Estimation of shipping noise from sparse measurements via generative adversarial networks

There is growing interest in prediction of anthropogenic noise levels in the ocean. Evidence suggests that sources of ambient noise such as shipping traffic may be destructive to marine organisms that rely on acoustics for communication. Characterizing and predicting ambient noise are also critical...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of the Acoustical Society of America 2019-03, Vol.145 (3), p.1672-1672
Hauptverfasser: Chen, Johnny L., Summers, Jason E.
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 1672
container_issue 3
container_start_page 1672
container_title The Journal of the Acoustical Society of America
container_volume 145
creator Chen, Johnny L.
Summers, Jason E.
description There is growing interest in prediction of anthropogenic noise levels in the ocean. Evidence suggests that sources of ambient noise such as shipping traffic may be destructive to marine organisms that rely on acoustics for communication. Characterizing and predicting ambient noise are also critical to effective naval operations. Understanding ocean noise is constrained by the limited ability to directly measure the spatial distribution of noise levels. In this work, we present a deep-learning method to estimate the spatial distribution of ambient noise due to shipping from a small number of measurements. Noise levels are typically estimated using forward models based on statistical information about shipping routes and source levels and predictions or measurements of environmental variables including bathymetry and sound-speed profile. Inverse methods are sometimes used to estimate input parameters from in situ data. In contrast, we demonstrate the robust estimation of shipping noise using a pretrained generative adversarial network (GAN) as a prior. By using a context and prior loss, our algorithm is able to accurately predict the entire spatial distribution of noise from sparsely sampled measurements. This can yield greater accuracy than a forward model based on environmental parameters taken from archival databases or inferred in situ.
doi_str_mv 10.1121/1.5101134
format Article
fullrecord <record><control><sourceid>scitation_cross</sourceid><recordid>TN_cdi_scitation_primary_10_1121_1_5101134</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>jasa</sourcerecordid><originalsourceid>FETCH-LOGICAL-c694-701814c0df2d952a9801844a064f8cb249ed65651dc1ba25d6046d162fef5fae3</originalsourceid><addsrcrecordid>eNp9kEtPwzAQhC0EEqFw4B_4ClKK17FNckRVKUiVuPTCKXLjdTE0D3lDEP8eQ3vmtA99M9IMY9cg5gAS7mCuQQAU6oRloKXISy3VKcuEEJCryphzdkH0nk5dFlXGXpc0htaOoe947zm9hWEI3Y53fSDkPvYtp8HGtLdo6TNii91IfAqW77DDmJQTcusmjGRjsHve4fjVxw-6ZGfe7gmvjnPGNo_LzeIpX7-snhcP67wxlcrvBZSgGuG8dJWWtirTQykrjPJls5WqQme00eAa2FqpnRHKODDSo9feYjFjNwfbJvZEEX09xBQoftcg6t9KaqiPlST29sBSE8a_zP_AP5RYYg8</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Estimation of shipping noise from sparse measurements via generative adversarial networks</title><source>AIP Journals Complete</source><source>Alma/SFX Local Collection</source><source>AIP Acoustical Society of America</source><creator>Chen, Johnny L. ; Summers, Jason E.</creator><creatorcontrib>Chen, Johnny L. ; Summers, Jason E.</creatorcontrib><description>There is growing interest in prediction of anthropogenic noise levels in the ocean. Evidence suggests that sources of ambient noise such as shipping traffic may be destructive to marine organisms that rely on acoustics for communication. Characterizing and predicting ambient noise are also critical to effective naval operations. Understanding ocean noise is constrained by the limited ability to directly measure the spatial distribution of noise levels. In this work, we present a deep-learning method to estimate the spatial distribution of ambient noise due to shipping from a small number of measurements. Noise levels are typically estimated using forward models based on statistical information about shipping routes and source levels and predictions or measurements of environmental variables including bathymetry and sound-speed profile. Inverse methods are sometimes used to estimate input parameters from in situ data. In contrast, we demonstrate the robust estimation of shipping noise using a pretrained generative adversarial network (GAN) as a prior. By using a context and prior loss, our algorithm is able to accurately predict the entire spatial distribution of noise from sparsely sampled measurements. This can yield greater accuracy than a forward model based on environmental parameters taken from archival databases or inferred in situ.</description><identifier>ISSN: 0001-4966</identifier><identifier>EISSN: 1520-8524</identifier><identifier>DOI: 10.1121/1.5101134</identifier><identifier>CODEN: JASMAN</identifier><language>eng</language><ispartof>The Journal of the Acoustical Society of America, 2019-03, Vol.145 (3), p.1672-1672</ispartof><rights>Acoustical Society of America</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://pubs.aip.org/jasa/article-lookup/doi/10.1121/1.5101134$$EHTML$$P50$$Gscitation$$H</linktohtml><link.rule.ids>207,208,314,778,782,792,1562,4500,27907,27908,76135</link.rule.ids></links><search><creatorcontrib>Chen, Johnny L.</creatorcontrib><creatorcontrib>Summers, Jason E.</creatorcontrib><title>Estimation of shipping noise from sparse measurements via generative adversarial networks</title><title>The Journal of the Acoustical Society of America</title><description>There is growing interest in prediction of anthropogenic noise levels in the ocean. Evidence suggests that sources of ambient noise such as shipping traffic may be destructive to marine organisms that rely on acoustics for communication. Characterizing and predicting ambient noise are also critical to effective naval operations. Understanding ocean noise is constrained by the limited ability to directly measure the spatial distribution of noise levels. In this work, we present a deep-learning method to estimate the spatial distribution of ambient noise due to shipping from a small number of measurements. Noise levels are typically estimated using forward models based on statistical information about shipping routes and source levels and predictions or measurements of environmental variables including bathymetry and sound-speed profile. Inverse methods are sometimes used to estimate input parameters from in situ data. In contrast, we demonstrate the robust estimation of shipping noise using a pretrained generative adversarial network (GAN) as a prior. By using a context and prior loss, our algorithm is able to accurately predict the entire spatial distribution of noise from sparsely sampled measurements. This can yield greater accuracy than a forward model based on environmental parameters taken from archival databases or inferred in situ.</description><issn>0001-4966</issn><issn>1520-8524</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNp9kEtPwzAQhC0EEqFw4B_4ClKK17FNckRVKUiVuPTCKXLjdTE0D3lDEP8eQ3vmtA99M9IMY9cg5gAS7mCuQQAU6oRloKXISy3VKcuEEJCryphzdkH0nk5dFlXGXpc0htaOoe947zm9hWEI3Y53fSDkPvYtp8HGtLdo6TNii91IfAqW77DDmJQTcusmjGRjsHve4fjVxw-6ZGfe7gmvjnPGNo_LzeIpX7-snhcP67wxlcrvBZSgGuG8dJWWtirTQykrjPJls5WqQme00eAa2FqpnRHKODDSo9feYjFjNwfbJvZEEX09xBQoftcg6t9KaqiPlST29sBSE8a_zP_AP5RYYg8</recordid><startdate>201903</startdate><enddate>201903</enddate><creator>Chen, Johnny L.</creator><creator>Summers, Jason E.</creator><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>201903</creationdate><title>Estimation of shipping noise from sparse measurements via generative adversarial networks</title><author>Chen, Johnny L. ; Summers, Jason E.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c694-701814c0df2d952a9801844a064f8cb249ed65651dc1ba25d6046d162fef5fae3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chen, Johnny L.</creatorcontrib><creatorcontrib>Summers, Jason E.</creatorcontrib><collection>CrossRef</collection><jtitle>The Journal of the Acoustical Society of America</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chen, Johnny L.</au><au>Summers, Jason E.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Estimation of shipping noise from sparse measurements via generative adversarial networks</atitle><jtitle>The Journal of the Acoustical Society of America</jtitle><date>2019-03</date><risdate>2019</risdate><volume>145</volume><issue>3</issue><spage>1672</spage><epage>1672</epage><pages>1672-1672</pages><issn>0001-4966</issn><eissn>1520-8524</eissn><coden>JASMAN</coden><abstract>There is growing interest in prediction of anthropogenic noise levels in the ocean. Evidence suggests that sources of ambient noise such as shipping traffic may be destructive to marine organisms that rely on acoustics for communication. Characterizing and predicting ambient noise are also critical to effective naval operations. Understanding ocean noise is constrained by the limited ability to directly measure the spatial distribution of noise levels. In this work, we present a deep-learning method to estimate the spatial distribution of ambient noise due to shipping from a small number of measurements. Noise levels are typically estimated using forward models based on statistical information about shipping routes and source levels and predictions or measurements of environmental variables including bathymetry and sound-speed profile. Inverse methods are sometimes used to estimate input parameters from in situ data. In contrast, we demonstrate the robust estimation of shipping noise using a pretrained generative adversarial network (GAN) as a prior. By using a context and prior loss, our algorithm is able to accurately predict the entire spatial distribution of noise from sparsely sampled measurements. This can yield greater accuracy than a forward model based on environmental parameters taken from archival databases or inferred in situ.</abstract><doi>10.1121/1.5101134</doi><tpages>1</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0001-4966
ispartof The Journal of the Acoustical Society of America, 2019-03, Vol.145 (3), p.1672-1672
issn 0001-4966
1520-8524
language eng
recordid cdi_scitation_primary_10_1121_1_5101134
source AIP Journals Complete; Alma/SFX Local Collection; AIP Acoustical Society of America
title Estimation of shipping noise from sparse measurements via generative adversarial networks
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-16T09%3A11%3A39IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-scitation_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Estimation%20of%20shipping%20noise%20from%20sparse%20measurements%20via%20generative%20adversarial%20networks&rft.jtitle=The%20Journal%20of%20the%20Acoustical%20Society%20of%20America&rft.au=Chen,%20Johnny%20L.&rft.date=2019-03&rft.volume=145&rft.issue=3&rft.spage=1672&rft.epage=1672&rft.pages=1672-1672&rft.issn=0001-4966&rft.eissn=1520-8524&rft.coden=JASMAN&rft_id=info:doi/10.1121/1.5101134&rft_dat=%3Cscitation_cross%3Ejasa%3C/scitation_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true