Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss

The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net ar...

Full description

Saved in:
Bibliographic Details
Main Authors: Rakhlin, Alexander, Davydow, Alex, Nikolenko, Sergey
Format: Conference Proceeding
Language:English
Subjects:
Computer architecture
Computer vision
Image segmentation
Satellites
Stochastic processes
Task analysis
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 2574
container_issue
container_start_page 257
container_title
container_volume
creator Rakhlin, Alexander
Davydow, Alex
Nikolenko, Sergey
description The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net architecture with the Lov´asz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.
doi_str_mv 10.1109/CVPRW.2018.00048
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_8575508</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8575508</ieee_id><sourcerecordid>8575508</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-9de019fc667ca9715f1e06ba3a7138a8cecab48f9f0ff3829b934a3d5993a9d3</originalsourceid><addsrcrecordid>eNotjMtKxDAYRqMgOIyzF9zkBVrzN81tKcXLQFFxqi6Hv22ikXYqTRhnfBufxRezoqvDd-A7hJwCSwGYOS-e7h-e04yBThljuT4gC6M0CK6lhMkcklkGkiVKgDwmixDeJglMC2H4jFQlblpaDFs70qLDELzzDUY_bKgbh56uMNqu89HSZY8vdtzTDx9f6WNyayP9vZbD9vsrfCarwcUed9MO4YQcOeyCXfxzTqqry6q4Scq762VxUSYelIiJaS0D4xopVYNGgXBgmayRowKuUTe2wTrXzjjmHNeZqQ3PkbfCGI6m5XNy9pf11tr1--h7HPdrLZQQTPMfTexRuA</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss</title><source>IEEE Xplore All Conference Series</source><creator>Rakhlin, Alexander ; Davydow, Alex ; Nikolenko, Sergey</creator><creatorcontrib>Rakhlin, Alexander ; Davydow, Alex ; Nikolenko, Sergey</creatorcontrib><description>The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net architecture with the Lov´asz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.</description><identifier>EISSN: 2160-7516</identifier><identifier>EISBN: 9781538661000</identifier><identifier>EISBN: 1538661004</identifier><identifier>DOI: 10.1109/CVPRW.2018.00048</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Computer architecture ; Computer vision ; Image segmentation ; Satellites ; Stochastic processes ; Task analysis ; Training</subject><ispartof>2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2018, p.257-2574</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8575508$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,27925,54555,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8575508$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Rakhlin, Alexander</creatorcontrib><creatorcontrib>Davydow, Alex</creatorcontrib><creatorcontrib>Nikolenko, Sergey</creatorcontrib><title>Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss</title><title>2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)</title><addtitle>CVPRW</addtitle><description>The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net architecture with the Lov´asz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.</description><subject>Computer architecture</subject><subject>Computer vision</subject><subject>Image segmentation</subject><subject>Satellites</subject><subject>Stochastic processes</subject><subject>Task analysis</subject><subject>Training</subject><issn>2160-7516</issn><isbn>9781538661000</isbn><isbn>1538661004</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2018</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotjMtKxDAYRqMgOIyzF9zkBVrzN81tKcXLQFFxqi6Hv22ikXYqTRhnfBufxRezoqvDd-A7hJwCSwGYOS-e7h-e04yBThljuT4gC6M0CK6lhMkcklkGkiVKgDwmixDeJglMC2H4jFQlblpaDFs70qLDELzzDUY_bKgbh56uMNqu89HSZY8vdtzTDx9f6WNyayP9vZbD9vsrfCarwcUed9MO4YQcOeyCXfxzTqqry6q4Scq762VxUSYelIiJaS0D4xopVYNGgXBgmayRowKuUTe2wTrXzjjmHNeZqQ3PkbfCGI6m5XNy9pf11tr1--h7HPdrLZQQTPMfTexRuA</recordid><startdate>201806</startdate><enddate>201806</enddate><creator>Rakhlin, Alexander</creator><creator>Davydow, Alex</creator><creator>Nikolenko, Sergey</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201806</creationdate><title>Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss</title><author>Rakhlin, Alexander ; Davydow, Alex ; Nikolenko, Sergey</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-9de019fc667ca9715f1e06ba3a7138a8cecab48f9f0ff3829b934a3d5993a9d3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer architecture</topic><topic>Computer vision</topic><topic>Image segmentation</topic><topic>Satellites</topic><topic>Stochastic processes</topic><topic>Task analysis</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Rakhlin, Alexander</creatorcontrib><creatorcontrib>Davydow, Alex</creatorcontrib><creatorcontrib>Nikolenko, Sergey</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore (IEEE/IET Electronic Library - IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Rakhlin, Alexander</au><au>Davydow, Alex</au><au>Nikolenko, Sergey</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss</atitle><btitle>2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)</btitle><stitle>CVPRW</stitle><date>2018-06</date><risdate>2018</risdate><spage>257</spage><epage>2574</epage><pages>257-2574</pages><eissn>2160-7516</eissn><eisbn>9781538661000</eisbn><eisbn>1538661004</eisbn><coden>IEEPAD</coden><abstract>The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net architecture with the Lov´asz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.</abstract><pub>IEEE</pub><doi>10.1109/CVPRW.2018.00048</doi><tpages>2318</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2160-7516
ispartof 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2018, p.257-2574
issn 2160-7516
language eng
recordid cdi_ieee_primary_8575508
source IEEE Xplore All Conference Series
subjects Computer architecture
Computer vision
Image segmentation
Satellites
Stochastic processes
Task analysis
Training
title Land Cover Classification from Satellite Imagery with U-Net and Lovász-Softmax Loss
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T18%3A59%3A51IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Land%20Cover%20Classification%20from%20Satellite%20Imagery%20with%20U-Net%20and%20Lov%C3%A1sz-Softmax%20Loss&rft.btitle=2018%20IEEE/CVF%20Conference%20on%20Computer%20Vision%20and%20Pattern%20Recognition%20Workshops%20(CVPRW)&rft.au=Rakhlin,%20Alexander&rft.date=2018-06&rft.spage=257&rft.epage=2574&rft.pages=257-2574&rft.eissn=2160-7516&rft.coden=IEEPAD&rft_id=info:doi/10.1109/CVPRW.2018.00048&rft.eisbn=9781538661000&rft.eisbn_list=1538661004&rft_dat=%3Cieee_CHZPO%3E8575508%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i175t-9de019fc667ca9715f1e06ba3a7138a8cecab48f9f0ff3829b934a3d5993a9d3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=8575508&rfr_iscdi=true