Seismic Stratum Segmentation Using an Encoder-Decoder Convolutional Neural Network

被引:15
作者
Wang, Detao [1 ,2 ]
Chen, Guoxiong [1 ,2 ]
机构
[1] China Univ Geosci, State Key Lab Geol Proc & Mineral Resources, Wuhan 430074, Peoples R China
[2] China Univ Geosci, Fac Earth Resources, Wuhan 430074, Peoples R China
基金
中国国家自然科学基金;
关键词
Seismic interpretation; Stratum segmentation; Deep learning; U-Net; FAULT-DETECTION;
D O I
10.1007/s11004-020-09916-8
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
As an essential step in reservoir characterization, seismic stratigraphic interpretation is often dependent on the development of powerful computer-based interpretation tools that can simulate the intelligence of experienced interpreters. With the success of machine/deep learning applications in many aspects of geoscience in recent decades, geophysicists have become more dedicated to exploring seismic big data in a smarter and more sophisticated way to better image subsurface reservoirs/structures. In this paper, a specific U-shaped fully convolutional network (U-Net) is established for automatic seismic stratigraphic interpretation. Specifically, this task is formulated as a semantic segmentation problem by identifying strata at the pixel level and classifying each pixel in the image into a specific stratum category. An experiment using the Netherlands F3 seismic dataset suggests that, compared with previously established deep learning models requiring a large number of training sets, the proposed U-Net method can achieve high evaluation indicators and better stratum segmentations in the case of a limited training set. During the test, the proposed U-Net model outperforms the Bayesian neural network (BNN) model for seismic stratum segmentation with regard to the training time, prediction speed, and segmentation accuracy. These results indicate the great potential of using U-Net-based deep learning for intelligent seismic stratigraphic interpretation.
引用
收藏
页码:1355 / 1374
页数:20
相关论文
共 33 条
[21]  
Nair Vinod., 2010, Proceedings of the 27th international conference on machine learning (ICML-10), P807, DOI [10.5555/3104322.3104425, DOI 10.5555/3104322.3104425]
[22]   Weakly- and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation [J].
Papandreou, George ;
Chen, Liang-Chieh ;
Murphy, Kevin P. ;
Yuille, Alan L. .
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2015, :1742-1750
[23]   U-Net: Convolutional Networks for Biomedical Image Segmentation [J].
Ronneberger, Olaf ;
Fischer, Philipp ;
Brox, Thomas .
MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION, PT III, 2015, 9351 :234-241
[24]  
Rumelhart DE, 1986, LEARNING INTERNAL RE, V1
[25]   SaltSeg: Automatic 3D salt segmentation using a deep convolutional neural network [J].
Shi, Yunzhi ;
Wu, Xinming ;
Fomel, Sergey .
INTERPRETATION-A JOURNAL OF SUBSURFACE CHARACTERIZATION, 2019, 7 (03) :SE113-SE122
[26]  
Silva, 2019, ARXIV190400770, P1
[27]  
Srivastava N, 2014, J MACH LEARN RES, V15, P1929
[28]  
Steventon M, 2019, 81 ANN INT C EXH EAG, V1, P1, DOI DOI 10.3997/2214-4609.201901510
[29]  
Sutton J, 2003, SEG TECHNICAL PROGRA, P2406
[30]   FaultSeg3D: Using synthetic data sets to train an end-to-end convolutional neural network for 3D seismic fault segmentation [J].
Wu, Xinming ;
Liang, Luming ;
Shi, Yunzhi ;
Fomel, Sergey .
GEOPHYSICS, 2019, 84 (03) :IM35-IM45