Jing Sun, Sigmund Slang, Thomas Elboth, Thomas Larsen Greiner, Steven McDonald, Leiv-J Gelius
{"title":"Attenuation of marine seismic interference noise employing a customized U-Net","authors":"Jing Sun, Sigmund Slang, Thomas Elboth, Thomas Larsen Greiner, Steven McDonald, Leiv-J Gelius","doi":"arxiv-2409.01673","DOIUrl":null,"url":null,"abstract":"Marine seismic interference noise occurs when energy from nearby marine\nseismic source vessels is recorded during a seismic survey. Such noise tends to\nbe well preserved over large distances and cause coherent artifacts in the\nrecorded data. Over the years, the industry has developed various denoising\ntechniques for seismic interference removal, but although well performing they\nare still time-consuming in use. Machine-learning based processing represents\nan alternative approach, which may significantly improve the computational\nefficiency. In case of conventional images, autoencoders are frequently\nemployed for denoising purposes. However, due to the special characteristics of\nseismic data as well as the noise, autoencoders failed in the case of marine\nseismic interference noise. We therefore propose the use of a customized U-Net\ndesign with element-wise summation as part of the skip-connection blocks to\nhandle the vanishing gradient problem and to ensure information fusion between\nhigh- and low-level features. To secure a realistic study, only seismic field\ndata were employed, including 25000 training examples. The customized U-Net was\nfound to perform well leaving only minor residuals, except for the case when\nseismic interference noise comes from the side. We further demonstrate that\nsuch noise can be treated by slightly increasing the depth of our network.\nAlthough our customized U-Net does not outperform a standard commercial\nalgorithm in quality, it can (after proper training) read and process one\nsingle shot gather in approximately 0.02s. This is significantly faster than\nany existing industry denoising algorithm. In addition, the proposed network\nprocesses shot gathers in a sequential order, which is an advantage compared\nwith industry algorithms that typically require a multi-shot input to break the\ncoherency of the noise.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"11 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Geophysics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.01673","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Marine seismic interference noise occurs when energy from nearby marine
seismic source vessels is recorded during a seismic survey. Such noise tends to
be well preserved over large distances and cause coherent artifacts in the
recorded data. Over the years, the industry has developed various denoising
techniques for seismic interference removal, but although well performing they
are still time-consuming in use. Machine-learning based processing represents
an alternative approach, which may significantly improve the computational
efficiency. In case of conventional images, autoencoders are frequently
employed for denoising purposes. However, due to the special characteristics of
seismic data as well as the noise, autoencoders failed in the case of marine
seismic interference noise. We therefore propose the use of a customized U-Net
design with element-wise summation as part of the skip-connection blocks to
handle the vanishing gradient problem and to ensure information fusion between
high- and low-level features. To secure a realistic study, only seismic field
data were employed, including 25000 training examples. The customized U-Net was
found to perform well leaving only minor residuals, except for the case when
seismic interference noise comes from the side. We further demonstrate that
such noise can be treated by slightly increasing the depth of our network.
Although our customized U-Net does not outperform a standard commercial
algorithm in quality, it can (after proper training) read and process one
single shot gather in approximately 0.02s. This is significantly faster than
any existing industry denoising algorithm. In addition, the proposed network
processes shot gathers in a sequential order, which is an advantage compared
with industry algorithms that typically require a multi-shot input to break the
coherency of the noise.