernel[J].Journal of Seismic Exploration,2011,20(2):161-176.
[12]
Boashash B,Mesbah M.Signal enhancement by time-frequency peak filtering[J].IEEE Transactions on Signal Processing,2004,52(4):929-937.
[13]
林红波,马海涛,许丽萍.压制空间非平稳地震勘探随机噪声的ROAD径向时频峰值滤波算法[J].地球物理学报,2015,58(7):2546-2555.LIN Hongbo,MA Haitao,XU Liping.A radial time-frequency peak filtering based on ROAD for suppressing spatially nonstationary random noise in seismic data[J].Chinese Journal of Geophysics,2015,58(7):2546-2555.
[14]
Burger H C,Schuler C J,Harmeling S.Image denoising:Can plain neural networks compete with BM3D?[C].IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2012,2392-2399.
[15]
Schmidt U,Roth S.Shrinkage fields for effective ima-ge restoration[C].IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2014,2774-2781.
[16]
Chen Y J,Yu W,Pock T.On learning optimized reaction diffusion processes for effective image restoration[C].IEEE,Conference on Computer Vision & Patten Recognition,2015,5261-5269.
[17]
Mao X J,Shen C H,Yang Y B.Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections[C].NIPS,2016,2802-2810.
[18]
Zhang J,Ghanem B.ISTA-Net:Interpretable optimization-inspired deep network for image compressive sensing[C].IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2018,1828-1837.
[19]
Zhang M,Liu Y,Li G,et al.Iterative scheme-inspired network for impulse noise removal[J].Pattern Analysis and Applications,2018,23(2):135-145.
[20]
Mun S,Fowler J E.Block compressed sensing of images using directional transforms[C].IEEE International Conference on Image Processing,2010.
[21]
Beck A,Teboulle M.A fast iterative shrinkage-thre-sholding algorithm for linear inverse problems[J].SIAM Journal on Imaging Sciences,2009,2(1):183-202.
[22]
Blumensath T,Davies M E.Iterative hard thresho-lding for compressed sensing[J].Applied & Computational Harmonic Analysis,2009,27(3):265-274.
[23]
张雅晨,刘洋,刘财,等.低信噪比地震资料FDOC-Seislet变换阈值消噪方法研究[J].地球物理学报,2019,62(3):1181-1192.ZHANG Yachen,LIU Yang,LIU Cai,et al.Seismic random noise attenuation using FDOC-Seislet transform and threshold for seismic data with low SNR[J].Chinese Journal of Geophysics,2019,62(3):1181-1192.
[24]
Jonas A,Ozan O.Solving ill-posed inverse problems using iterative deep neural networks[J].Inverse Problems,2017,doi:10.1088/136-6420/aa9581.
[25]
Yan Y,Huibin L,Zongben X,et al.Deep ADMM-Net for compressive sensing MRI[C].NIPS,2016,10-18.
[26]
Boyd S,Parikh N,Chu E,et al.Distributed optimization and statistical learning via the alternating direction method of multipliers[J].Foundations & Trends in Machine Learning,2010,3(1):1-122.