基于数据增广和CNN的地震随机噪声压制
|
摘要
卷积神经网络(Convolutional Neural Network,CNN)是一种基于数据驱动的学习算法,简化了传统从特征提取到分类的两阶段式处理任务,被广泛应用于计算机科学的各个领域.在标注数据不足的地震数据去噪领域,CNN的推广应用受到限制.针对这一问题,本文提出了一种基于数据生成和增广的地震数据CNN去噪框架.对于合成数据,本文对无噪地震数据添加不同方差的高斯噪声,增广后构成训练集,实现基于小样本的CNN训练.对于实际地震数据,由于无法获得真实的干净数据和噪声来生成训练样本集,本文提出一种直接从无标签实际有噪数据生成标签数据集的方法.在所提出的方法中,我们利用目前已有的去噪方法从实际地震数据中分别获得估计干净数据和估计噪声,前者与未知的干净数据具有相似纹理,后者与实际噪声具有相似的概率分布.人工合成数据和实际数据实验结果表明,相较于F-X反褶积,BM3D和自适应频域滤波算法,本文方法能更好地压制随机噪声和保护有效信号.最后,本文采用神经网络可视化方法对去噪CNN的机理进行了探索,一定程度上解释了网络每一层的学习内容.
Convolutional neural network(CNN)has been widely adopted in various research fields of computer science.Combining the process of feature extracting and classification,CNN greatly simplifies traditional data processing task.However,as a data-driven algorithm,the generalization ability of CNN is limited in the problem of seismic noise attenuation which lacks labeled data.Tosolve this problem,we propose a CNN training framework based on data generation and augmentation for seismic noise attenuation.When processing synthetic data,we add Gaussian noise with different variance levels to clean seismic data and further augment training datasets to increase the diversity of features.For real seismic data,the clean data and corresponding noise are hard to acquire,thus we propose a method to generate labeled datasets directly from unlabeled noisy seismic data.In the proposed method,we apply existing denoising method to obtain the estimated clean data and estimated noise from real seismic data.The estimated data retains similar texture characteristics with clean data and the estimated noise has similar probability distribution with real seismic noise.We compare our method with F-X deconvolution,BM3D and adaptive frequency domain filtering method.The experiment results demonstrate that our method can efficiently attenuate random noise while preserving signals.Finally,we adopt neural network visualization methods to our CNN model and the visualization results explain the texture patterns learned by each layer of our network to some extent.
Convolutional neural network(CNN)has been widely adopted in various research fields of computer science.Combining the process of feature extracting and classification,CNN greatly simplifies traditional data processing task.However,as a data-driven algorithm,the generalization ability of CNN is limited in the problem of seismic noise attenuation which lacks labeled data.Tosolve this problem,we propose a CNN training framework based on data generation and augmentation for seismic noise attenuation.When processing synthetic data,we add Gaussian noise with different variance levels to clean seismic data and further augment training datasets to increase the diversity of features.For real seismic data,the clean data and corresponding noise are hard to acquire,thus we propose a method to generate labeled datasets directly from unlabeled noisy seismic data.In the proposed method,we apply existing denoising method to obtain the estimated clean data and estimated noise from real seismic data.The estimated data retains similar texture characteristics with clean data and the estimated noise has similar probability distribution with real seismic noise.We compare our method with F-X deconvolution,BM3D and adaptive frequency domain filtering method.The experiment results demonstrate that our method can efficiently attenuate random noise while preserving signals.Finally,we adopt neural network visualization methods to our CNN model and the visualization results explain the texture patterns learned by each layer of our network to some extent.
引文
Bonar D,Sacchi M.2012.Denoising seismic data using the nonlocal means algorithm.Geophysics,77(1):A5-A8,doi:10.1190/geo2011-0235.1.
Canales L L.1984.Random noise reduction.∥54th Ann.Internat Mtg.,Soc.Expi.Geophys..Expanded Abstracts,doi:10.1190/1.1894168.
Chaudhury K N,Sage D,Unser M.2011.Fast O(1)bilateral filtering using trigonometric range kernels.IEEE Transactions on Image Processing,20(12):3376-3382,doi:10.1109/TIP.2011.2159234.
Cheng S J,Han L G,Yu J L,et al.2018.Seismic data denoising based on improved K-SVD dictionary learning method.Global Geology(in Chinese),37(2):627-635.
Dabov K,Foi A,Katkovnik V,et al.2007.Image denoising by sparse 3-D transform-domain collaborative filtering.IEEETransactions on Image Processing,16(8):2080-2095,doi:10.1109/TIP.2007.901238.
Das V,Pollack A,Wollner U,et al.2018.Convolutional neural network for seismic impedance inversion.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.Expanded Abstracts,2071-2075.
Dong W S,Shi G M,Li X.2013.Nonlocal image restoration with bilateral variance estimation:A low-rank approach.IEEETransactions on Image Processing,22(2):700-711,doi:10.1109/TIP.2012.2221729.
Erhan D,Bengio Y,Courville A,et al.2009.Visualizing higherlayer features of a deep network.Technical report 1341,University of Montreal,1341(3):1.
Fomel S,Liu Y.2010.Seislet transform and seislet frame.Geophysics,75(3):V25-V38.
Freire S L,Ulrych T J.1988.Application of singular value decomposition to vertical seismic profiling.Geophysics,53(6):778-785,doi:10.1190/1.1442513.
Gilboa G,Zeevi Y Y,Sochen N.2003.Texture preserving variational denoising using an adaptive fidelity term.∥Proceedings of the VLSM.Nice,France.
Gulunay N.1986.FXDECON and complex Wiener prediction filter.∥56th Ann.Internat Mtg.,Soc.Expi.Geophys.,Expanded Abstracts,doi:10.1190/1.1893128.
He K M,Zhang X Y,Ren S Q,et al.2016.Deep residual learning for image recognition.∥2016IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,NV,USA:IEEE770-778,doi:10.1109/CVPR.2016.90.
Hennenfent G,Herrmann F J.2006.Seismic denoising with nonuniformly sampled curvelets.Computing in Science&Engineering,8(3):16-25,doi:10.1109/MCSE.2006.49.
Hornbostel S C.1991.Spatial prediction filtering in the t-xand f-x domains.Geophysics,56(12):2019-2026,doi:10.1190/1.1443014.
Hu J,Shen L,Albanie S,et al.2017.Squeeze-and-excitation networks.arXiv preprint arXiv:1709.01507.
Huang G,Liu Z,Van Der Maaten L,et al.2017.Densely connected convolutional networks.∥2017IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,HI,USA:IEEE,doi:10.1109/CVPR.2017.243.
Kingma D P,Ba J.2014.Adam:A method for stochastic optimization.arXiv preprint arXiv:1412.6980.
Krizhevsky A,Sutskever I,Hinton G E.2012.Imagenet classification with deep convolutional neural networks.∥Proceedings of the 25th International Conference on Neural Information Processing Systems.Lake Tahoe,Nevada:Curran Associates Inc.,doi:10.1145/3065386.
Liu J,Lu W,Zhang P.2018.Random noise attenuation using convolutional neural networks.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.Expanded Abstracts.
Liu X,He Z H,Huang D J.2006.Seismic data denoising research based on wavelet transform.Petroleum Geophysics(in Chinese),4(4):15-18.
Lu W K.2006.Adaptive noise attenuation of seismic images based on singular value decomposition and texture direction detection.Journal of Geophysics and Engineering,3(1):28-34,doi:10.1088/1742-2132/3/1/004.
Ma J.2018.Deep learning for attenuating random and coherence noise simultaneously.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.,Expanded Abstracts.
Mordvintsev A,Olah C,Tyka M.2015.Inceptionism:Going deeper into neural networks.Google Research Blog.
Odena A,Dumoulin V,Olah C.2016.Deconvolution and checkerboard artifacts.Distill,doi:10.23915/distill.
Peng C,Chang Z,Zhu S J.2008.Noise elimination method based on curvelet transform.Geophysical Prospecting for Petroleum(in Chinese),47(5):461-464,doi:10.3969/j.issn.1000-1441.2008.05.006.
Ronneberger O,Fischer P,Brox T.2015.U-net:Convolutional networks for biomedical image segmentation.∥International Conference on Medical Image Computing and Computer-Assisted Intervention,Munich,Germany:Springer.
Rudin L I,Osher S,Fatemi E.1992.Nonlinear total variation based noise removal algorithms.Physica D:Nonlinear Phenomena,60(1-4):259-268,doi:10.1016/0167-2789(92)90242-F.
Shao J,Sun C Y,Tang J,et al.2016.Micro-seismic data denoising based on sparse representations over learned dictionary in the wavelet domain.Oil Geophysical Prospecting(in Chinese),51(2):254-260,doi:10.13810/j.cnki.issn.1000-7210.2016.02.007.
Shelhamer E,Long J,Darrell T.2014.Fully convolutional networks for semantic segmentation.IEEE Transactions on Pattern Analysis and Machine Intelligence,39(4):640-651,doi:10.1109/TPAMI.2016.2572683.
Simonyan K,Zisserman A.2014.Very deep convolutional networks for large-scale image recognition.arXiv preprint arXiv:1409.1556.
Springenberg J T,Dosovitskiy A,Brox T,et al.2014.Striving for simplicity:The all convolutional net.arXiv preprint arXiv:1412.6806.
Tang G,Ma J W,Yang H Z.2012.Seismic data denoising based on learning-type overcomplete dictionaries.Applied Geophysics,9(1):27-32,doi:10.1007/s11770-012-0310-z.
Wang B,Zhang N,Lu W,et al.2018.Deep learning based seismic data interpolation:A preliminary result.Geophysics,84(1):11-20.
Wang J J,Yuan L,Liu W R,et al.2016.Dual-tree complex wavelet domain bivariate method for seismic signal random noise attenuation.Chinese Journal of Geophysics(in Chinese),59(8):3046-3055,doi:10.6038/cjg20160827.
Xiong W,Ji X,Ma Y,et al.2018.Seismic fault detection with convolutional neural network.Geophysics,83(5):O97-O103.
Zhang K,Zuo W M,Chen Y J,et al.2017a.Beyond a gaussian denoiser:Residual learning of deep cnn for image denoising.IEEE Transactions on Image Processing,26(7):3142-3155,doi:10.1109/TIP.2017.2662206.
Zhang K,Zuo W M,Zhang L.2018.FFDnet:Toward a fast and flexible solution for CNN-based image denoising.IEEE Transactions on Image Processing,27(9):4608-4622,doi:10.1109/TIP.2018.2839891.
Zhang R F,Ulrych T J.2003.Physical wavelet frame denoising.Geophysics,68(1):225-231,doi:10.1190/1.1543209.
Zhang X Y,Zhou X Y,Lin M X,et al.2017b.Shufflenet:An extremely efficient convolutional neural network for mobile devices.arXiv preprint arXiv:1707.01083.
Zhou J X,Lu W K,He J W,et al.2015.A data-dependent fourier filter based on image segmentation for random seismic noise attenuation.Journal of Applied Geophysics,114:224-231,doi:10.1016/j.jappgeo.2015.01.020.
Zhu L C,Liu E T,McClellan J H.2015.Seismic data denoising through multiscale and sparsity-promoting dictionary learning.Geophysics,80(6):WD45-WD57,doi:10.1190/geo2015-0047.1.
Zhu L C,Liu E T,McClellan J H.2017.Joint seismic data denoising and interpolation with double-sparsity dictionary learning.Journal of Geophysics and Engineering,14(4):802-810,doi:10.1088/1742-2140/aa6491.
程时俊,韩立国,于江龙等.2018.基于改进K-SVD字典学习方法的地震数据去噪.世界地质,37(2):627-635.
刘鑫,贺振华,黄德济.2006.基于小波变换的地震资料去噪处理研究.油气地球物理,4(4):15-18.
彭才,常智,朱仕军.2008.基于曲波变换的地震数据去噪方法.石油物探,47(5):461-464,doi:10.3969/j.issn.1000-1441.2008.05.006.
邵婕,孙成禹,唐杰等.2016.基于字典训练的小波域稀疏表示微地震去噪方法.石油地球物理勘探,51(2):254-260,doi:10.13810/j.cnki.issn.1000-7210.2016.02.007.
汪金菊,袁力,刘婉如等.2016.地震信号随机噪声压制的双树复小波域双变量方法.地球物理学报,59(8):3046-3055,doi:10.6038/cjg20160827.
Canales L L.1984.Random noise reduction.∥54th Ann.Internat Mtg.,Soc.Expi.Geophys..Expanded Abstracts,doi:10.1190/1.1894168.
Chaudhury K N,Sage D,Unser M.2011.Fast O(1)bilateral filtering using trigonometric range kernels.IEEE Transactions on Image Processing,20(12):3376-3382,doi:10.1109/TIP.2011.2159234.
Cheng S J,Han L G,Yu J L,et al.2018.Seismic data denoising based on improved K-SVD dictionary learning method.Global Geology(in Chinese),37(2):627-635.
Dabov K,Foi A,Katkovnik V,et al.2007.Image denoising by sparse 3-D transform-domain collaborative filtering.IEEETransactions on Image Processing,16(8):2080-2095,doi:10.1109/TIP.2007.901238.
Das V,Pollack A,Wollner U,et al.2018.Convolutional neural network for seismic impedance inversion.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.Expanded Abstracts,2071-2075.
Dong W S,Shi G M,Li X.2013.Nonlocal image restoration with bilateral variance estimation:A low-rank approach.IEEETransactions on Image Processing,22(2):700-711,doi:10.1109/TIP.2012.2221729.
Erhan D,Bengio Y,Courville A,et al.2009.Visualizing higherlayer features of a deep network.Technical report 1341,University of Montreal,1341(3):1.
Fomel S,Liu Y.2010.Seislet transform and seislet frame.Geophysics,75(3):V25-V38.
Freire S L,Ulrych T J.1988.Application of singular value decomposition to vertical seismic profiling.Geophysics,53(6):778-785,doi:10.1190/1.1442513.
Gilboa G,Zeevi Y Y,Sochen N.2003.Texture preserving variational denoising using an adaptive fidelity term.∥Proceedings of the VLSM.Nice,France.
Gulunay N.1986.FXDECON and complex Wiener prediction filter.∥56th Ann.Internat Mtg.,Soc.Expi.Geophys.,Expanded Abstracts,doi:10.1190/1.1893128.
He K M,Zhang X Y,Ren S Q,et al.2016.Deep residual learning for image recognition.∥2016IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,NV,USA:IEEE770-778,doi:10.1109/CVPR.2016.90.
Hennenfent G,Herrmann F J.2006.Seismic denoising with nonuniformly sampled curvelets.Computing in Science&Engineering,8(3):16-25,doi:10.1109/MCSE.2006.49.
Hornbostel S C.1991.Spatial prediction filtering in the t-xand f-x domains.Geophysics,56(12):2019-2026,doi:10.1190/1.1443014.
Hu J,Shen L,Albanie S,et al.2017.Squeeze-and-excitation networks.arXiv preprint arXiv:1709.01507.
Huang G,Liu Z,Van Der Maaten L,et al.2017.Densely connected convolutional networks.∥2017IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,HI,USA:IEEE,doi:10.1109/CVPR.2017.243.
Kingma D P,Ba J.2014.Adam:A method for stochastic optimization.arXiv preprint arXiv:1412.6980.
Krizhevsky A,Sutskever I,Hinton G E.2012.Imagenet classification with deep convolutional neural networks.∥Proceedings of the 25th International Conference on Neural Information Processing Systems.Lake Tahoe,Nevada:Curran Associates Inc.,doi:10.1145/3065386.
Liu J,Lu W,Zhang P.2018.Random noise attenuation using convolutional neural networks.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.Expanded Abstracts.
Liu X,He Z H,Huang D J.2006.Seismic data denoising research based on wavelet transform.Petroleum Geophysics(in Chinese),4(4):15-18.
Lu W K.2006.Adaptive noise attenuation of seismic images based on singular value decomposition and texture direction detection.Journal of Geophysics and Engineering,3(1):28-34,doi:10.1088/1742-2132/3/1/004.
Ma J.2018.Deep learning for attenuating random and coherence noise simultaneously.∥88th Ann.Internat Mtg.,Soc.Expi.Geophys.,Expanded Abstracts.
Mordvintsev A,Olah C,Tyka M.2015.Inceptionism:Going deeper into neural networks.Google Research Blog.
Odena A,Dumoulin V,Olah C.2016.Deconvolution and checkerboard artifacts.Distill,doi:10.23915/distill.
Peng C,Chang Z,Zhu S J.2008.Noise elimination method based on curvelet transform.Geophysical Prospecting for Petroleum(in Chinese),47(5):461-464,doi:10.3969/j.issn.1000-1441.2008.05.006.
Ronneberger O,Fischer P,Brox T.2015.U-net:Convolutional networks for biomedical image segmentation.∥International Conference on Medical Image Computing and Computer-Assisted Intervention,Munich,Germany:Springer.
Rudin L I,Osher S,Fatemi E.1992.Nonlinear total variation based noise removal algorithms.Physica D:Nonlinear Phenomena,60(1-4):259-268,doi:10.1016/0167-2789(92)90242-F.
Shao J,Sun C Y,Tang J,et al.2016.Micro-seismic data denoising based on sparse representations over learned dictionary in the wavelet domain.Oil Geophysical Prospecting(in Chinese),51(2):254-260,doi:10.13810/j.cnki.issn.1000-7210.2016.02.007.
Shelhamer E,Long J,Darrell T.2014.Fully convolutional networks for semantic segmentation.IEEE Transactions on Pattern Analysis and Machine Intelligence,39(4):640-651,doi:10.1109/TPAMI.2016.2572683.
Simonyan K,Zisserman A.2014.Very deep convolutional networks for large-scale image recognition.arXiv preprint arXiv:1409.1556.
Springenberg J T,Dosovitskiy A,Brox T,et al.2014.Striving for simplicity:The all convolutional net.arXiv preprint arXiv:1412.6806.
Tang G,Ma J W,Yang H Z.2012.Seismic data denoising based on learning-type overcomplete dictionaries.Applied Geophysics,9(1):27-32,doi:10.1007/s11770-012-0310-z.
Wang B,Zhang N,Lu W,et al.2018.Deep learning based seismic data interpolation:A preliminary result.Geophysics,84(1):11-20.
Wang J J,Yuan L,Liu W R,et al.2016.Dual-tree complex wavelet domain bivariate method for seismic signal random noise attenuation.Chinese Journal of Geophysics(in Chinese),59(8):3046-3055,doi:10.6038/cjg20160827.
Xiong W,Ji X,Ma Y,et al.2018.Seismic fault detection with convolutional neural network.Geophysics,83(5):O97-O103.
Zhang K,Zuo W M,Chen Y J,et al.2017a.Beyond a gaussian denoiser:Residual learning of deep cnn for image denoising.IEEE Transactions on Image Processing,26(7):3142-3155,doi:10.1109/TIP.2017.2662206.
Zhang K,Zuo W M,Zhang L.2018.FFDnet:Toward a fast and flexible solution for CNN-based image denoising.IEEE Transactions on Image Processing,27(9):4608-4622,doi:10.1109/TIP.2018.2839891.
Zhang R F,Ulrych T J.2003.Physical wavelet frame denoising.Geophysics,68(1):225-231,doi:10.1190/1.1543209.
Zhang X Y,Zhou X Y,Lin M X,et al.2017b.Shufflenet:An extremely efficient convolutional neural network for mobile devices.arXiv preprint arXiv:1707.01083.
Zhou J X,Lu W K,He J W,et al.2015.A data-dependent fourier filter based on image segmentation for random seismic noise attenuation.Journal of Applied Geophysics,114:224-231,doi:10.1016/j.jappgeo.2015.01.020.
Zhu L C,Liu E T,McClellan J H.2015.Seismic data denoising through multiscale and sparsity-promoting dictionary learning.Geophysics,80(6):WD45-WD57,doi:10.1190/geo2015-0047.1.
Zhu L C,Liu E T,McClellan J H.2017.Joint seismic data denoising and interpolation with double-sparsity dictionary learning.Journal of Geophysics and Engineering,14(4):802-810,doi:10.1088/1742-2140/aa6491.
程时俊,韩立国,于江龙等.2018.基于改进K-SVD字典学习方法的地震数据去噪.世界地质,37(2):627-635.
刘鑫,贺振华,黄德济.2006.基于小波变换的地震资料去噪处理研究.油气地球物理,4(4):15-18.
彭才,常智,朱仕军.2008.基于曲波变换的地震数据去噪方法.石油物探,47(5):461-464,doi:10.3969/j.issn.1000-1441.2008.05.006.
邵婕,孙成禹,唐杰等.2016.基于字典训练的小波域稀疏表示微地震去噪方法.石油地球物理勘探,51(2):254-260,doi:10.13810/j.cnki.issn.1000-7210.2016.02.007.
汪金菊,袁力,刘婉如等.2016.地震信号随机噪声压制的双树复小波域双变量方法.地球物理学报,59(8):3046-3055,doi:10.6038/cjg20160827.