计算机科学 ›› 2021, Vol. 48 ›› Issue (8): 24-31.doi: 10.11896/jsjkx.200900034

• 数据库&大数据&数据科学* • 上一篇    下一篇

基于残差注意力网络的地震数据超分辨率方法

周文辉1,2, 石敏3, 朱登明1, 周军4   

  1. 1 中国科学院计算技术研究所 北京100190
    2 中国科学院大学 北京101408
    3 华北电力大学控制与计算机工程学院 北京102206
    4 中国石油集团测井有限公司技术中心 西安710065
  • 收稿日期:2020-09-04 修回日期:2020-11-17 发布日期:2021-08-10
  • 通讯作者: 朱登明(mdzhu@ict.ac.cn)
  • 基金资助:
    国家重大科技专项(2017ZX05019005)

Seismic Data Super-resolution Method Based on Residual Attention Network

ZHOU Wen-hui1,2, SHI Min3, ZHU Deng-ming1, ZHOU Jun4   

  1. 1 Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China;
    2 University of Chinese Academy of Sciences,Beijing 101408,China;
    3 College of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China;
    4 China National Petroleum Logging Co.,Ltd.Technology Center,Xi'an 710065,China
  • Received:2020-09-04 Revised:2020-11-17 Published:2021-08-10
  • About author:ZHOU Wen-hui,born in 1995,master.His main research interests include computer vision and computer gra-phics.(765647930@qq.com)ZHU Deng-ming,born in 1973,Ph.D,associate researcher,master supervisor,is a member of China Computer Federation.His main research interests include virtual reality and human-computer interaction.
  • Supported by:
    National Major Science and Technology Project(2017ZX05019005).

摘要: 地震数据在油气勘探、地质勘探领域发挥着至关重要的作用。精确详细的地震数据有助于对油气勘探做出精确指导,减小勘探的风险,从而产生巨大的社会效益和经济效益。在提升地震数据分辨率方面,现有的方法在面对海量数据时,在高分辨恢复、去噪性能和效率上效果欠佳,难以恢复出细节丰富的地质信息,无法满足实际需求。地震数据能够反映地质构造以及地层的组成,具有局部相关性高、全局相关性低的特点。同时,地震数据高频部分通常蕴含着地质勘探等重要信息,如分层、断层信息等。针对地震数据的特点,文中将地震数据重建问题转化为图像超分辨率问题,提出了采用基于生成对抗网络的地震数据超分辨方法。针对地震数据分布具有局部相关性高、全局相关性低的特点,设计残差注意力模块,挖掘地震数据的内在相关性,通过训练含有相对生成对抗损失函数的生成对抗网络模型,来对地震数据进行超分辨率恢复,以得到更加精确的地震数据。在真实的地震数据集上进行了实验验证,结果表明,所提方法在地震数据超分辨上效果良好,在性能指标PSNR和SSIM上有3%~4%的提升,具有较强的实用性。

关键词: 残差注意力模块, 超分辨率, 地震数据, 生成对抗网络, 相对生成对抗损失

Abstract: Seismic data plays a vital role in oil and gas exploration and geological exploration.Accurate and detailed seismic data can help to provide accurate guidance for oil and gas exploration,reduce the risk of exploration,and generate huge social and economic benefits.In terms of improving the resolution of seismic data,the existing methods are difficult to recover detailed geolo-gical information when facing large amounts of data,and have poor results in high-resolution recovery,denoising performance and efficiency,and it is difficult to meet the actual needs.Seismic data can reflect the composition of geological structures and strata,and have the characteristics of high local correlation and low global correlation.At the same time,the high frequency part of seismic data usually contain important geological exploration information such as layering,chasm,etc.In view of the characteristics of seismic data,this paper innovatively transforms the problem of seismic data reconstruction into the problem of image super-resolution,and proposes a method for super-resolution of seismic data based on the generative adversarial networks.In view of the characteristics of high local correlation and low global correlation of seismic data distribution,a residual attention module is designed to mine the inherent correlation of seismic data,so as to recover more refined seismic data.By training a generative adversarial network model with a relative generative adversarial loss function,the generative network is used to perform super-resolution recovery of the seismic data to recover more refined seismic data.In this paper,experimental verification is carried out on real seismic data,and the experimental results show that the proposed method has a good effect on super-resolution of seismic data and has strong practicability.

Key words: Relative generative adversarial loss, Generative adversarial network, Residual attention module, Seismic data, Super resolution

中图分类号: 

  • TP391.41
[1]JIAO H C.Research on seismic phase division method in 3Dseismic data[D].Chengdu:University of Electronic Science and Technology of China,2019.
[2]ZHAO N,YAN X P,HAO C L.Application of comprehensivegeological exploration methods in geological exploration[J].World Nonferrous Metals,2020(3):233,235.
[3]CHEN S X.Analysis on the development of petroleum geolo-gical seismic exploration technology[J].Petrochemical Technology,2019,26(3):193.
[4]SHANG J K.Application of three-dimensional seismic exploration technology under complex geological conditions[J].Jiangxi Coal Science and Technology,2020(3):143-146.
[5]ZHAN Z D,XU D P,WU H Y,et al.In-depth analysis of the accuracy of SEGY format seismic data conversion[J].Physical and Geochemical Exploration,2011,35(6):851-854.
[6]XIONG Z.Seismic exploration methods and techniques for stratigraphic and lithologic reservoirs[J].Petroleum Geophysical Prospecting,2012,47(1):1-18.
[7]REILLY D L,COOPER L N.An Overview of Neural Net-works:Early Models to Real World Systems[M].San Diego:Academic Press,2015:7-15.
[8]SPITZ S.Seismic trace interpolation in the F-X domain [J].Geophysics,1991,56(6):785.
[9]GULUNAY N,CHAMBERS R E.Unaliased fk domain trace interpolation[C]//SEG Technical Program Expanded Abstracts.Tulsa:Society of Exploration Geophysicists,1996:1461-1464.
[10]RONEN J.Wave-equation trace interpolation[J].Geophysics,2012,52(7):973-984.
[11]ZHAO Z Y,LI Z C,ZHANG M.Discrete orthogonal S-transform seismic data reconstruction using compressed sensing technology[J].Petroleum Geophysical Prospecting,2020,55(1):29-35.
[12]JIAO X M,DU Q Z,ZHAO Q.The compressive sensing seismic trace reconstruction method based on the maximum correlation entropy criterion[J].Journal of China University of Petroleum (Edition of Natural Science),2020,44(3):38-46.
[13]XU S,ZHANG Y,PHAM D,et al.Antileakage Fourier trans-form for seismic data regularizetion[J].Geophysics,2005,70(4):V87-V95.
[14]HUANG X W,WU L,NIU B H.Seismic trace reconstructionbased on parabolic Radon transformation[J].Journal of China University of Mining & Technology,2003(5):534-539.
[15]HERRMANN F J,HENNENFENT G.Non-parametric seismic data recovery with curvelet frames[J].Geophysical Journal of the Royal Astronomical Society,2010,173(1):233-248.
[16]LIU C,LI P,LIU Y,et al.Anti-aliasing iterative data interpolation method based on Seislet transform[J].Chinese Journal of Geophysics,2013,56(5):1619-1627.
[17]ZHANG L,HAN L G,XU D X,et al.Shearlet transform reconstruction of seismic data based on compressed sensing technology[J].Petroleum Geophysical Prospecting,2017,52(2):220-225.
[18]HUANG W,WU R S,WANG R.Damped dreamlet representation for exploration seismic data interpolation and denoising[J].IEEE Transactions on Geoscience & Remote Sensing,2018,56(6):3159-3172.
[19]DONG C,LOY C C,HE K,et al.Learning a deep convolutional network for image super-resolution[C]//IEEE International Conference on Computer Vision.2014.
[20]DONG C,LOY C C,HE K,et al.Image super-resolution using deep convolutional networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(2):295-307.
[21]KIM J,LEE J K,LEE K M.Accurate image super-resolutionusing very deep convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:1646-1654.
[22] LEDIG C,THEIS L,HUSZAR F,et al.Photo-Realistic SingleImage Super-Resolution Using a Generative Adversarial Network[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence.2017.
[23]ROLAND P.Cortical regulation of selective attention in man.A regional cerebral blood flow study[J].Journal of Neurophy-siology,1982,48(5):1059.
[24]MORAN J,DE SIMONE R.Selective attention gates visual processing in the extrastriate cortex[J].Science,1985,229(4715):782-784.
[25]MNIH V,HEESS N,GRAVES A,et al.Recurrent Models of Visual Attention[C]//Conference and Workshop on Neural Information Processing Systems.2014.
[26]MALINOWSKI M,DOERSCH C,SANTORO A,et al.Learning visual question answering by bootstrapping hard attention[C]//Proceedings of the European Conference on Computer Vision (ECCV).2018:3-20.
[27]MCCLENNY L,BRAGA-NETO U.Self-Adaptive Physics-Informed Neural Networks using a Soft Attention Mechanism[J].arXiv:2009.04544,2020.
[28]ZHANG B B,ZHANG Z J,FU J R,et al.Recovering low-frequency information using high-frequency components of seismic data[C]//2014 China Earth Sciences Joint Academic Annual Conference-Topic 18:Collected Papers on Reservoir Geophy-sics.2014.
[29]HE K,ZHANG X,REN S,et al.Deep residual learning forimage recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017.
[30]GOODFELLOW L,POUGET-ABADIE J,MIRZA M,et al.Generative adversary yal nets[C]//Conference and Workshop on Neural Information Processing Systems.2016.
[31]ZHAO Z,SUN Q,YANG H,et al.Compression artifacts reduction by improved generative adversarial networks[J].EURASIP Journal on Image and Video Processing,2019,2019(1).
[32]ZHAO Z S,GAO H X,SUN Q,et al.The theoretical framework of generative confrontation network,the latest development of derivative models and applications[J].Small Microcomputer System,2018,39(12):44-48.
[33]WANG X,YU K,WU S,et al.Esrgan:Enhanced super-resolution generative adversarial networks[C]//Proceedings of the European Conference on Computer Vision (ECCV).2018.
[34]ZHANG Y,LI K P,LI K,et al.Image super-resolution using very deep residual channel attention networks[C]//Proceedings of the European Conference on Computer Vision (ECCV).2018:286-301.
[1] 张佳, 董守斌.
基于评论方面级用户偏好迁移的跨领域推荐算法
Cross-domain Recommendation Based on Review Aspect-level User Preference Transfer
计算机科学, 2022, 49(9): 41-47. https://doi.org/10.11896/jsjkx.220200131
[2] 孙奇, 吉根林, 张杰.
基于非局部注意力生成对抗网络的视频异常事件检测方法
Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection
计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[3] 戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮.
基于DNGAN的磁共振图像超分辨率重建算法
Super-resolution Reconstruction of MRI Based on DNGAN
计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105
[4] 尹文兵, 高戈, 曾邦, 王霄, 陈怡.
基于时频域生成对抗网络的语音增强算法
Speech Enhancement Based on Time-Frequency Domain GAN
计算机科学, 2022, 49(6): 187-192. https://doi.org/10.11896/jsjkx.210500114
[5] 徐辉, 康金梦, 张加万.
基于特征感知的数字壁画复原方法
Digital Mural Inpainting Method Based on Feature Perception
计算机科学, 2022, 49(6): 217-223. https://doi.org/10.11896/jsjkx.210500105
[6] 高志宇, 王天荆, 汪悦, 沈航, 白光伟.
基于生成对抗网络的5G网络流量预测方法
Traffic Prediction Method for 5G Network Based on Generative Adversarial Network
计算机科学, 2022, 49(4): 321-328. https://doi.org/10.11896/jsjkx.210300240
[7] 黎思泉, 万永菁, 蒋翠玲.
基于生成对抗网络去影像的多基频估计算法
Multiple Fundamental Frequency Estimation Algorithm Based on Generative Adversarial Networks for Image Removal
计算机科学, 2022, 49(3): 179-184. https://doi.org/10.11896/jsjkx.201200081
[8] 周颖, 常明新, 叶红, 张燕.
基于元迁移的太阳能电池板缺陷图像超分辨率重建方法
Super Resolution Reconstruction Method of Solar Panel Defect Images Based on Meta-transfer
计算机科学, 2022, 49(3): 185-191. https://doi.org/10.11896/jsjkx.210100234
[9] 李建, 郭延明, 于天元, 武与伦, 王翔汉, 老松杨.
基于生成对抗网络的多目标类别对抗样本生成算法
Multi-target Category Adversarial Example Generating Algorithm Based on GAN
计算机科学, 2022, 49(2): 83-91. https://doi.org/10.11896/jsjkx.210800130
[10] 谈馨悦, 何小海, 王正勇, 罗晓东, 卿粼波.
基于Transformer交叉注意力的文本生成图像技术
Text-to-Image Generation Technology Based on Transformer Cross Attention
计算机科学, 2022, 49(2): 107-115. https://doi.org/10.11896/jsjkx.210600085
[11] 陈贵强, 何军.
自然场景下遥感图像超分辨率重建算法研究
Study on Super-resolution Reconstruction Algorithm of Remote Sensing Images in Natural Scene
计算机科学, 2022, 49(2): 116-122. https://doi.org/10.11896/jsjkx.210700095
[12] 冷佳旭, 王佳, 莫梦竟成, 陈泰岳, 高新波.
基于深度学习的视频超分辨率重构进展综述
Survey on Video Super-resolution Based on Deep Learning
计算机科学, 2022, 49(2): 123-133. https://doi.org/10.11896/jsjkx.211000007
[13] 石达, 芦天亮, 杜彦辉, 张建岭, 暴雨轩.
基于改进CycleGAN的人脸性别伪造图像生成模型
Generation Model of Gender-forged Face Image Based on Improved CycleGAN
计算机科学, 2022, 49(2): 31-39. https://doi.org/10.11896/jsjkx.210600012
[14] 唐雨潇, 王斌君.
基于深度生成模型的人脸编辑研究进展
Research Progress of Face Editing Based on Deep Generative Model
计算机科学, 2022, 49(2): 51-61. https://doi.org/10.11896/jsjkx.210400108
[15] 张玮琪, 汤轶丰, 李林燕, 胡伏原.
基于场景图的段落生成序列图像方法
Image Stream From Paragraph Method Based on Scene Graph
计算机科学, 2022, 49(1): 233-240. https://doi.org/10.11896/jsjkx.201100207
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!