BEARNet: A novel buildings edge-aware refined network for building extraction from high-resolution remote sensing images
Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart cities. At present, the knowledge-driven building extraction methods mostly rely on building prior knowledge of manual design, resulting in low au...
Saved in:
Published in | IEEE geoscience and remote sensing letters Vol. 20; p. 1 |
---|---|
Main Authors | , , , , |
Format | Journal Article |
Language | English |
Published |
Piscataway
IEEE
01.01.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart cities. At present, the knowledge-driven building extraction methods mostly rely on building prior knowledge of manual design, resulting in low automation and poor versatility. The data-driven methods usually rely too much on training samples, resulting in insufficient pertinence for the features of building extraction and low generalization ability of the model. Therefore, this study proposes a novel buildings edge-aware refined deep learning network (BEARNet) for building extraction from high-resolution remote sensing images. The network takes the building edge as a priori knowledge, learns the building edge features by decoupling the building body and edge, and further optimizes the network by combining the multi-objective loss function to strengthen the pertinence of building edge features extraction. Experimental results show that on the WHU building dataset, which is less difficult to extract buildings, compared with other methods, BEARNet has the highest Precision, F1, IoU and OA values of 97.70%, 97.42%, 95.3% and 98.67%. On the Massachusetts building dataset, where building extraction is difficult, BEARNet has the highest Precision, Recall, F1, IoU and OA compared to other methods, with values of 84.92%, 85.27%, 85.09%, 75.82% and 93.99%, respectively. Our proposed method is more accurate in extracting complex shapes and dense small-scale buildings, and the building edges are more refined and complete. |
---|---|
AbstractList | Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart cities. At present, the knowledge-driven building extraction methods mostly rely on building prior knowledge of manual design, resulting in low automation and poor versatility. The data-driven methods usually rely too much on training samples, resulting in insufficient pertinence for the features of building extraction and low generalization ability of the model. Therefore, this study proposes a novel buildings edge-aware refined deep learning network (BEARNet) for building extraction from high-resolution remote sensing images. The network takes the building edge as a priori knowledge, learns the building edge features by decoupling the building body and edge, and further optimizes the network by combining the multi-objective loss function to strengthen the pertinence of building edge features extraction. Experimental results show that on the WHU building dataset, which is less difficult to extract buildings, compared with other methods, BEARNet has the highest Precision, F1, IoU and OA values of 97.70%, 97.42%, 95.3% and 98.67%. On the Massachusetts building dataset, where building extraction is difficult, BEARNet has the highest Precision, Recall, F1, IoU and OA compared to other methods, with values of 84.92%, 85.27%, 85.09%, 75.82% and 93.99%, respectively. Our proposed method is more accurate in extracting complex shapes and dense small-scale buildings, and the building edges are more refined and complete. Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart cities. At present, the knowledge-driven building extraction methods mostly rely on building prior knowledge of manual design, resulting in low automation and poor versatility. The data-driven methods usually rely too much on training samples, resulting in insufficient pertinence for the features of building extraction and low generalization ability of the model. Therefore, this study proposes a novel buildings edge-aware refined network (BEARNet) for building extraction from high-resolution remote sensing images. The network takes the building edge as a priori knowledge, learns the building edge features by decoupling the building body and edge, and further optimizes the network by combining the multiobjective loss function to strengthen the pertinence of building edge features extraction. Experimental results show that on the WHU building dataset, which is less difficult to extract buildings, compared with other methods, BEARNet has the highest Precision, F1, IoU, and OA values of 97.70%, 97.42%, 95.3%, and 98.67%. On the Massachusetts building dataset, where building extraction is difficult, BEARNet has the highest Precision, Recall, F1, IoU, and OA compared to other methods, with values of 84.92%, 85.27%, 85.09%, 75.82%, and 93.99%, respectively. Our proposed method is more accurate in extracting complex shapes and dense small-scale buildings, and the building edges are more refined and complete. |
Author | Yu, Hongye Zheng, Nanshan Lin, Huijing Hao, Ming Luo, Weiqiang |
Author_xml | – sequence: 1 givenname: Huijing orcidid: 0000-0002-8561-2502 surname: Lin fullname: Lin, Huijing organization: Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou, China – sequence: 2 givenname: Ming orcidid: 0000-0002-4676-7309 surname: Hao fullname: Hao, Ming organization: Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou, China – sequence: 3 givenname: Weiqiang surname: Luo fullname: Luo, Weiqiang organization: Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou, China – sequence: 4 givenname: Hongye surname: Yu fullname: Yu, Hongye organization: Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou, China – sequence: 5 givenname: Nanshan orcidid: 0000-0002-5474-1854 surname: Zheng fullname: Zheng, Nanshan organization: Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou, China |
BookMark | eNpNkFtLwzAUgINMcJv-AMGHgM-duTRt6tsccwpDYSr4Ftr0tOvckpm0bv57WzfEp3M4fOf2DVDPWAMIXVIyopQkN_PZ4mXECOMjzmLGBT9BfSqEDIiIaa_LQxGIRL6foYH3K0JYKGXcR_u76XjxBPUtHmNjv2CNs6Za55UpPYa8hCDdpQ6wg6IykGMD9c66D1xY9wdi2Ncu1XVlDS6c3eBlVS4DB96um9-ig42tAXswvsOrTVqCP0enRbr2cHGMQ_R2P32dPATz59njZDwPNEvCOmBhFulYxpEmOdGayCInCY8EJWksZJ6lnHJNCxFLUoDmIomiTHMZ0aTVwjjjQ3R9mLt19rMBX6uVbZxpVyomaShaWVHYUvRAaWe9b79VW9fe6b4VJaoTrDrBqhOsjoLbnqtDTwUA_3hK2_0R_wHsfHme |
CODEN | IGRSBY |
CitedBy_id | crossref_primary_10_3390_rs15245638 |
Cites_doi | 10.1016/j.landusepol.2020.105201 10.3390/rs13163297 10.1109/JSTARS.2022.3175200 10.1109/TPAMI.2020.2983686 10.3390/rs13183710 10.1109/TGRS.2012.2200689 10.1080/01431161.2011.606852 10.1007/978-3-030-22808-8_38 10.1109/3DV.2016.79 10.3390/rs14194889 10.1109/LGRS.2022.3197319 10.1080/17538947.2020.1831087 10.1109/TGRS.2018.2858817 10.3390/rs10030407 10.3390/rs14020330 |
ContentType | Journal Article |
Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023 |
Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023 |
DBID | 97E RIA RIE AAYXX CITATION 7SC 7SP 7TG 7UA 8FD C1K F1W FR3 H8D H96 JQ2 KL. KR7 L.G L7M L~C L~D |
DOI | 10.1109/LGRS.2023.3272353 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Meteorological & Geoastrophysical Abstracts Water Resources Abstracts Technology Research Database Environmental Sciences and Pollution Management ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Aerospace Database Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources ProQuest Computer Science Collection Meteorological & Geoastrophysical Abstracts - Academic Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional |
DatabaseTitle | CrossRef Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Water Resources Abstracts Environmental Sciences and Pollution Management Computer and Information Systems Abstracts Professional Aerospace Database Meteorological & Geoastrophysical Abstracts Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Meteorological & Geoastrophysical Abstracts - Academic |
DatabaseTitleList | Civil Engineering Abstracts |
Database_xml | – sequence: 1 dbid: RIE name: IEEE Xplore url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Geography Geology |
EISSN | 1558-0571 |
EndPage | 1 |
ExternalDocumentID | 10_1109_LGRS_2023_3272353 10113866 |
Genre | orig-research |
GrantInformation_xml | – fundername: National Natural Science Foundation of China grantid: 42271368; U22A20569 funderid: 10.13039/501100001809 – fundername: Fundamental Research Funds for the Central Universities grantid: 2021YCPY0113 |
GroupedDBID | 0R~ 29I 4.4 5GY 6IK 97E AAJGR AASAJ ABQJQ ABVLG ACGFO ACIWK AENEX AFRAH AKJIK ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 EBS HZ~ H~9 IFIPE IPLJI JAVBF LAI M43 O9- OCL P2P RIA RIE RIG RNS ~02 5VS AAYXX AETIX AIBXA CITATION EJD 7SC 7SP 7TG 7UA 8FD C1K F1W FR3 H8D H96 JQ2 KL. KR7 L.G L7M L~C L~D |
ID | FETCH-LOGICAL-c294t-24b6c7876c0d0cc08fd0936510a758dba313c1f5780fec35966bc386191102323 |
IEDL.DBID | RIE |
ISSN | 1545-598X |
IngestDate | Fri Sep 13 02:27:42 EDT 2024 Fri Aug 23 03:19:59 EDT 2024 Wed Jun 26 19:28:54 EDT 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c294t-24b6c7876c0d0cc08fd0936510a758dba313c1f5780fec35966bc386191102323 |
ORCID | 0000-0002-4676-7309 0000-0002-8561-2502 0000-0002-5474-1854 |
PQID | 2814572364 |
PQPubID | 75725 |
PageCount | 1 |
ParticipantIDs | crossref_primary_10_1109_LGRS_2023_3272353 proquest_journals_2814572364 ieee_primary_10113866 |
PublicationCentury | 2000 |
PublicationDate | 2023-01-01 |
PublicationDateYYYYMMDD | 2023-01-01 |
PublicationDate_xml | – month: 01 year: 2023 text: 2023-01-01 day: 01 |
PublicationDecade | 2020 |
PublicationPlace | Piscataway |
PublicationPlace_xml | – name: Piscataway |
PublicationTitle | IEEE geoscience and remote sensing letters |
PublicationTitleAbbrev | LGRS |
PublicationYear | 2023 |
Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
References | li (ref12) 2020 ref15 ref14 simonyan (ref11) 2015 ref20 woo (ref10) 2018 ref21 ref2 ref1 ref16 ref19 ref18 ref8 ref7 jaderberg (ref13) 2015 ref4 ref3 ref6 ref5 mnih (ref17) 2013 yu (ref9) 2022; 112 |
References_xml | – ident: ref1 doi: 10.1016/j.landusepol.2020.105201 – ident: ref3 doi: 10.3390/rs13163297 – start-page: 1 year: 2015 ident: ref11 article-title: Very deep convolutional networks for large-scale image recognition publication-title: Proc Int Conf Learn Represent contributor: fullname: simonyan – ident: ref21 doi: 10.1109/JSTARS.2022.3175200 – ident: ref20 doi: 10.1109/TPAMI.2020.2983686 – ident: ref7 doi: 10.3390/rs13183710 – ident: ref5 doi: 10.1109/TGRS.2012.2200689 – ident: ref2 doi: 10.1080/01431161.2011.606852 – start-page: 775 year: 2020 ident: ref12 publication-title: Semantic Flow for Fast and Accurate Scene Parsing contributor: fullname: li – ident: ref15 doi: 10.1007/978-3-030-22808-8_38 – start-page: 3 year: 2018 ident: ref10 article-title: CBAM: Convolutional block attention module publication-title: Proc Eur Conf Comput Vis (ECCV) contributor: fullname: woo – year: 2013 ident: ref17 publication-title: Machine Learning for Aerial Image Labeling contributor: fullname: mnih – start-page: 1 year: 2015 ident: ref13 article-title: Spatial transformer networks publication-title: Proc Adv Neural Inf Process Syst (NIPS) 29th Annu Conf Neural Inf Process Syst contributor: fullname: jaderberg – ident: ref14 doi: 10.1109/3DV.2016.79 – ident: ref18 doi: 10.3390/rs14194889 – ident: ref8 doi: 10.1109/LGRS.2022.3197319 – ident: ref19 doi: 10.1080/17538947.2020.1831087 – volume: 112 year: 2022 ident: ref9 article-title: SNNFD, spiking neural segmentation network in frequency domain using high spatial resolution images for building extraction publication-title: Int J Appl Earth Observ Geoinf contributor: fullname: yu – ident: ref16 doi: 10.1109/TGRS.2018.2858817 – ident: ref6 doi: 10.3390/rs10030407 – ident: ref4 doi: 10.3390/rs14020330 |
SSID | ssj0024887 |
Score | 2.4217725 |
Snippet | Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart... |
SourceID | proquest crossref ieee |
SourceType | Aggregation Database Publisher |
StartPage | 1 |
SubjectTerms | Automation building edge building extraction Buildings Data mining Datasets Decoupling deep learning Feature extraction High resolution Image edge detection Image resolution Indexes Methods prior knowledge Remote sensing remote sensing images Resolution Shape Spatial resolution |
Title | BEARNet: A novel buildings edge-aware refined network for building extraction from high-resolution remote sensing images |
URI | https://ieeexplore.ieee.org/document/10113866 https://www.proquest.com/docview/2814572364/abstract/ |
Volume | 20 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB61lSq4FFqKWCjIB05IDo7tZOPetlUfqsoeCpX2FsX2rKiAbLWbBdpf37HjLS8h9ZbDOLI9D8_YM_MBvC1KJKm1lvtiqLiW0nCryRiStUTrlZs6DNXIH8bl6aU-mxSTVKwea2EQMSafYRY-41u-n7lluCojDc9zVZXlOqxXQvbFWr8a61URDS-4BLww1SQ9YebCvD8_ufiYBZzwTMmhVIX64xCKqCr_mOJ4vhw_gfFqZn1ayZds2dnM3f7VtPHBU38KW8nTZKNeNLZhDdsdeJRAzz_f7MDmSUT1vXkGPw-ORhdj7PbZiLWz7_iV2YSWvWDhvo03P5o5MloH-aSetX3qOCN_956QkZGf90USLJSssNAHmVMsn0SbBpNQIFuEhHkiv_pGhmyxC5fHR58OT3mCZOBOGt1xqW3pSMdLJ7xwTlRTL4wqSbEbCjy8bVSuXD4lMyCm6FRBwZR1tG6KCkOPCKmew0Y7a_EFMGOtQE8OkpaoXeWNcSY091eqKSvduAG8W_Govu47b9QxYhGmDgytA0PrxNAB7IY9_42w3-4B7K3YWiflXNSyynUxDJ3zX_5n2Ct4HP7eX7XswUY3X-Jrcj46-yYK3R1-VdXY |
link.rule.ids | 315,786,790,802,27957,27958,55109 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3db9MwED-NTWi8wBhDFMbwA09IDontpDFvZdpWoOvD2KS-RbF9FdMgRW3KNv56zo47YGgSb3k4K7bvw3f23f0AXucFktQaw13el1wJoblRZAzJWqJx0k4t-mrk43ExPFMfJ_kkFquHWhhEDMlnmPjP8JbvZnbpr8pIw7NMlkVxDzbooE91V671u7VeGfDwvFPAc11O4iMmEb4dHZ18TjxSeCJFX8hc_nUMBVyVf4xxOGEOH8F4NbcuseQiWbYmsT9vtW3878lvwcPoa7JBJxyPYQ2bbdiMsOdfrrfh_lHA9b1-AlfvDwYnY2zfsQFrZj_wKzMRL3vB_I0bry_rOTJaB3mljjVd8jgjj_eGkJGZn3dlEswXrTDfCZlTNB-FmwaTWCBb-JR5Ij__RqZssQNnhwen-0MeQRm4FVq1XChTWNLywqYutTYtpy7VsiB21BR6OFPLTNpsSoYgnaKVOYVTxtK6KS70XSKEfArrzazBZ8C0MSk6cpGUQGVLp7XVvr2_lHVRqtr24M2KR9X3rvdGFWKWVFeeoZVnaBUZ2oMdv-d_EHbb3YPdFVurqJ6LSpSZyvu-d_7zO4a9gs3h6fGoGn0Yf3oBD_yfuouXXVhv50t8Sa5Ia_aCAP4C8a3ZLg |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=BEARNet%3A+A+novel+buildings+edge-aware+refined+network+for+building+extraction+from+high-resolution+remote+sensing+images&rft.jtitle=IEEE+geoscience+and+remote+sensing+letters&rft.au=Lin%2C+Huijing&rft.au=Hao%2C+Ming&rft.au=Luo%2C+Weiqiang&rft.au=Yu%2C+Hongye&rft.date=2023-01-01&rft.pub=IEEE&rft.issn=1545-598X&rft.eissn=1558-0571&rft.spage=1&rft.epage=1&rft_id=info:doi/10.1109%2FLGRS.2023.3272353&rft.externalDocID=10113866 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1545-598X&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1545-598X&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1545-598X&client=summon |