A novel K-SVD dictionary learning approach for seismic data denoising
ZHOU Zixiang1,2, WU Juan1,2, YUAN Cheng3, BAI Min1,2, GUI Zhixian1,2
1. Key Laboratory of Exploration Technologies for Oil and Gas Resources (Yangtze University), Ministry of Education, Wuhan, Hubei 430100, China; 2. College of Geophysics and Petroleum Resources, Yangtze University, Wuhan, Hubei 430100, China; 3. Research Institute of Petroleum Exploration & Development-Northwest (NWGI), PetroChina, Lanzhou, Gansu 730020, China
Abstract：Denoising of seismic data is one of the key steps of seismic data processing, and high signal-to-noise ratio data are the basis of high-quality processing and interpretation. At present, there are many types of denoising methods, among which the sparse representation method can represent most or all of the original signal with less linear combination of basic signals. In other words, the information contained in the data is fully mined, and the advantages of the data are maximized. However, the K-SVD dictionary learning algorithm used for sparse representation has some problems such as the loss of the original signal in the denoising result, and its calculation efficiency is not ideal. In order to optimize these problems, a novel K-SVD dictionary learning method for seismic data denoising is studied in this paper. First, the dictionary is initialized by extracting randomly located blocks from the sample data and deleting the blank blocks. Then, dictionary learning is carried out, and the latest dictionary of sparse representation data is constructed adaptively from the features of seismic data. Finally, the dictionary is used to denoise the noisy seismic data in blocks, average the denoised blocks, reconstruct the image blocks to get the denoised images, and complete the denoising of seismic data. The experimental results of synthetic data and field data show that the denoising results of the novel K-SVD dictionary learning algorithm are more advantageous than those of the current K-SVD dictionary learning algorithm for sparse representation in the aspects of signal-to-noise ratio, computation time, and local feature preservation of seismic data.
李雪英, 孔祥琦, 侯相辉. 基于高通滤波的频率-空间域经验模态分解压制高频噪声[J]. 地球物理学进展, 2012, 27(3):1070-1077.LI Xueying, KONG Xiangqi, HOU Xianghui. High frequency noise suppression by empirical mode decomposition in the frequency-spatial domain based on high-pass filter[J]. Progress in Geophysics, 2012, 27(3):1070-1077.
[2]
LIU Y K. Noise reduction by vector median filtering[J]. Geophysics, 2013, 78(3):V79-V87.
[3]
路鹏飞, 郭爱华, 何月顺, 等. 曲波变换与傅里叶变换联合压制面波方法研究[J]. 地球物理学进展, 2020, 35(6):2181-2187.LU Pengfei, GUO Aihua, HE Yueshun, et al. Research on surface wave suppression combined with Curvelet transform and Fourier transform[J]. Progress in Geophysics, 2020, 35(6):2181-2187.
[4]
夏洪瑞, 朱勇, 周开明. 小波变换及其在去噪中的应用[J]. 石油地球物理勘探, 1994, 29(3):274-285.XIA Hongrui, ZHU Yong, ZHOU Kaiming. Wavelet transform and its application to noise elimination[J]. Oil Geophysical Prospecting, 1994, 29(3):274-285.
[5]
张华, 陈小宏, 杨海燕. 地震信号去噪的最优小波基选取方法[J]. 石油地球物理勘探, 2011, 46(1):70-75.ZHANG Hua, CHEN Xiaohong, YANG Haiyan. Optimistic wavelet basis selection in seismic signal noise elimination[J]. Oil Geophysical Prospecting, 2011, 46(1):70-75.
[6]
张华, 陈小宏, 李红星, 等. 曲波变换三维地震数据去噪技术[J]. 石油地球物理勘探, 2017, 52(2):226-232.ZHANG Hua, CHEN Xiaohong, LI Hongxing, et al. 3D seismic data denoising approach based on Curvelet transform[J]. Oil Geophysical Prospecting, 2017, 52(2):226-232.
[7]
刘洋, FOMELS, 刘财, 等. 高阶seislet变换及其在随机噪声消除中的应用[J]. 地球物理学报, 2009, 52(8):2142-2151.LIU Yang, FOMEL S, LIU Cai, et al. High-order seislet transform and its application of random noise attenuation[J]. Chinese Journal of Geophysics, 2009, 52(8):2142-2151.
[8]
陈毅军, 程浩, 巩恩普, 等. 基于Shearlet变换的尺度方向自适应阈值地震数据随机噪声压制方法[J]. 吉林大学学报(地球科学版), 2021, 51(4):1231-1242.CHEN Yijun, CHENG Hao, GONG Enpu, et al. Random noise suppression of seismic data with scale-oriented adaptive threshold based on shearlet transform[J]. Journal of Jilin University(Earth Science Edition), 2021, 51(4):1231-1242.
[9]
WANG C, ZHU Z, GU H, et al. Hankel low-rank approximation for seismic noise attenuation[J]. IEEE Transactions on Geoscience and Remote Sen-sing, 2019, 57(1):561-573.
[10]
LIU Z, Ma J, Yong X.Line survey joint denoi-sing via low-rank minimization[J]. Geophysics, 2019, 84(1):V21-V32.
[11]
练秋生, 石保顺, 陈书贞. 字典学习模型、算法及其应用研究进展[J]. 自动化学报, 2015, 41(2):240-260.LIAN Qiusheng, SHI Baoshun, CHEN Shuzhen.Research advances on dictionary learning models, algorithms and applications[J]. Acta Automatica Sinica, 2015, 41(2):240-260.
[12]
RUBINSTEIN R, ZIBULEVSKY M, ELAD M. Efficient implementation of the K-SVD algorithm using batch orthogonal matching pursuit[J]. Technical Report-CS Technion, 2008, 40(8):1-15.
[13]
贺月, 柳建新, 王显莹, 等. 基于自适应字典学习的插值去噪的应用[J]. 地球物理学进展, 2021, 36(6):2454-2461.HE Yue, LIU Jianxin, WANG Xianying, et al. Application of interpolation denoising based on adaptive dictionary learning[J]. Progress in Geophysics, 2021, 36(6):2454-2461.
[14]
FAOUZI ZIZI M O, TURQUAIS P. A dictionary learning method for seismic data compression[J]. Geophysics, 2022, 87(2):V101-V116.
[15]
张岩, 李新月, 王斌, 等. 基于深度学习的鲁棒地震数据去噪[J]. 石油地球物理勘探, 2022, 57(1):12-25.ZHANG Yan, LI Xinyue, WANG Bin, et al. Robust seismic data denoising based on deep learning[J]. Oil Geophysical Prospecting, 2022, 57(1):12-25.
[16]
武国宁, 于萌萌, 王君仙, 等. 应用平稳小波变换与深度残差网络压制地震随机噪声[J]. 石油地球物理勘探, 2022, 57(1):43-51.WU Guoning, YU Mengmeng, WANG Junxian, et al. Seismic random noise attenuation based on stationary wavelet transform and deep residual neural network[J]. Oil Geophysical Prospecting, 2022, 57(1):43-51.
[17]
刘璐, 刘洋, 刘财, 等. 地震随机噪声压缩感知迭代压制方法[J]. 地球物理学报, 2021, 64(12):4629-4643.LIU Lu, LIU Yang, LIU Cai, et al.Iterative seismic random noise suppression method based on compressive sensing[J]. Chinese Journal of Geophysics, 2021, 64(12):4629-4643.
[18]
陈兴飞, 孙红梅. 基于压缩感知理论的地震数据降噪方法[J]. 地球物理学进展, 2019, 34(3):1025-1031.CHEN Xingfei, SUN Hongmei.Seismic data denoising method based on compressed sensing theory[J]. Progress in Geophysics, 2019, 34(3):1025-1031.
[19]
ELAD M, AHARON M. Image denoising via sparse and redundant representations over learned dictionaries[J]. IEEE Transactions on Image Processing, 2006, 15(12):3736-3745.
[20]
OLSHAUSEN B A, FIELD D J.Emergence of simple-cell receptive field properties by learning a sparse code for natural images[J].Nature, 1996, 381(6583):607-609.
[21]
AHARON M, ELAD M, BRUCKSTEIN A. K-SVD:An algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11):4311-4322.
[22]
CHEN Y K. Fast dictionary learning for noise attenuation of multidimensional seismic data[J]. Geophysical Journal International, 2020, 222(1):1717-1727.
[23]
BRUCKSTEIN A M, DONOHO D L, ELAD M. From sparse solutions of systems of equations to sparse modeling of signals and images[J]. SIAM Review, 2009, 51(1):34-81.
[24]
DONOHO D L, ELAD M, TEMLYAKOV V N. Stable recovery of sparse overcomplete representations in the presence of noise[J]. IEEE Transactions on Information Theory, 2006, 52(1):6-18.
[25]
CHEN S S, DONOHO D L, SAUNDERS M A. Atomic decomposition by basis pursuit[J]. SIAM Journal on Scientific Computing, 1998, 20(1):33-61.
[26]
SMITH L N, ELAD M. Improving dictionary learning:multiple dictionary updates and coefficient reuse[J].IEEE Signal Processing Letters, 2013, 20(1):79-82.
[27]
SADEGHI M, BABAIE-ZADEH M, JUTTEN C. Dictionary learning for sparse representation:A novel approach[J]. IEEE Signal Processing Letters, 2013, 20(12):1195-1198.
[28]
RUSU C, DUMITRESCU B.Stagewise K-SVD to design efficient dictionaries for sparse representations[J]. IEEE Signal Processing Letters, 2012, 19(10):631-634.
[29]
BECKOUCHE S, MA J W. Simultaneous dictionary learning and denoising for seismic data[J]. Geophysics, 2014, 79(3):A27-A31.
[30]
思明波. 基于稀疏编码和字典重构的地震信号去噪方法研究[D]. 北京:中国石油大学(北京), 2020.SI Mingbo.Research on Seismic Signal Denoising Method Based on Sparse Coding and Dictionary ReconStruction[D].China University of Petroleum(Beijing), Beijing, 2020.
