Show simple item record

dc.contributor.authorOvcharenko, Oleg
dc.contributor.authorKazei, Vladimir
dc.contributor.authorPeter, Daniel
dc.contributor.authorSilvestrov, Ilya
dc.contributor.authorBakulin, Andrey
dc.contributor.authorAlkhalifah, Tariq Ali
dc.date.accessioned2021-09-06T06:37:51Z
dc.date.available2021-09-06T06:37:51Z
dc.date.issued2021-09-01
dc.identifier.citationOvcharenko, O., Kazei, V., Peter, D., Silvestrov, I., Bakulin, A., & Alkhalifah, T. (2021). Dual-band generative learning for low-frequency extrapolation in seismic land data. First International Meeting for Applied Geoscience & Energy Expanded Abstracts. doi:10.1190/segam2021-3579442.1
dc.identifier.doi10.1190/segam2021-3579442.1
dc.identifier.urihttp://hdl.handle.net/10754/670950
dc.description.abstractThe presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia. The presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia.
dc.description.sponsorshipThe research reported in this publication was supported by funding from Saudi Aramco and King Abdullah University of Science and Technology (KAUST).
dc.publisherSociety of Exploration Geophysicists
dc.relation.urlhttps://library.seg.org/doi/10.1190/segam2021-3579442.1
dc.rightsArchived with thanks to Society of Exploration Geophysicists
dc.titleDual-band generative learning for low-frequency extrapolation in seismic land data
dc.typeConference Paper
dc.contributor.departmentEarth Science and Engineering Program
dc.contributor.departmentExtreme Computing Research Center
dc.contributor.departmentPhysical Science and Engineering (PSE) Division
dc.contributor.departmentSeismic Wave Analysis Group
dc.eprint.versionPre-print
dc.contributor.institutionAramco Services Company
dc.contributor.institutionSaudi Aramco
kaust.personOvcharenko, Oleg
kaust.personPeter, Daniel
kaust.personAlkhalifah, Tariq Ali
refterms.dateFOA2021-09-07T06:29:58Z
dc.date.published-online2021-09-01
dc.date.published-print2021-09-01


Files in this item

Thumbnail
Name:
seg_2021_dual.pdf
Size:
2.899Mb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record