Random noise attenuation of sparker seismic oceanography data with machine learning
<p>Seismic oceanography (SO) acquires water column reflections using controlled source seismology and provides high lateral resolution that enables the tracking of the thermohaline structure of the oceans. Most SO studies obtain data using air guns, which can produce acoustic energy below 100&...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2020-11-01
|
| Series: | Ocean Science |
| Online Access: | https://os.copernicus.org/articles/16/1367/2020/os-16-1367-2020.pdf |
| _version_ | 1818136644842160128 |
|---|---|
| author | H. Jun H.-T. Jou C.-H. Kim S. H. Lee H.-J. Kim |
| author_facet | H. Jun H.-T. Jou C.-H. Kim S. H. Lee H.-J. Kim |
| author_sort | H. Jun |
| collection | DOAJ |
| description | <p>Seismic oceanography (SO) acquires water column
reflections using controlled source seismology and provides high lateral
resolution that enables the tracking of the thermohaline structure of the
oceans. Most SO studies obtain data using air guns, which can produce
acoustic energy below 100 Hz bandwidth, with vertical resolution of
approximately 10 m or more. For higher-frequency bands, with vertical
resolution ranging from several centimeters to several meters, a smaller,
low-cost seismic exploration system may be used, such as a sparker source
with central frequencies of 250 Hz or higher. However, the sparker source
has a relatively low energy compared to air guns and consequently produces
data with a lower signal-to-noise (<span class="inline-formula">S∕N</span>) ratio. To attenuate the random noise
and extract reliable signal from the low <span class="inline-formula">S∕N</span> ratio of sparker SO data without
distorting the true shape and amplitude of water column reflections, we
applied machine learning. Specifically, we used a denoising convolutional
neural network (DnCNN) that efficiently suppresses random noise in a natural
image. One of the most important factors of machine learning is the
generation of an appropriate training dataset. We generated two different
training datasets using synthetic and field data. Models trained with the
different training datasets were applied to the test data, and the denoised
results were quantitatively compared. To demonstrate the technique, the
trained models were applied to an SO sparker seismic dataset acquired in the Ulleung Basin, East Sea (Sea of Japan), and the denoised seismic sections were evaluated. The results show
that machine learning can successfully attenuate the random noise in sparker
water column seismic reflection data.</p> |
| first_indexed | 2024-12-11T09:43:41Z |
| format | Article |
| id | doaj.art-c5e721b446744871a93eb116af7877cc |
| institution | Directory Open Access Journal |
| issn | 1812-0784 1812-0792 |
| language | English |
| last_indexed | 2024-12-11T09:43:41Z |
| publishDate | 2020-11-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | Ocean Science |
| spelling | doaj.art-c5e721b446744871a93eb116af7877cc2022-12-22T01:12:37ZengCopernicus PublicationsOcean Science1812-07841812-07922020-11-01161367138310.5194/os-16-1367-2020Random noise attenuation of sparker seismic oceanography data with machine learningH. JunH.-T. JouC.-H. KimS. H. LeeH.-J. Kim<p>Seismic oceanography (SO) acquires water column reflections using controlled source seismology and provides high lateral resolution that enables the tracking of the thermohaline structure of the oceans. Most SO studies obtain data using air guns, which can produce acoustic energy below 100 Hz bandwidth, with vertical resolution of approximately 10 m or more. For higher-frequency bands, with vertical resolution ranging from several centimeters to several meters, a smaller, low-cost seismic exploration system may be used, such as a sparker source with central frequencies of 250 Hz or higher. However, the sparker source has a relatively low energy compared to air guns and consequently produces data with a lower signal-to-noise (<span class="inline-formula">S∕N</span>) ratio. To attenuate the random noise and extract reliable signal from the low <span class="inline-formula">S∕N</span> ratio of sparker SO data without distorting the true shape and amplitude of water column reflections, we applied machine learning. Specifically, we used a denoising convolutional neural network (DnCNN) that efficiently suppresses random noise in a natural image. One of the most important factors of machine learning is the generation of an appropriate training dataset. We generated two different training datasets using synthetic and field data. Models trained with the different training datasets were applied to the test data, and the denoised results were quantitatively compared. To demonstrate the technique, the trained models were applied to an SO sparker seismic dataset acquired in the Ulleung Basin, East Sea (Sea of Japan), and the denoised seismic sections were evaluated. The results show that machine learning can successfully attenuate the random noise in sparker water column seismic reflection data.</p>https://os.copernicus.org/articles/16/1367/2020/os-16-1367-2020.pdf |
| spellingShingle | H. Jun H.-T. Jou C.-H. Kim S. H. Lee H.-J. Kim Random noise attenuation of sparker seismic oceanography data with machine learning Ocean Science |
| title | Random noise attenuation of sparker seismic oceanography data with machine learning |
| title_full | Random noise attenuation of sparker seismic oceanography data with machine learning |
| title_fullStr | Random noise attenuation of sparker seismic oceanography data with machine learning |
| title_full_unstemmed | Random noise attenuation of sparker seismic oceanography data with machine learning |
| title_short | Random noise attenuation of sparker seismic oceanography data with machine learning |
| title_sort | random noise attenuation of sparker seismic oceanography data with machine learning |
| url | https://os.copernicus.org/articles/16/1367/2020/os-16-1367-2020.pdf |
| work_keys_str_mv | AT hjun randomnoiseattenuationofsparkerseismicoceanographydatawithmachinelearning AT htjou randomnoiseattenuationofsparkerseismicoceanographydatawithmachinelearning AT chkim randomnoiseattenuationofsparkerseismicoceanographydatawithmachinelearning AT shlee randomnoiseattenuationofsparkerseismicoceanographydatawithmachinelearning AT hjkim randomnoiseattenuationofsparkerseismicoceanographydatawithmachinelearning |