Improving faults continuity for extraction by transfer learning based on synthetic data
- Liyuan Xing (Norwegian University of Science and Technology) | Victor Aarre (Schlumberger AS) | Theoharis Theoharis (Norwegian University of Science and Technology)
- Document ID
- Society of Exploration Geophysicists
- 2018 SEG International Exposition and Annual Meeting, 14-19 October, Anaheim, California, USA
- Publication Date
- Document Type
- Conference Paper
- 2018. Society of Exploration Geophysicists
- Continuation, Machine learning, Faults
- 0 in the last 30 days
- 10 since 2007
- Show more detail
Fault interpretation requires significant manual effort. Automatic fault interpretation is the process of automatically interpreting fault surfaces from 3D seismic volumes and usually involves the following steps: (1) compute edge attributes, (2) calculate fault likelihoods and (3) extract fault surfaces. However, a human editing or merging procedure is usually demanded in step (3) to extract large surfaces from small discontinuous regions. A possible step toward improving region merging is to enhance the continuity (refers to the extent of the fault regions, not to the seismic reflection itself) beforehand. The proposed transfer learning method is trained on synthetic circle data with different connectivity which is balanced over all dips and curvatures, and applied to fault likelihood data with discontinuity and no labelling. Specifically, two neural networks (single and combined), including convolutional and fully connected layers, are investigated. The first aims to train directly from discontinuous to continuous circle data, while the second approaches the same objective in two steps: blurring (which improves continuity) and skeletonization. Three synthetic circle datasets (discontinuity, continuity and blur) are generated and fed into the networks for training accordingly, and the trained models are applied on real ant tracking faults for testing. Continuity is evaluated quantitatively both on the synthetic circle datasets and on real fault datasets, showing promising results. The approach is generalizable to the continuity of other geobodies (e.g. salt bodies, horizon).
Presentation Date: Tuesday, October 16, 2018
Start Time: 8:30:00 AM
Location: 204B (Anaheim Convention Center)
Presentation Type: Oral
|File Size||400 KB||Number of Pages||5|
Admasu,F.,S.Back, andK.Toennies,2006,Autotracking of faults on 3D seismic data:Geophysics,71,no.6,A49–A53,10.1190/1.2358399.
Barnes,A. E.,2006,A filter to improve seismic discontinuity data for fault interpretation:Geophysics,71,no.3,P1–P4,10.1190/1.2195988.
Bounaim,A.,T. H.Bø,W.Athmer,L.Sonneland, andO.Knoth,2013,Large fault extraction using point cloud approach to a seismic enhanced discontinuity cube:75th Annual International Conference and Exhibition, EAGE,Extended Abstracts,10.3997/2214-4609.20130750.
Gibson,D.,M.Spann,J.Turner, andT.Wright,2005,Fault surface detection in 3-D seismic data:IEEE Transactions on Geoscience and Remote Sensing,43,2094–2102,10.1109/TGRS.2005.852769.
Hale,D.,2013,Methods to compute fault images, extract fault surfaces, and estimate fault throws from 3D seismic images:Geophysics,78,no.2,O33–O43,10.1190/geo2012-0331.1.
Kadlec,B.,G.Dorn,H.Tufo, andD.Yuen,2008,Interactive 3-D computation of fault surfaces using level sets:Visual Geosciences,13,133–138,10.1007/s10069-008-0016-9.
Pedersen,S. I.,T.Randen,L.Sønneland, andø.Steen,2002,Automatic fault extraction using artificial ants:72th Annual International Meeting, SEG,Expanded Abstracts,512–515,10.1190/1.1817297.
Pedersen,S. I.,T.Skov,A.Hetlelid,P.Fayemendy,T.Randen, andL.Sønneland,2003,New paradigm of fault interpretation:73th Annual International Meeting, SEG,Expanded Abstracts,350–353,10.1190/1.1817918.
Randen,T.,S. I.Pedersen, andL.Sønneland,2001,Automatic extraction of fault surfaces from three dimensional seismic data:71th Annual International Meeting, SEG,Expanded Abstracts,551–554,10.1190/1.1816675.
Wikipedia,2018,Transfer learning,https://en.wikipedia.org/wiki/Transfer_learning,accessed 22 March 2018.