A cyclic learning approach for improving pre-stack seismic processing
Abstract Current seismic processing workflows in the oil and gas industry involve several interactions between different experts to optimize the overall data quality in various tasks, such as noise attenuation, velocity analysis and horizon picking. While many machine learning-based approaches have...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2021-04-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-87794-8 |
| _version_ | 1818719122183159808 |
|---|---|
| author | Dario Augusto Borges Oliveira Daniela Szwarcman Rodrigo da Silva Ferreira Semen Zaytsev Daniil Semin |
| author_facet | Dario Augusto Borges Oliveira Daniela Szwarcman Rodrigo da Silva Ferreira Semen Zaytsev Daniil Semin |
| author_sort | Dario Augusto Borges Oliveira |
| collection | DOAJ |
| description | Abstract Current seismic processing workflows in the oil and gas industry involve several interactions between different experts to optimize the overall data quality in various tasks, such as noise attenuation, velocity analysis and horizon picking. While many machine learning-based approaches have been proposed to support each of those steps, most of them disregard expert interactions to guide the overall optimization. This paper presents geocycles, a cyclic learning approach that mimics this iterative process, which can be applied to different pre-stack seismic processing tasks. Our method refactor these processes considering training, testing, and evaluation sub-tasks, which allow the selection of samples for greedy sequential processes targeting an overall optimum quality for very large seismic datasets. We present encouraging results showing that a cyclic structure and efficient quality metrics improved overall outcomes in up to 128% for two different seismic processing tasks in comparison to a 1-cycle machine learning approach. |
| first_indexed | 2024-12-17T20:01:55Z |
| format | Article |
| id | doaj.art-3058d7ac66554cf2a714ebf9122e8ffb |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-12-17T20:01:55Z |
| publishDate | 2021-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-3058d7ac66554cf2a714ebf9122e8ffb2022-12-21T21:34:26ZengNature PortfolioScientific Reports2045-23222021-04-0111111310.1038/s41598-021-87794-8A cyclic learning approach for improving pre-stack seismic processingDario Augusto Borges Oliveira0Daniela Szwarcman1Rodrigo da Silva Ferreira2Semen Zaytsev3Daniil Semin4IBM ResearchIBM ResearchIBM ResearchGazprom NeftGazprom NeftAbstract Current seismic processing workflows in the oil and gas industry involve several interactions between different experts to optimize the overall data quality in various tasks, such as noise attenuation, velocity analysis and horizon picking. While many machine learning-based approaches have been proposed to support each of those steps, most of them disregard expert interactions to guide the overall optimization. This paper presents geocycles, a cyclic learning approach that mimics this iterative process, which can be applied to different pre-stack seismic processing tasks. Our method refactor these processes considering training, testing, and evaluation sub-tasks, which allow the selection of samples for greedy sequential processes targeting an overall optimum quality for very large seismic datasets. We present encouraging results showing that a cyclic structure and efficient quality metrics improved overall outcomes in up to 128% for two different seismic processing tasks in comparison to a 1-cycle machine learning approach.https://doi.org/10.1038/s41598-021-87794-8 |
| spellingShingle | Dario Augusto Borges Oliveira Daniela Szwarcman Rodrigo da Silva Ferreira Semen Zaytsev Daniil Semin A cyclic learning approach for improving pre-stack seismic processing Scientific Reports |
| title | A cyclic learning approach for improving pre-stack seismic processing |
| title_full | A cyclic learning approach for improving pre-stack seismic processing |
| title_fullStr | A cyclic learning approach for improving pre-stack seismic processing |
| title_full_unstemmed | A cyclic learning approach for improving pre-stack seismic processing |
| title_short | A cyclic learning approach for improving pre-stack seismic processing |
| title_sort | cyclic learning approach for improving pre stack seismic processing |
| url | https://doi.org/10.1038/s41598-021-87794-8 |
| work_keys_str_mv | AT darioaugustoborgesoliveira acycliclearningapproachforimprovingprestackseismicprocessing AT danielaszwarcman acycliclearningapproachforimprovingprestackseismicprocessing AT rodrigodasilvaferreira acycliclearningapproachforimprovingprestackseismicprocessing AT semenzaytsev acycliclearningapproachforimprovingprestackseismicprocessing AT daniilsemin acycliclearningapproachforimprovingprestackseismicprocessing AT darioaugustoborgesoliveira cycliclearningapproachforimprovingprestackseismicprocessing AT danielaszwarcman cycliclearningapproachforimprovingprestackseismicprocessing AT rodrigodasilvaferreira cycliclearningapproachforimprovingprestackseismicprocessing AT semenzaytsev cycliclearningapproachforimprovingprestackseismicprocessing AT daniilsemin cycliclearningapproachforimprovingprestackseismicprocessing |