A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution
Super-resolution (SR) brings an excellent opportunity to improve a wide range of different remote sensing applications. SR techniques are concerned about increasing the image resolution while providing finer spatial details than those captured by the original acquisition instrument. Therefore, SR te...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on geoscience and remote sensing 2018-11, Vol.56 (11), p.6792-6810 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 6810 |
|---|---|
| container_issue | 11 |
| container_start_page | 6792 |
| container_title | IEEE transactions on geoscience and remote sensing |
| container_volume | 56 |
| creator | Haut, Juan Mario Fernandez-Beltran, Ruben Paoletti, Mercedes E. Plaza, Javier Plaza, Antonio Pla, Filiberto |
| description | Super-resolution (SR) brings an excellent opportunity to improve a wide range of different remote sensing applications. SR techniques are concerned about increasing the image resolution while providing finer spatial details than those captured by the original acquisition instrument. Therefore, SR techniques are particularly useful to cope with the increasing demand remote sensing imaging applications requiring fine spatial resolution. Even though different machine learning paradigms have been successfully applied in SR, more research is required to improve the SR process without the need of external high-resolution (HR) training examples. This paper proposes a new convolutional generator model to super-resolve low-resolution (LR) remote sensing data from an unsupervised perspective. That is, the proposed generative network is able to initially learn relationships between the LR and HR domains throughout several convolutional, downsampling, batch normalization, and activation layers. Then, the data are symmetrically projected to the target resolution while guaranteeing a reconstruction constraint over the LR input image. An experimental comparison is conducted using 12 different unsupervised SR methods over different test images. Our experiments reveal the potential of the proposed approach to improve the resolution of remote sensing imagery. |
| doi_str_mv | 10.1109/TGRS.2018.2843525 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_proquest_journals_2126663520</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8400496</ieee_id><sourcerecordid>2126663520</sourcerecordid><originalsourceid>FETCH-LOGICAL-c402t-de6c03f59f3ad2eedeb46e4d1dad4f9de10d138f5293f90cadf542b5780794523</originalsourceid><addsrcrecordid>eNo9kE1LAzEQhoMoWKs_QLwEPG-dZJN0cyxVa0EU-nHyELabSdnabtZkt8V_75YWLzPw8rwz8BByz2DAGOinxWQ2H3Bg2YBnIpVcXpAekzJLQAlxSXrAtEp4pvk1uYlxA8CEZMMe-RrRDzzQZ8SaTrDCkDflHrusOfjwTZ0PdFnFtsawLyNaOsOdb5DOsYpltabzbmwxme7ydRcesWSG0W_bpvTVLbly-Tbi3Xn3yfL1ZTF-S94_J9Px6D0pBPAmsagKSJ3ULs0tR7S4EgqFZTa3wmmLDCxLMye5Tp2GIrdOCr6SwwyGWkie9snj6W4d_E-LsTEb34aqe2k440qpzgd0FDtRRfAxBnSmDuUuD7-GgTk6NEeH5ujQnB12nYdTp0TEfz4TAEKr9A_aPG4Q</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2126663520</pqid></control><display><type>article</type><title>A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution</title><source>IEEE Electronic Library (IEL)</source><creator>Haut, Juan Mario ; Fernandez-Beltran, Ruben ; Paoletti, Mercedes E. ; Plaza, Javier ; Plaza, Antonio ; Pla, Filiberto</creator><creatorcontrib>Haut, Juan Mario ; Fernandez-Beltran, Ruben ; Paoletti, Mercedes E. ; Plaza, Javier ; Plaza, Antonio ; Pla, Filiberto</creatorcontrib><description>Super-resolution (SR) brings an excellent opportunity to improve a wide range of different remote sensing applications. SR techniques are concerned about increasing the image resolution while providing finer spatial details than those captured by the original acquisition instrument. Therefore, SR techniques are particularly useful to cope with the increasing demand remote sensing imaging applications requiring fine spatial resolution. Even though different machine learning paradigms have been successfully applied in SR, more research is required to improve the SR process without the need of external high-resolution (HR) training examples. This paper proposes a new convolutional generator model to super-resolve low-resolution (LR) remote sensing data from an unsupervised perspective. That is, the proposed generative network is able to initially learn relationships between the LR and HR domains throughout several convolutional, downsampling, batch normalization, and activation layers. Then, the data are symmetrically projected to the target resolution while guaranteeing a reconstruction constraint over the LR input image. An experimental comparison is conducted using 12 different unsupervised SR methods over different test images. Our experiments reveal the potential of the proposed approach to improve the resolution of remote sensing imagery.</description><identifier>ISSN: 0196-2892</identifier><identifier>EISSN: 1558-0644</identifier><identifier>DOI: 10.1109/TGRS.2018.2843525</identifier><identifier>CODEN: IGRSD2</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Convolutional neural networks (CNNs) ; Data models ; Detection ; Domains ; Image processing ; Image reconstruction ; Image resolution ; Imagery ; Imaging ; Imaging techniques ; Learning algorithms ; Machine learning ; Remote sensing ; Resolution ; Spatial discrimination ; Spatial resolution ; super-resolution (SR) ; Test procedures ; Training</subject><ispartof>IEEE transactions on geoscience and remote sensing, 2018-11, Vol.56 (11), p.6792-6810</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c402t-de6c03f59f3ad2eedeb46e4d1dad4f9de10d138f5293f90cadf542b5780794523</citedby><cites>FETCH-LOGICAL-c402t-de6c03f59f3ad2eedeb46e4d1dad4f9de10d138f5293f90cadf542b5780794523</cites><orcidid>0000-0001-6701-961X ; 0000-0002-9613-1659 ; 0000-0003-1374-8416 ; 0000-0003-0054-3489 ; 0000-0002-2384-9141 ; 0000-0003-1030-3729</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8400496$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,796,27924,27925,54758</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8400496$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Haut, Juan Mario</creatorcontrib><creatorcontrib>Fernandez-Beltran, Ruben</creatorcontrib><creatorcontrib>Paoletti, Mercedes E.</creatorcontrib><creatorcontrib>Plaza, Javier</creatorcontrib><creatorcontrib>Plaza, Antonio</creatorcontrib><creatorcontrib>Pla, Filiberto</creatorcontrib><title>A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution</title><title>IEEE transactions on geoscience and remote sensing</title><addtitle>TGRS</addtitle><description>Super-resolution (SR) brings an excellent opportunity to improve a wide range of different remote sensing applications. SR techniques are concerned about increasing the image resolution while providing finer spatial details than those captured by the original acquisition instrument. Therefore, SR techniques are particularly useful to cope with the increasing demand remote sensing imaging applications requiring fine spatial resolution. Even though different machine learning paradigms have been successfully applied in SR, more research is required to improve the SR process without the need of external high-resolution (HR) training examples. This paper proposes a new convolutional generator model to super-resolve low-resolution (LR) remote sensing data from an unsupervised perspective. That is, the proposed generative network is able to initially learn relationships between the LR and HR domains throughout several convolutional, downsampling, batch normalization, and activation layers. Then, the data are symmetrically projected to the target resolution while guaranteeing a reconstruction constraint over the LR input image. An experimental comparison is conducted using 12 different unsupervised SR methods over different test images. Our experiments reveal the potential of the proposed approach to improve the resolution of remote sensing imagery.</description><subject>Convolutional neural networks (CNNs)</subject><subject>Data models</subject><subject>Detection</subject><subject>Domains</subject><subject>Image processing</subject><subject>Image reconstruction</subject><subject>Image resolution</subject><subject>Imagery</subject><subject>Imaging</subject><subject>Imaging techniques</subject><subject>Learning algorithms</subject><subject>Machine learning</subject><subject>Remote sensing</subject><subject>Resolution</subject><subject>Spatial discrimination</subject><subject>Spatial resolution</subject><subject>super-resolution (SR)</subject><subject>Test procedures</subject><subject>Training</subject><issn>0196-2892</issn><issn>1558-0644</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kE1LAzEQhoMoWKs_QLwEPG-dZJN0cyxVa0EU-nHyELabSdnabtZkt8V_75YWLzPw8rwz8BByz2DAGOinxWQ2H3Bg2YBnIpVcXpAekzJLQAlxSXrAtEp4pvk1uYlxA8CEZMMe-RrRDzzQZ8SaTrDCkDflHrusOfjwTZ0PdFnFtsawLyNaOsOdb5DOsYpltabzbmwxme7ydRcesWSG0W_bpvTVLbly-Tbi3Xn3yfL1ZTF-S94_J9Px6D0pBPAmsagKSJ3ULs0tR7S4EgqFZTa3wmmLDCxLMye5Tp2GIrdOCr6SwwyGWkie9snj6W4d_E-LsTEb34aqe2k440qpzgd0FDtRRfAxBnSmDuUuD7-GgTk6NEeH5ujQnB12nYdTp0TEfz4TAEKr9A_aPG4Q</recordid><startdate>20181101</startdate><enddate>20181101</enddate><creator>Haut, Juan Mario</creator><creator>Fernandez-Beltran, Ruben</creator><creator>Paoletti, Mercedes E.</creator><creator>Plaza, Javier</creator><creator>Plaza, Antonio</creator><creator>Pla, Filiberto</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>7UA</scope><scope>8FD</scope><scope>C1K</scope><scope>F1W</scope><scope>FR3</scope><scope>H8D</scope><scope>H96</scope><scope>KR7</scope><scope>L.G</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0001-6701-961X</orcidid><orcidid>https://orcid.org/0000-0002-9613-1659</orcidid><orcidid>https://orcid.org/0000-0003-1374-8416</orcidid><orcidid>https://orcid.org/0000-0003-0054-3489</orcidid><orcidid>https://orcid.org/0000-0002-2384-9141</orcidid><orcidid>https://orcid.org/0000-0003-1030-3729</orcidid></search><sort><creationdate>20181101</creationdate><title>A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution</title><author>Haut, Juan Mario ; Fernandez-Beltran, Ruben ; Paoletti, Mercedes E. ; Plaza, Javier ; Plaza, Antonio ; Pla, Filiberto</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c402t-de6c03f59f3ad2eedeb46e4d1dad4f9de10d138f5293f90cadf542b5780794523</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Convolutional neural networks (CNNs)</topic><topic>Data models</topic><topic>Detection</topic><topic>Domains</topic><topic>Image processing</topic><topic>Image reconstruction</topic><topic>Image resolution</topic><topic>Imagery</topic><topic>Imaging</topic><topic>Imaging techniques</topic><topic>Learning algorithms</topic><topic>Machine learning</topic><topic>Remote sensing</topic><topic>Resolution</topic><topic>Spatial discrimination</topic><topic>Spatial resolution</topic><topic>super-resolution (SR)</topic><topic>Test procedures</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Haut, Juan Mario</creatorcontrib><creatorcontrib>Fernandez-Beltran, Ruben</creatorcontrib><creatorcontrib>Paoletti, Mercedes E.</creatorcontrib><creatorcontrib>Plaza, Javier</creatorcontrib><creatorcontrib>Plaza, Antonio</creatorcontrib><creatorcontrib>Pla, Filiberto</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>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>Civil Engineering Abstracts</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) Professional</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>IEEE transactions on geoscience and remote sensing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Haut, Juan Mario</au><au>Fernandez-Beltran, Ruben</au><au>Paoletti, Mercedes E.</au><au>Plaza, Javier</au><au>Plaza, Antonio</au><au>Pla, Filiberto</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution</atitle><jtitle>IEEE transactions on geoscience and remote sensing</jtitle><stitle>TGRS</stitle><date>2018-11-01</date><risdate>2018</risdate><volume>56</volume><issue>11</issue><spage>6792</spage><epage>6810</epage><pages>6792-6810</pages><issn>0196-2892</issn><eissn>1558-0644</eissn><coden>IGRSD2</coden><abstract>Super-resolution (SR) brings an excellent opportunity to improve a wide range of different remote sensing applications. SR techniques are concerned about increasing the image resolution while providing finer spatial details than those captured by the original acquisition instrument. Therefore, SR techniques are particularly useful to cope with the increasing demand remote sensing imaging applications requiring fine spatial resolution. Even though different machine learning paradigms have been successfully applied in SR, more research is required to improve the SR process without the need of external high-resolution (HR) training examples. This paper proposes a new convolutional generator model to super-resolve low-resolution (LR) remote sensing data from an unsupervised perspective. That is, the proposed generative network is able to initially learn relationships between the LR and HR domains throughout several convolutional, downsampling, batch normalization, and activation layers. Then, the data are symmetrically projected to the target resolution while guaranteeing a reconstruction constraint over the LR input image. An experimental comparison is conducted using 12 different unsupervised SR methods over different test images. Our experiments reveal the potential of the proposed approach to improve the resolution of remote sensing imagery.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TGRS.2018.2843525</doi><tpages>19</tpages><orcidid>https://orcid.org/0000-0001-6701-961X</orcidid><orcidid>https://orcid.org/0000-0002-9613-1659</orcidid><orcidid>https://orcid.org/0000-0003-1374-8416</orcidid><orcidid>https://orcid.org/0000-0003-0054-3489</orcidid><orcidid>https://orcid.org/0000-0002-2384-9141</orcidid><orcidid>https://orcid.org/0000-0003-1030-3729</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 0196-2892 |
| ispartof | IEEE transactions on geoscience and remote sensing, 2018-11, Vol.56 (11), p.6792-6810 |
| issn | 0196-2892 1558-0644 |
| language | eng |
| recordid | cdi_proquest_journals_2126663520 |
| source | IEEE Electronic Library (IEL) |
| subjects | Convolutional neural networks (CNNs) Data models Detection Domains Image processing Image reconstruction Image resolution Imagery Imaging Imaging techniques Learning algorithms Machine learning Remote sensing Resolution Spatial discrimination Spatial resolution super-resolution (SR) Test procedures Training |
| title | A New Deep Generative Network for Unsupervised Remote Sensing Single-Image Super-Resolution |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-05T06%3A39%3A07IST&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=A%20New%20Deep%20Generative%20Network%20for%20Unsupervised%20Remote%20Sensing%20Single-Image%20Super-Resolution&rft.jtitle=IEEE%20transactions%20on%20geoscience%20and%20remote%20sensing&rft.au=Haut,%20Juan%20Mario&rft.date=2018-11-01&rft.volume=56&rft.issue=11&rft.spage=6792&rft.epage=6810&rft.pages=6792-6810&rft.issn=0196-2892&rft.eissn=1558-0644&rft.coden=IGRSD2&rft_id=info:doi/10.1109/TGRS.2018.2843525&rft_dat=%3Cproquest_RIE%3E2126663520%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=2126663520&rft_id=info:pmid/&rft_ieee_id=8400496&rfr_iscdi=true |