Cargando…

Diffusion Model for DAS-VSP Data Denoising

Distributed acoustic sensing (DAS) has emerged as a transformational technology for seismic data acquisition. However, noise remains a major impediment, necessitating advanced denoising techniques. This study pioneers the application of diffusion models, a type of generative model, for DAS vertical...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhu, Donglin, Fu, Lei, Kazei, Vladimir, Li, Weichang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10611154/
https://www.ncbi.nlm.nih.gov/pubmed/37896712
http://dx.doi.org/10.3390/s23208619
_version_ 1785128424906424320
author Zhu, Donglin
Fu, Lei
Kazei, Vladimir
Li, Weichang
author_facet Zhu, Donglin
Fu, Lei
Kazei, Vladimir
Li, Weichang
author_sort Zhu, Donglin
collection PubMed
description Distributed acoustic sensing (DAS) has emerged as a transformational technology for seismic data acquisition. However, noise remains a major impediment, necessitating advanced denoising techniques. This study pioneers the application of diffusion models, a type of generative model, for DAS vertical seismic profile (VSP) data denoising. The diffusion network is trained on a new generated synthetic dataset that accommodates variations in the acquisition parameters. The trained model is applied to suppress noise in synthetic and field DAS-VSP data. The results demonstrate the model’s effectiveness in removing various noise types with minimal signal leakage, outperforming conventional methods. This research signifies diffusion models’ potential for DAS processing.
format Online
Article
Text
id pubmed-10611154
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-106111542023-10-28 Diffusion Model for DAS-VSP Data Denoising Zhu, Donglin Fu, Lei Kazei, Vladimir Li, Weichang Sensors (Basel) Article Distributed acoustic sensing (DAS) has emerged as a transformational technology for seismic data acquisition. However, noise remains a major impediment, necessitating advanced denoising techniques. This study pioneers the application of diffusion models, a type of generative model, for DAS vertical seismic profile (VSP) data denoising. The diffusion network is trained on a new generated synthetic dataset that accommodates variations in the acquisition parameters. The trained model is applied to suppress noise in synthetic and field DAS-VSP data. The results demonstrate the model’s effectiveness in removing various noise types with minimal signal leakage, outperforming conventional methods. This research signifies diffusion models’ potential for DAS processing. MDPI 2023-10-21 /pmc/articles/PMC10611154/ /pubmed/37896712 http://dx.doi.org/10.3390/s23208619 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhu, Donglin
Fu, Lei
Kazei, Vladimir
Li, Weichang
Diffusion Model for DAS-VSP Data Denoising
title Diffusion Model for DAS-VSP Data Denoising
title_full Diffusion Model for DAS-VSP Data Denoising
title_fullStr Diffusion Model for DAS-VSP Data Denoising
title_full_unstemmed Diffusion Model for DAS-VSP Data Denoising
title_short Diffusion Model for DAS-VSP Data Denoising
title_sort diffusion model for das-vsp data denoising
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10611154/
https://www.ncbi.nlm.nih.gov/pubmed/37896712
http://dx.doi.org/10.3390/s23208619
work_keys_str_mv AT zhudonglin diffusionmodelfordasvspdatadenoising
AT fulei diffusionmodelfordasvspdatadenoising
AT kazeivladimir diffusionmodelfordasvspdatadenoising
AT liweichang diffusionmodelfordasvspdatadenoising