NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations
Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial featur...
Gespeichert in:
| Veröffentlicht in: | IEEE geoscience and remote sensing letters 2022, Vol.19, p.1-5 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 5 |
|---|---|
| container_issue | |
| container_start_page | 1 |
| container_title | IEEE geoscience and remote sensing letters |
| container_volume | 19 |
| creator | Wang, Yooseung Seo, Junghoon Jeon, Taegyun |
| description | Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial feature point to refer to all other contextual information and results in more accurate road segmentation. In detail, our single model without any postprocessing like conditional random field (CRF) refinement performed better than any other published state-of-the-art ensemble model in the official DeepGlobe Challenge. Moreover, our nonlocal LinkNet (NL-LinkNet) beat the D-LinkNet, the winner of the DeepGlobe challenge (Demir et al. , 2018), with 43% less parameters, less giga floating-point operations per seconds (GFLOPs), and shorter training convergence time. We also present empirical analyses on the proper usages of NLBs for the baseline model. |
| doi_str_mv | 10.1109/LGRS.2021.3050477 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_9336223</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9336223</ieee_id><sourcerecordid>2610182527</sourcerecordid><originalsourceid>FETCH-LOGICAL-c341t-dbdc694bc3d1892fa0982651db22ca8e0d86667c722b90cd8f3f7f07c9e497523</originalsourceid><addsrcrecordid>eNo9kF1LwzAUhoMoOKc_QLwJeN2Zj6ZJvJsyp1A3mBt6F9IkdZ2zmWmK-u9t2fDqvBye9xx4ALjEaIQxkjf5dPEyIojgEUUMpZwfgQFmTCSIcXzc55QlTIq3U3DWNBuESCoEH4DVLE_yqv6YuXgLl_5bBwvz6n0dXYB3bYTPPjg4NqYNOjq48NrCyU8M2sTK1_C1ims48_XWG72F853rqG7fnIOTUm8bd3GYQ7B6mCzvH5N8Pn26H-eJoSmOiS2syWRaGGqxkKTUSAqSMWwLQowWDlmRZRk3nJBCImNFSUteIm6kSyVnhA7B9f7uLviv1jVRbXwb6u6lIhlGWBBGeEfhPWWCb5rgSrUL1acOvwoj1dtTvT3V21MHe13nat-pnHP_vKQ0I4TSP822aow</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2610182527</pqid></control><display><type>article</type><title>NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations</title><source>IEEE Electronic Library (IEL)</source><creator>Wang, Yooseung ; Seo, Junghoon ; Jeon, Taegyun</creator><creatorcontrib>Wang, Yooseung ; Seo, Junghoon ; Jeon, Taegyun</creatorcontrib><description>Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial feature point to refer to all other contextual information and results in more accurate road segmentation. In detail, our single model without any postprocessing like conditional random field (CRF) refinement performed better than any other published state-of-the-art ensemble model in the official DeepGlobe Challenge. Moreover, our nonlocal LinkNet (NL-LinkNet) beat the D-LinkNet, the winner of the DeepGlobe challenge (Demir et al. , 2018), with 43% less parameters, less giga floating-point operations per seconds (GFLOPs), and shorter training convergence time. We also present empirical analyses on the proper usages of NLBs for the baseline model.</description><identifier>ISSN: 1545-598X</identifier><identifier>EISSN: 1558-0571</identifier><identifier>DOI: 10.1109/LGRS.2021.3050477</identifier><identifier>CODEN: IGRSBY</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Computer architecture ; Conditional random fields ; Convolutional neural networks (CNNs) ; Data mining ; Decoding ; Empirical analysis ; Feature extraction ; Floating point arithmetic ; Image resolution ; Image segmentation ; nonlocal LinkNet (NL-LinkNet) ; Remote sensing ; road extraction ; Roads ; Satellite imagery ; Satellites ; Spaceborne remote sensing ; Training</subject><ispartof>IEEE geoscience and remote sensing letters, 2022, Vol.19, p.1-5</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c341t-dbdc694bc3d1892fa0982651db22ca8e0d86667c722b90cd8f3f7f07c9e497523</citedby><cites>FETCH-LOGICAL-c341t-dbdc694bc3d1892fa0982651db22ca8e0d86667c722b90cd8f3f7f07c9e497523</cites><orcidid>0000-0001-8320-9050 ; 0000-0002-2341-0251</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9336223$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,777,781,793,4010,27904,27905,27906,54739</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9336223$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Wang, Yooseung</creatorcontrib><creatorcontrib>Seo, Junghoon</creatorcontrib><creatorcontrib>Jeon, Taegyun</creatorcontrib><title>NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations</title><title>IEEE geoscience and remote sensing letters</title><addtitle>LGRS</addtitle><description>Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial feature point to refer to all other contextual information and results in more accurate road segmentation. In detail, our single model without any postprocessing like conditional random field (CRF) refinement performed better than any other published state-of-the-art ensemble model in the official DeepGlobe Challenge. Moreover, our nonlocal LinkNet (NL-LinkNet) beat the D-LinkNet, the winner of the DeepGlobe challenge (Demir et al. , 2018), with 43% less parameters, less giga floating-point operations per seconds (GFLOPs), and shorter training convergence time. We also present empirical analyses on the proper usages of NLBs for the baseline model.</description><subject>Computer architecture</subject><subject>Conditional random fields</subject><subject>Convolutional neural networks (CNNs)</subject><subject>Data mining</subject><subject>Decoding</subject><subject>Empirical analysis</subject><subject>Feature extraction</subject><subject>Floating point arithmetic</subject><subject>Image resolution</subject><subject>Image segmentation</subject><subject>nonlocal LinkNet (NL-LinkNet)</subject><subject>Remote sensing</subject><subject>road extraction</subject><subject>Roads</subject><subject>Satellite imagery</subject><subject>Satellites</subject><subject>Spaceborne remote sensing</subject><subject>Training</subject><issn>1545-598X</issn><issn>1558-0571</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kF1LwzAUhoMoOKc_QLwJeN2Zj6ZJvJsyp1A3mBt6F9IkdZ2zmWmK-u9t2fDqvBye9xx4ALjEaIQxkjf5dPEyIojgEUUMpZwfgQFmTCSIcXzc55QlTIq3U3DWNBuESCoEH4DVLE_yqv6YuXgLl_5bBwvz6n0dXYB3bYTPPjg4NqYNOjq48NrCyU8M2sTK1_C1ims48_XWG72F853rqG7fnIOTUm8bd3GYQ7B6mCzvH5N8Pn26H-eJoSmOiS2syWRaGGqxkKTUSAqSMWwLQowWDlmRZRk3nJBCImNFSUteIm6kSyVnhA7B9f7uLviv1jVRbXwb6u6lIhlGWBBGeEfhPWWCb5rgSrUL1acOvwoj1dtTvT3V21MHe13nat-pnHP_vKQ0I4TSP822aow</recordid><startdate>2022</startdate><enddate>2022</enddate><creator>Wang, Yooseung</creator><creator>Seo, Junghoon</creator><creator>Jeon, Taegyun</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7TG</scope><scope>7UA</scope><scope>8FD</scope><scope>C1K</scope><scope>F1W</scope><scope>FR3</scope><scope>H8D</scope><scope>H96</scope><scope>JQ2</scope><scope>KL.</scope><scope>KR7</scope><scope>L.G</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0001-8320-9050</orcidid><orcidid>https://orcid.org/0000-0002-2341-0251</orcidid></search><sort><creationdate>2022</creationdate><title>NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations</title><author>Wang, Yooseung ; Seo, Junghoon ; Jeon, Taegyun</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c341t-dbdc694bc3d1892fa0982651db22ca8e0d86667c722b90cd8f3f7f07c9e497523</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer architecture</topic><topic>Conditional random fields</topic><topic>Convolutional neural networks (CNNs)</topic><topic>Data mining</topic><topic>Decoding</topic><topic>Empirical analysis</topic><topic>Feature extraction</topic><topic>Floating point arithmetic</topic><topic>Image resolution</topic><topic>Image segmentation</topic><topic>nonlocal LinkNet (NL-LinkNet)</topic><topic>Remote sensing</topic><topic>road extraction</topic><topic>Roads</topic><topic>Satellite imagery</topic><topic>Satellites</topic><topic>Spaceborne remote sensing</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Wang, Yooseung</creatorcontrib><creatorcontrib>Seo, Junghoon</creatorcontrib><creatorcontrib>Jeon, Taegyun</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Meteorological & Geoastrophysical Abstracts</collection><collection>Water Resources Abstracts</collection><collection>Technology Research Database</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Engineering Research Database</collection><collection>Aerospace Database</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources</collection><collection>ProQuest Computer Science Collection</collection><collection>Meteorological & Geoastrophysical Abstracts - Academic</collection><collection>Civil Engineering Abstracts</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) Professional</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>IEEE geoscience and remote sensing letters</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Wang, Yooseung</au><au>Seo, Junghoon</au><au>Jeon, Taegyun</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations</atitle><jtitle>IEEE geoscience and remote sensing letters</jtitle><stitle>LGRS</stitle><date>2022</date><risdate>2022</risdate><volume>19</volume><spage>1</spage><epage>5</epage><pages>1-5</pages><issn>1545-598X</issn><eissn>1558-0571</eissn><coden>IGRSBY</coden><abstract>Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial feature point to refer to all other contextual information and results in more accurate road segmentation. In detail, our single model without any postprocessing like conditional random field (CRF) refinement performed better than any other published state-of-the-art ensemble model in the official DeepGlobe Challenge. Moreover, our nonlocal LinkNet (NL-LinkNet) beat the D-LinkNet, the winner of the DeepGlobe challenge (Demir et al. , 2018), with 43% less parameters, less giga floating-point operations per seconds (GFLOPs), and shorter training convergence time. We also present empirical analyses on the proper usages of NLBs for the baseline model.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/LGRS.2021.3050477</doi><tpages>5</tpages><orcidid>https://orcid.org/0000-0001-8320-9050</orcidid><orcidid>https://orcid.org/0000-0002-2341-0251</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1545-598X |
| ispartof | IEEE geoscience and remote sensing letters, 2022, Vol.19, p.1-5 |
| issn | 1545-598X 1558-0571 |
| language | eng |
| recordid | cdi_ieee_primary_9336223 |
| source | IEEE Electronic Library (IEL) |
| subjects | Computer architecture Conditional random fields Convolutional neural networks (CNNs) Data mining Decoding Empirical analysis Feature extraction Floating point arithmetic Image resolution Image segmentation nonlocal LinkNet (NL-LinkNet) Remote sensing road extraction Roads Satellite imagery Satellites Spaceborne remote sensing Training |
| title | NL-LinkNet: Toward Lighter But More Accurate Road Extraction With Nonlocal Operations |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T13%3A28%3A25IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=NL-LinkNet:%20Toward%20Lighter%20But%20More%20Accurate%20Road%20Extraction%20With%20Nonlocal%20Operations&rft.jtitle=IEEE%20geoscience%20and%20remote%20sensing%20letters&rft.au=Wang,%20Yooseung&rft.date=2022&rft.volume=19&rft.spage=1&rft.epage=5&rft.pages=1-5&rft.issn=1545-598X&rft.eissn=1558-0571&rft.coden=IGRSBY&rft_id=info:doi/10.1109/LGRS.2021.3050477&rft_dat=%3Cproquest_RIE%3E2610182527%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2610182527&rft_id=info:pmid/&rft_ieee_id=9336223&rfr_iscdi=true |