Boundary-enhanced dual-stream network for semantic segmentation of high-resolution remote sensing images

Deep convolutional neural networks (DCNNs) have been successfully used in semantic segmentation of high-resolution remote sensing images (HRSIs). However, this task still suffers from intra-class inconsistency and boundary blur due to high intra-class heterogeneity and inter-class homogeneity, consi...

Full description

Saved in:
Bibliographic Details
Published inGIScience and remote sensing Vol. 61; no. 1
Main Authors Li, Xinghua, Xie, Linglin, Wang, Caifeng, Miao, Jianhao, Shen, Huanfeng, Zhang, Liangpei
Format Journal Article
LanguageEnglish
Published Taylor & Francis 31.12.2024
Taylor & Francis Group
Subjects
boundary blur
CNN
HRSIs
intra-class inconsistency
Semantic segmentation
Online AccessGet full text

Cover

Abstract Deep convolutional neural networks (DCNNs) have been successfully used in semantic segmentation of high-resolution remote sensing images (HRSIs). However, this task still suffers from intra-class inconsistency and boundary blur due to high intra-class heterogeneity and inter-class homogeneity, considerable scale variance, and spatial information loss in conventional DCNN-based methods. Therefore, a novel boundary-enhanced dual-stream network (BEDSN) is proposed, in which an edge detection branch stream (EDBS) with a composite loss function is introduced to compensate for boundary loss in semantic segmentation branch stream (SSBS). EDBS and SSBS are integrated by highly coupled encoder and feature extractor. A lightweight multilevel information fusion module guided by channel attention mechanism is designed to reuse intermediate boundary information effectively. For aggregating multiscale contextual information, SSBS is enhanced by multiscale feature extraction module and hybrid atrous convolution module. Extensive experiments have been tested on ISPRS Vaihingen and Potsdam datasets. Results show that BEDSN can achieve significant improvements in intra-class consistency and boundary refinement. Compared with 11 state-of-the-art methods, BEDSN exhibits higher-level performance in both quantitative and visual assessments with low model complexity. The code will be available at https://github.com/lixinghua5540/BEDSN .
AbstractList Deep convolutional neural networks (DCNNs) have been successfully used in semantic segmentation of high-resolution remote sensing images (HRSIs). However, this task still suffers from intra-class inconsistency and boundary blur due to high intra-class heterogeneity and inter-class homogeneity, considerable scale variance, and spatial information loss in conventional DCNN-based methods. Therefore, a novel boundary-enhanced dual-stream network (BEDSN) is proposed, in which an edge detection branch stream (EDBS) with a composite loss function is introduced to compensate for boundary loss in semantic segmentation branch stream (SSBS). EDBS and SSBS are integrated by highly coupled encoder and feature extractor. A lightweight multilevel information fusion module guided by channel attention mechanism is designed to reuse intermediate boundary information effectively. For aggregating multiscale contextual information, SSBS is enhanced by multiscale feature extraction module and hybrid atrous convolution module. Extensive experiments have been tested on ISPRS Vaihingen and Potsdam datasets. Results show that BEDSN can achieve significant improvements in intra-class consistency and boundary refinement. Compared with 11 state-of-the-art methods, BEDSN exhibits higher-level performance in both quantitative and visual assessments with low model complexity. The code will be available at https://github.com/lixinghua5540/BEDSN.
Deep convolutional neural networks (DCNNs) have been successfully used in semantic segmentation of high-resolution remote sensing images (HRSIs). However, this task still suffers from intra-class inconsistency and boundary blur due to high intra-class heterogeneity and inter-class homogeneity, considerable scale variance, and spatial information loss in conventional DCNN-based methods. Therefore, a novel boundary-enhanced dual-stream network (BEDSN) is proposed, in which an edge detection branch stream (EDBS) with a composite loss function is introduced to compensate for boundary loss in semantic segmentation branch stream (SSBS). EDBS and SSBS are integrated by highly coupled encoder and feature extractor. A lightweight multilevel information fusion module guided by channel attention mechanism is designed to reuse intermediate boundary information effectively. For aggregating multiscale contextual information, SSBS is enhanced by multiscale feature extraction module and hybrid atrous convolution module. Extensive experiments have been tested on ISPRS Vaihingen and Potsdam datasets. Results show that BEDSN can achieve significant improvements in intra-class consistency and boundary refinement. Compared with 11 state-of-the-art methods, BEDSN exhibits higher-level performance in both quantitative and visual assessments with low model complexity. The code will be available at https://github.com/lixinghua5540/BEDSN .
Author Shen, Huanfeng
Zhang, Liangpei
Wang, Caifeng
Li, Xinghua
Xie, Linglin
Miao, Jianhao
Author_xml – sequence: 1
  givenname: Xinghua
  surname: Li
  fullname: Li, Xinghua
  organization: Wuhan University
– sequence: 2
  givenname: Linglin
  surname: Xie
  fullname: Xie, Linglin
  email: xll@img.net
  organization: Ministry of Natural Resources
– sequence: 3
  givenname: Caifeng
  surname: Wang
  fullname: Wang, Caifeng
  organization: Wuhan University
– sequence: 4
  givenname: Jianhao
  surname: Miao
  fullname: Miao, Jianhao
  organization: Wuhan University
– sequence: 5
  givenname: Huanfeng
  surname: Shen
  fullname: Shen, Huanfeng
  organization: Collaborative Innovation Center of Geospatial Technology
– sequence: 6
  givenname: Liangpei
  surname: Zhang
  fullname: Zhang, Liangpei
  organization: Wuhan University
BookMark eNqFkc9uEzEQxi1UJNrCIyDtC2zqP2tnLS6FqoVKlbjA2ZrY443Lro1sRyhvXydpOXAopxmNvu8bzW8uyFlMEQn5yOiK0ZFeMTmMTFGx4pQPKy6kElK-IedMD6Jfc67OWt80_UH0jlyU8kipkIzJc7L9knbRQd73GLcQLbrO7WDuS80ISxex_kn5V-dT7gouEGuwrZkWjBVqSLFLvtuGadtnLGneHUcZl1SxyWIJcerCAhOW9-Sth7ngh-d6SX7e3f64-dY_fP96f_P5obcDF7VXjlmrgDq2EQ6HkUvuuPBaa-A4KslgDcLLUTmhqQDtmXZc2REFMk2lFZfk_pTrEjya37ltz3uTIJjjIOXJQG5XzGioB_CjZnJD2YCDg7UWfoRBblBuuMKWJU9ZNqdSMvq_eYyaA3rzgt4c0Jtn9M336R-fDSdcNUOY_-u-PrlDbNgXaA-Ynamwn1P2uf0oFCNej3gCTdKg3A
CitedBy_id crossref_primary_10_1145_3721984
crossref_primary_10_1016_j_dsp_2024_104885
crossref_primary_10_1109_JSTARS_2025_3525634
crossref_primary_10_1109_JSTARS_2024_3471638
crossref_primary_10_1109_JSTARS_2024_3444773
crossref_primary_10_1007_s10489_025_06433_1
crossref_primary_10_1080_15481603_2024_2426589
crossref_primary_10_1109_JSTARS_2025_3528650
crossref_primary_10_1016_j_image_2024_117238
crossref_primary_10_1109_JSTARS_2024_3470316
crossref_primary_10_1109_TCSVT_2024_3495769
crossref_primary_10_1109_JSTARS_2024_3456854
crossref_primary_10_1109_TGRS_2025_3526247
crossref_primary_10_1364_JOSAA_526142
crossref_primary_10_1109_TGRS_2024_3477749
crossref_primary_10_1109_TGRS_2024_3502401
crossref_primary_10_1016_j_ecoinf_2024_102818
crossref_primary_10_1016_j_jag_2024_104083
crossref_primary_10_1109_TGRS_2024_3492715
crossref_primary_10_1109_TGRS_2024_3507784
crossref_primary_10_1109_JSTARS_2024_3485239
crossref_primary_10_1109_TGRS_2024_3507274
Cites_doi 10.1109/CVPR.2017.549
10.1007/978-3-319-24574-4_28
10.3390/ijgi10010022
10.1109/CVPR.2018.00199
10.1109/TGRS.2004.843193
10.3390/rs12040701
10.1007/978-3-030-00889-5_1
10.48550/arXiv.1906.11428
10.1007/s11263-021-01515-2
10.1080/17476938708814211
10.1109/ICCV.2015.164
10.1109/ICCV.2011.6126474
10.3390/rs9060522
10.1007/s11063-019-10174-x
10.1109/TPAMI.2016.2644615
10.1109/WACV.2018.00163
10.1016/j.jag.2011.06.008
10.1109/CVPRW.2018.00050
10.1109/CVPR.2017.660
10.1109/JSTARS.2021.3076035
10.1109/CVPR.2018.00813
10.1109/tits.2020.2972974
10.1109/LGRS.2023.3234257
10.1007/978-3-030-01234-2_1
10.48550/arXiv.1511.07122
10.1109/JSTARS.2021.3071353
10.1016/j.isprsjprs.2017.06.001
10.1109/CVPR.2016.492
10.3390/rs10091339
10.48550/arXiv.1706.05587
10.1109/ACCESS.2021.3065695
10.3390/rs11151774
10.1109/LGRS.2020.2988294
10.1109/TGRS.2020.2964675
10.1016/j.isprsjprs.2020.01.013
10.1109/CVPR.2016.90
10.1109/ICECCT.2017.8117946
10.1109/ICIP.2019.8803132
10.1109/JSTARS.2021.3073935
10.3390/ijgi9100601
10.1109/CVPR.2019.00326
10.1109/TPAMI.2017.2699184
10.1016/j.isprsjprs.2017.11.009
10.1109/JSTARS.2018.2833382
10.1109/JSTARS.2017.2747599
10.1109/CVPR.2015.7299173
10.23915/distill.00003
10.1109/CVPR.2017.353
10.1109/TGRS.2020.2994150
10.1109/TGRS.2003.813271
10.1109/TGRS.2021.3119537
10.3390/app122111248
10.1109/CVPR.2018.00745
10.1109/ISM46123.2019.00049
10.3390/rs11070830
10.3390/rs12040633
10.1080/01431161.2020.1871100
10.1109/cvpr.2015.7298594
10.1109/CVPR.2017.243
10.1016/j.neucom.2019.02.003
10.1061/(ASCE)CP.1943-5487.0000947
10.1109/TPAMI.2020.2983686
10.1016/j.rse.2010.12.017
10.1109/ACCESS.2019.2917952
10.1109/TPAMI.2018.2878849
10.1109/TPAMI.2016.2572683
10.1109/TGRS.2021.3065112
10.1109/TPAMI.2020.3007032
10.1007/978-3-030-01234-2_49
10.1080/17538947.2020.1831087
ContentType Journal Article
Copyright 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. 2024
Copyright_xml – notice: 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. 2024
DBID 0YH
AAYXX
CITATION
DOA
DOI 10.1080/15481603.2024.2356355
DatabaseName Taylor & Francis Open Access
CrossRef
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
DatabaseTitleList

Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: 0YH
  name: Taylor & Francis Open Access
  url: https://www.tandfonline.com
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Astronomy & Astrophysics
EISSN 1943-7226
ExternalDocumentID oai_doaj_org_article_0faaf8915b014e4da793f8a45be5b26e
10_1080_15481603_2024_2356355
2356355
Genre Method
GrantInformation_xml – fundername: Hubei Luojia Laboratory
  grantid: 220100055
– fundername: National Natural Science Foundation of China
  grantid: 42171302
– fundername: Open Fund of Key Laboratory of Natural Resources Monitoring and Supervision in Southern Hilly Region, Ministry of Natural Resources
  grantid: NRMSSHR2022Z03
GroupedDBID 0YH
30N
4.4
5GY
AAHBH
AAJMT
ABCCY
ABFIM
ABPEM
ABTAI
ACGFS
ACTIO
ADCVX
AEISY
AENEX
AEYOC
AIJEM
ALMA_UNASSIGNED_HOLDINGS
ALQZU
AQRUH
AVBZW
BLEHA
CCCUG
CS3
DGEBU
DKSSO
DU5
EBS
E~A
E~B
GROUPED_DOAJ
GTTXZ
H13
HZ~
H~P
IPNFZ
KYCEM
LJTGL
M4Z
O9-
OK1
RIG
S-T
SNACF
TDBHL
TEI
TFL
TFT
TFW
TTHFI
UT5
~02
AAYXX
AIYEW
CITATION
ID FETCH-LOGICAL-c423t-6d1cc6a0d1b3de48252d23f999a2e8651a7a3f586d3903a9f19d26c8e3e1905c3
IEDL.DBID DOA
ISSN 1548-1603
IngestDate Wed Aug 27 01:27:32 EDT 2025
Tue Jul 01 02:27:29 EDT 2025
Thu Apr 24 23:10:22 EDT 2025
Wed Dec 25 09:04:03 EST 2024
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 1
Language English
License open-access: http://creativecommons.org/licenses/by-nc/4.0/: This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c423t-6d1cc6a0d1b3de48252d23f999a2e8651a7a3f586d3903a9f19d26c8e3e1905c3
OpenAccessLink https://doaj.org/article/0faaf8915b014e4da793f8a45be5b26e
ParticipantIDs informaworld_taylorfrancis_310_1080_15481603_2024_2356355
crossref_citationtrail_10_1080_15481603_2024_2356355
doaj_primary_oai_doaj_org_article_0faaf8915b014e4da793f8a45be5b26e
crossref_primary_10_1080_15481603_2024_2356355
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2024-12-31
PublicationDateYYYYMMDD 2024-12-31
PublicationDate_xml – month: 12
  year: 2024
  text: 2024-12-31
  day: 31
PublicationDecade 2020
PublicationTitle GIScience and remote sensing
PublicationYear 2024
Publisher Taylor & Francis
Taylor & Francis Group
Publisher_xml – name: Taylor & Francis
– name: Taylor & Francis Group
References e_1_3_3_52_1
e_1_3_3_50_1
e_1_3_3_71_1
Molchanov P. (e_1_3_3_41_1) 2016
e_1_3_3_18_1
e_1_3_3_39_1
e_1_3_3_14_1
e_1_3_3_37_1
e_1_3_3_16_1
e_1_3_3_35_1
e_1_3_3_58_1
e_1_3_3_10_1
e_1_3_3_33_1
e_1_3_3_56_1
e_1_3_3_12_1
e_1_3_3_31_1
e_1_3_3_54_1
e_1_3_3_73_1
e_1_3_3_40_1
e_1_3_3_63_1
e_1_3_3_61_1
e_1_3_3_7_1
e_1_3_3_9_1
e_1_3_3_29_1
e_1_3_3_25_1
e_1_3_3_48_1
e_1_3_3_27_1
e_1_3_3_46_1
e_1_3_3_69_1
e_1_3_3_3_1
e_1_3_3_21_1
e_1_3_3_44_1
e_1_3_3_67_1
e_1_3_3_5_1
e_1_3_3_23_1
e_1_3_3_42_1
e_1_3_3_65_1
e_1_3_3_30_1
e_1_3_3_51_1
e_1_3_3_70_1
e_1_3_3_17_1
e_1_3_3_19_1
e_1_3_3_13_1
e_1_3_3_38_1
e_1_3_3_59_1
e_1_3_3_15_1
e_1_3_3_36_1
e_1_3_3_57_1
e_1_3_3_34_1
e_1_3_3_55_1
e_1_3_3_72_1
e_1_3_3_11_1
e_1_3_3_32_1
e_1_3_3_53_1
e_1_3_3_74_1
e_1_3_3_62_1
e_1_3_3_60_1
e_1_3_3_6_1
e_1_3_3_8_1
e_1_3_3_28_1
e_1_3_3_24_1
e_1_3_3_49_1
e_1_3_3_26_1
e_1_3_3_47_1
e_1_3_3_68_1
e_1_3_3_2_1
e_1_3_3_20_1
e_1_3_3_45_1
e_1_3_3_66_1
e_1_3_3_4_1
e_1_3_3_22_1
e_1_3_3_43_1
e_1_3_3_64_1
References_xml – ident: e_1_3_3_31_1
  doi: 10.1109/CVPR.2017.549
– ident: e_1_3_3_48_1
  doi: 10.1007/978-3-319-24574-4_28
– ident: e_1_3_3_68_1
  doi: 10.3390/ijgi10010022
– ident: e_1_3_3_67_1
  doi: 10.1109/CVPR.2018.00199
– ident: e_1_3_3_27_1
  doi: 10.1109/TGRS.2004.843193
– ident: e_1_3_3_69_1
  doi: 10.3390/rs12040701
– ident: e_1_3_3_73_1
  doi: 10.1007/978-3-030-00889-5_1
– ident: e_1_3_3_71_1
  doi: 10.48550/arXiv.1906.11428
– ident: e_1_3_3_24_1
– ident: e_1_3_3_65_1
  doi: 10.1007/s11263-021-01515-2
– ident: e_1_3_3_5_1
  doi: 10.1080/17476938708814211
– ident: e_1_3_3_61_1
  doi: 10.1109/ICCV.2015.164
– ident: e_1_3_3_16_1
  doi: 10.1109/ICCV.2011.6126474
– ident: e_1_3_3_36_1
  doi: 10.3390/rs9060522
– ident: e_1_3_3_54_1
  doi: 10.1007/s11063-019-10174-x
– ident: e_1_3_3_2_1
  doi: 10.1109/TPAMI.2016.2644615
– ident: e_1_3_3_55_1
  doi: 10.1109/WACV.2018.00163
– ident: e_1_3_3_59_1
  doi: 10.1016/j.jag.2011.06.008
– ident: e_1_3_3_49_1
  doi: 10.1109/CVPRW.2018.00050
– ident: e_1_3_3_70_1
  doi: 10.1109/CVPR.2017.660
– ident: e_1_3_3_51_1
– ident: e_1_3_3_44_1
  doi: 10.1109/JSTARS.2021.3076035
– ident: e_1_3_3_56_1
  doi: 10.1109/CVPR.2018.00813
– ident: e_1_3_3_15_1
  doi: 10.1109/tits.2020.2972974
– ident: e_1_3_3_28_1
  doi: 10.1109/LGRS.2023.3234257
– ident: e_1_3_3_60_1
  doi: 10.1007/978-3-030-01234-2_1
– ident: e_1_3_3_66_1
  doi: 10.48550/arXiv.1511.07122
– ident: e_1_3_3_72_1
  doi: 10.1109/JSTARS.2021.3071353
– ident: e_1_3_3_39_1
  doi: 10.1016/j.isprsjprs.2017.06.001
– ident: e_1_3_3_3_1
  doi: 10.1109/CVPR.2016.492
– ident: e_1_3_3_34_1
  doi: 10.3390/rs10091339
– ident: e_1_3_3_7_1
  doi: 10.48550/arXiv.1706.05587
– ident: e_1_3_3_57_1
  doi: 10.1109/ACCESS.2021.3065695
– ident: e_1_3_3_64_1
  doi: 10.3390/rs11151774
– ident: e_1_3_3_32_1
  doi: 10.1109/LGRS.2020.2988294
– ident: e_1_3_3_11_1
  doi: 10.1109/TGRS.2020.2964675
– ident: e_1_3_3_9_1
  doi: 10.1016/j.isprsjprs.2020.01.013
– ident: e_1_3_3_19_1
  doi: 10.1109/CVPR.2016.90
– ident: e_1_3_3_50_1
  doi: 10.1109/ICECCT.2017.8117946
– start-page: 1
  volume-title: International Conference on Learning Representations (ICLR)
  year: 2016
  ident: e_1_3_3_41_1
– ident: e_1_3_3_38_1
  doi: 10.1109/ICIP.2019.8803132
– ident: e_1_3_3_22_1
  doi: 10.1109/JSTARS.2021.3073935
– ident: e_1_3_3_52_1
  doi: 10.3390/ijgi9100601
– ident: e_1_3_3_17_1
  doi: 10.1109/CVPR.2019.00326
– ident: e_1_3_3_6_1
  doi: 10.1109/TPAMI.2017.2699184
– ident: e_1_3_3_40_1
  doi: 10.1016/j.isprsjprs.2017.11.009
– ident: e_1_3_3_30_1
  doi: 10.1109/JSTARS.2018.2833382
– ident: e_1_3_3_4_1
  doi: 10.1109/JSTARS.2017.2747599
– ident: e_1_3_3_18_1
  doi: 10.1109/CVPR.2015.7299173
– ident: e_1_3_3_45_1
  doi: 10.23915/distill.00003
– ident: e_1_3_3_47_1
  doi: 10.1109/CVPR.2017.353
– ident: e_1_3_3_10_1
  doi: 10.1109/TGRS.2020.2994150
– ident: e_1_3_3_14_1
  doi: 10.1109/TGRS.2003.813271
– ident: e_1_3_3_74_1
  doi: 10.1109/TGRS.2021.3119537
– ident: e_1_3_3_26_1
  doi: 10.3390/app122111248
– ident: e_1_3_3_23_1
  doi: 10.1109/CVPR.2018.00745
– ident: e_1_3_3_25_1
  doi: 10.1109/ISM46123.2019.00049
– ident: e_1_3_3_35_1
  doi: 10.3390/rs11070830
– ident: e_1_3_3_63_1
  doi: 10.3390/rs12040633
– ident: e_1_3_3_62_1
  doi: 10.1080/01431161.2020.1871100
– ident: e_1_3_3_53_1
  doi: 10.1109/cvpr.2015.7298594
– ident: e_1_3_3_20_1
  doi: 10.1109/CVPR.2017.243
– ident: e_1_3_3_29_1
  doi: 10.1016/j.neucom.2019.02.003
– ident: e_1_3_3_46_1
  doi: 10.1061/(ASCE)CP.1943-5487.0000947
– ident: e_1_3_3_58_1
  doi: 10.1109/TPAMI.2020.2983686
– ident: e_1_3_3_42_1
  doi: 10.1016/j.rse.2010.12.017
– ident: e_1_3_3_12_1
  doi: 10.1109/ACCESS.2019.2917952
– ident: e_1_3_3_33_1
  doi: 10.1109/TPAMI.2018.2878849
– ident: e_1_3_3_37_1
  doi: 10.1109/TPAMI.2016.2572683
– ident: e_1_3_3_43_1
  doi: 10.1109/TGRS.2021.3065112
– ident: e_1_3_3_21_1
  doi: 10.1109/TPAMI.2020.3007032
– ident: e_1_3_3_8_1
  doi: 10.1007/978-3-030-01234-2_49
– ident: e_1_3_3_13_1
  doi: 10.1080/17538947.2020.1831087
SSID ssj0035115
Score 2.5432281
Snippet Deep convolutional neural networks (DCNNs) have been successfully used in semantic segmentation of high-resolution remote sensing images (HRSIs). However, this...
SourceID doaj
crossref
informaworld
SourceType Open Website
Enrichment Source
Index Database
Publisher
SubjectTerms boundary blur
CNN
HRSIs
intra-class inconsistency
Semantic segmentation
SummonAdditionalLinks – databaseName: Taylor & Francis Open Access
  dbid: 0YH
  link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT9wwELYoXHqpeFVsC8gHxM008SMbH6ECrThwYiU4RX4CUpOtNtvD_vvOOM4KKrU9cEsiO4k843l55htCzsBjwD5TBTOl90waK5hRXoIwlIILEXXtUrbFXTWby9sHNWYT9jmtEn3oOABFJFmNm9vYfsyI-4ZWNnZHBu-OywsuFCrND2SHI7cCSxePs1EY4zGZSpCpEpwlmDMW8fztNW_UU0Lx_wPD9JX2udkln7LZSC8HOu-RrdDtk6PLHgPZi3ZNz2m6HuIU_QF5vkrtkpZrFrrndMhPseiKYWmIaWk3JH9T-CDtQwur--Lg4qnNlUgdXUSKSMYMvPHMnHQZgKwBhnUYX6AvLYii_pDMb67vv89YbqrAHFhOK1b50rnKFL60wgcJDiL3HGiiteGhrlRppkZEVVde6EIYHUvteeXqIALYDsqJz2S7W3ThiFBYPKfKOPUqcBntVHvQiDVoXVlxbp2ZEDmuZeMy4jg2vvjRlBmYdCRBgyRoMgkm5GIz7ecAufG_CVdIqM1gRMxODxbLpyZvwKaIxsRal8qCUxikNyCYYm2kskFZXoUJ0a_J3KxSwCQO3U0a8c8f-PKOuV_JR7wdsCOPyfZq-SucgJ2zsqeJk38Dg4HyXg
  priority: 102
  providerName: Taylor & Francis
Title Boundary-enhanced dual-stream network for semantic segmentation of high-resolution remote sensing images
URI https://www.tandfonline.com/doi/abs/10.1080/15481603.2024.2356355
https://doaj.org/article/0faaf8915b014e4da793f8a45be5b26e
Volume 61
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07T8MwELaAiQXxFOUlD4jNkPjVeAQEqhiYQIIpcvwAJJKitgz8e-7sFBUGWNgsy06su9O97PuOkGOIGLDPVMFs6T2TthHMKi9BGUrBhYimcum1xa0e3cubB_Ww0OoL34RleOBMuLMiWhsrU6oGnPkgvQWBipWVqgmq4Tqg9gWbNw-msg7G2zGVkFIlxEi6EPPanao4wzmcgtiQy1MuFJrcb1Ypgff_gC5dMDrX62St9xbpeT7lBlkK3SbZPZ9i_nrcftATmsY5PTHdIs8XqUvS5IOF7jnd7VOstWJYEWJb2uU33xR-SKehBaK-OBg8tX0BUkfHkSKAMYMgvJdJOgnAzQDLOkwr0JcWNNB0m9xfX91djljfS4E5cJhmTPvSOW0LXzbCBwlxIfccWGGM5aHSqrRDK6KqtBemENbE0niuXRVEAJdBObFDVrpxF3YJBeI5VcahV4HL2AyNB0NYgbGVmvPG2QGRc1rWrgcax34Xr3XZ45HOWVAjC-qeBQNy-rXtLSNt_LXhAhn1tRiBstMEiE_di0_9l_gMiFlkcz1LeZKYm5rU4tcD7P3HAfbJKn4zY0cekJXZ5D0cgp8za47IcvE4OkqC_Qlt7_dY
linkProvider Directory of Open Access Journals
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Lb9QwELagHOAC5aUuTx8QNy-JXxsfW0S1QNlTK_UW-dlWbbLVbnoov54ZJ6mWSpRDb1ZiR_bYnldmviHkE1gMWGeqYLYMgUnrBLMqSGCGUnAhkql8jrZY6PmR_HGsjjdyYTCsEm3o1ANFZF6Nlxud0WNI3BdUs7E8Mph3XE65UCg1H5JHyugZVjEQxWLkxvifTGXMVAnWEowZs3j-9Zm_5FOG8b8FYrohfvafET9OvI86OZ9edW7qf9_CdLzfyrbJ00E7pbv9cXpOHsT2BdnZXaO_fNlc0880t3t3yPolOd3LVZlW1yy2pzmWgGJuF8MMFNvQto8xp7Asuo4NbOKZh8ZJMyQ8tXSZKAImMzD6hztAVxFOT4RuLbox6FkDHG_9ihztfzv8OmdD7QbmQUHrmA6l99oWoXQiRAl2KA8ctt4Yy2OlVWlnViRV6SBMIaxJpQlc-yqKCCqK8uI12WqXbdwhFOjhVZlmQUUuk5uZAIK3AuEuNefO2wmR447VfgA2x_oaF3U54J-OVK2RqvVA1QmZ3gy77JE9_jdgD4_DTWcE5s4PlquTerjndZGsTZUplQPbM8pggf-lykrlonJcxwkxm4ep7rJfJvVFVGpx5wTe3GPsR_J4fvjroD74vvj5ljzBVz1c5Tuy1a2u4ntQrTr3Id-dPzoJFeY
linkToPdf http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Jb9UwELagSIgLlE19LMUHxM2PxEtecmyhT2XREwcqcbO8thUkr3pJD-XXM-M4VakEPfRmJZ7Iy3gWZ-YbQt6Cx4B1pgpmSu-ZNFYwo7wEYSgFFyI2tUvRFqvq8Eh-_qGmaMI-h1WiDx1HoIgkq_Fwn_k4RcS9RysbqyODd8flnAuFSvMuuVcheDhmcRSrSRjjbzKVIFMlOEtAMyXx_Oszf6mnhOJ_DcP0ivZZPiJ2GvcYdPJzfj7Yuft9DdLxVhPbJg-zbUr3RmZ6TO6E7gnZ2evxtnzdXtB3NLXHy5D-KTnZTzWZNhcsdCcpkoBiZhfD_BPT0m6MMKcwK9qHFrbw1EHjuM3pTh1dR4pwyQxc_nwC6CYA7wTo1uElBj1tQd71z8jR8uD7h0OWKzcwB-bZwCpfOleZwpdW-CDBC-Wew8Y3jeGhrlRpFkZEVVdeNIUwTSwbzytXBxHAQFFOPCdb3boLO4TCejhVxoVXgctoF40HtVuDapcV59aZGZHThmmXYc2xusYvXWb002lVNa6qzqs6I_NLsrMR1-Mmgn3khsvOCMudHqw3xzqfcl1EY2LdlMqC5xmkNyD9Ym2kskFZXoUZaa7ykh7SrUwcS6ho8d8BvLgF7Rty_9vHpf76afXlJXmAb0asyldka9ich9dgVw12N52cP7TlFIo
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=Boundary-enhanced+dual-stream+network+for+semantic+segmentation+of+high-resolution+remote+sensing+images&rft.jtitle=GIScience+and+remote+sensing&rft.au=Xinghua+Li&rft.au=Linglin+Xie&rft.au=Caifeng+Wang&rft.au=Jianhao+Miao&rft.date=2024-12-31&rft.pub=Taylor+%26+Francis+Group&rft.issn=1548-1603&rft.eissn=1943-7226&rft.volume=61&rft.issue=1&rft_id=info:doi/10.1080%2F15481603.2024.2356355&rft.externalDBID=DOA&rft.externalDocID=oai_doaj_org_article_0faaf8915b014e4da793f8a45be5b26e
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1548-1603&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1548-1603&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1548-1603&client=summon