Multi-attribute automatic interpretation of salt domes based on deep learning
ZHANG Yuxi1,2, LIU Yang1,2,3, ZHANG Haoran1,2, LYU Wenjie1, XUE Hao4
1. State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum(Beijing), Beijing 102249, China; 2. CNPC Key Laboratory of Geophysical Prospectiong, China University of Petroleum(Beijing), Beijing 102249, China; 3. Karamay Campus, China University of Petroleum(Beijing), Karamay, Xinjiang 834000, China; 4. CNOOC Research Institute Co., Beijing 100028, China
Abstract:It is difficult and inefficient for salt dome interpretation using 3D seismic data.A new workflow uses different seismic attributes to automatically inpterpret salt domes based on a small amount of 2D seismic data as training samples and testing models after deep learning.The wrokflow consists of three parts.First,according to the characteristics of a salt dome on seismic data,extract three types of sensitive attributes including chaotic and RMS amplitude,and variance.For each type of attribute,select a small amount of inline and time slices as training samples and use a data augmentation method to automatically generate massive samples.Second,construct a convolutional neural network based on anencoder-decoder architecture,and input two types of samples with different attributes for training and testing models to obtain multiple independent models.Finally,to comprehensively consider the features of all attributes and obtain more accurate classified results,use an ensemble learning method to merge the models and acquire optimized results.The results indicate that the boundaries of salt domes are clear and the classification errors can be significantly removed.This method can efficiently realize automatic segmentation of salt domes in 3D data setand further improve the prediction ability of models.
Brown A R.Interpretation of three-dimensional seismic data[J].Geological Magazine,1998,doi:http://dx.doi.org/.
[2]
Shi J,Malik J.Normalized cuts and image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905.
[3]
Lomask J,Clapp R G,Biondi B.Parallel implementa-tion of image segmentation for tracking 3D salt boundaries[C].Extended Abstracts of 68th EAGE Conference and Exhibition,2006,1567-1571.
[4]
Jing Z,Zhang Y,Zhang Z.Detecting boundary of salt dome in seismic data withedge-detection technique[C].SEG Technical Program Expanded Abstracts,2007,26:1392-1396.
[5]
Aqrawi A A,Boe T H,Barros S.Detecting saltdomes using a dip guided 3D sobel seismic attribute[C].SEG Technical Program Expanded Abstracts,2011,30:1014-1018.
[6]
Berthelot A,Solberg A H,Gelius L J.Texture attri-butes for detection of salt[J].Journal of Applied Geophysics,2013,doi:10.1016/j.jappgeo.2012.09.006.
[7]
Shafiq M A,Wang Z,Amin A,et al.Detection of salt-dome boundary surfaces in migrated seismic volumes using gradient of textures[C].SEG Technical Program Expanded Abstracts,2015,34:1811-1815.
[8]
王清振,张金淼,姜秀娣,等.利用梯度结构张量检测盐丘与断层[J].石油地球物理勘探,2018,53(4):826-831.WANG Qingzhen,ZHANG Jinmiao,JIANG Xiudi,et al.Salt dome and fault detection based on the gradient-structure tensor[J].Oil Geophysical Prospecting,2018,53(4):826-831.
[9]
Roberts A.Curvature attributes and their application to 3D interpreted horizons[J].First Break,2001,doi:10.1046/j.0263-5046.2001.00142.x.
[10]
Tingdahl K,de Rooij M.Semi-automatic detection of faults in 3D seismic data[J].Geophysical Prospecting,2005,53(4):533-542.
[11]
Pham N,Fomel S,Dunlap D.Automatic channel de-tection using deep learning[C].SEG Technical Program Expanded Abstracts,2018,37:2026-2030.
[12]
Di H,Wang Z,Alregib G.Real-time seismic image interpretation via deconvolutional neural network[C].SEG Technical Program Expanded Abstracts,2018,37:2051-2055.
[13]
Wu X,Fomel S.Automatic fault interpretation with optimal surface voting[J].Geophysics,2018,83(5):O67-O82.
[14]
Badrinarayanan V,Kendall V,Cipolla R.SegNet:A deep convolutional encoder-decoder architecture for scene segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481-2495.
[15]
彭文绪,王应斌,吴奎,等.盐构造的识别、分类及与油气的关系[J].石油地球物理勘探,2008,43(6):689-698.PENG Wenxu,WANG Yingbin,WU Kui,et al.Recognition and classification of salt structure and relative to oil/gas[J].Oil Geophysical Prospecting,2008,43(6):689-698.
[16]
Schroot B M,Schüttenhelm R T E.Expressions of shallow gas in the Netherlands North Sea[J].Netherlands Journal of Geosciences,2003,82(1):91-105.
[17]
Dasarathy B V.A composite classifier system design:Concepts and methodology[J].Proceedings of the IEEE,1979,67(5):708-713.
[18]
Littlestone N,Manfred K W.The weighted majority algorithm[J].Computer Engineering & Information Sciences,1994,108(2):212-261.