The Weakly Supervised Learning Method for 3D Fault Detection

Donglin Zhu,Rui Guo,Lei Li,Shifan Zhan,Chunfeng Tao,Yingnan Gao
DOI: https://doi.org/10.1190/segam2021-3580914.1
2021-01-01
Abstract:PreviousNext No AccessFirst International Meeting for Applied Geoscience & Energy Expanded AbstractsThe weakly supervised learning method for 3D fault detectionAuthors: Donglin ZhuRui GuoLei LiShifan ZhanChunfeng TaoYingnan GaoDonglin ZhuBGP Inc., China National Petroleum CorporationSearch for more papers by this author, Rui GuoBGP Inc., China National Petroleum CorporationSearch for more papers by this author, Lei LiBGP Inc., China National Petroleum CorporationSearch for more papers by this author, Shifan ZhanBGP Inc., China National Petroleum CorporationSearch for more papers by this author, Chunfeng TaoBGP Inc., China National Petroleum CorporationSearch for more papers by this author, and Yingnan GaoBGP Inc., China National Petroleum CorporationSearch for more papers by this authorhttps://doi.org/10.1190/segam2021-3580914.1 SectionsSupplemental MaterialAboutPDF/ePub ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InRedditEmail AbstractIn this work, we propose a weakly supervised learning method which could utilize sparse manual interpretation results as training data for 3D fault detection task. Following a weakly supervised learning setting, we design the masked training data, which are gathered from field seismic volumes, and a sparse loss function for training process. Synthetic seismic data and field seismic volumes are applied to testify the proposed method. While we make no claim that these results from weakly supervised learning method are better than results predicted by full supervised methods, we believe that weakly supervised method can provide at least competitive results with the supervised models in the literature and highlight the potential of the weakly supervised learning framework.Keywords: faults, interpretation, 3DPermalink: https://doi.org/10.1190/segam2021-3580914.1FiguresReferencesRelatedDetails First International Meeting for Applied Geoscience & Energy Expanded AbstractsISSN (print):1052-3812 ISSN (online):1949-4645Copyright: 2021 Pages: 3561 publication data© 2021 Published in electronic format with permission by the Society of Exploration GeophysicistsPublisher:Society of Exploration Geophysicists HistoryPublished: 01 Sep 2021 CITATION INFORMATION Donglin Zhu, Rui Guo, Lei Li, Shifan Zhan, Chunfeng Tao, and Yingnan Gao, (2021), "The weakly supervised learning method for 3D fault detection," SEG Technical Program Expanded Abstracts : 1576-1579. https://doi.org/10.1190/segam2021-3580914.1 Plain-Language Summary Keywordsfaultsinterpretation3DPDF DownloadLoading ...
What problem does this paper attempt to address?