DF2Net: Differential Feature Fusion Network for Hyperspectral Image Classification
Recently, hybrid networks, combining graph convolutional networks (GCNs) and convolutional neural networks into a unified framework, have garnered significant attention in hyperspectral image (HSI) classification. However, existing hybrid networks have the following limitations. 1) Existing methods...
Saved in:
Published in | IEEE journal of selected topics in applied earth observations and remote sensing Vol. 17; pp. 10660 - 10673 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
Piscataway
IEEE
2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Recently, hybrid networks, combining graph convolutional networks (GCNs) and convolutional neural networks into a unified framework, have garnered significant attention in hyperspectral image (HSI) classification. However, existing hybrid networks have the following limitations. 1) Existing methods primarily utilize simple fusion strategies such as concatenation or direct addition, resulting in the ineffective utilization of advantageous features. 2) Traditional GCNs only consider the relationship between pairs of vertices, limiting their ability to capture complex high-order and long-range correlations. In this work, a novel differential feature fusion network (DF2Net) is proposed for HSI classification. Specifically, DF2Net utilizes two subnetworks to learn features at different abstraction levels: 1) the spectral-spatial hypergraph convolutional network for capturing complex high-order and long-range correlations, and the spectral-spatial convolution network for pixel-level local information extraction. Subsequently, we introduce an advantageous feature differential enhancement fusion module, in which mutual enhancement of advantageous features from different network structures is performed, thereby improving the classification robustness of different regions in HSI. The experiments on four HSI benchmark datasets demonstrate that our DF2Net exhibits superior advantages over state-of-the-art models, particularly when the training samples are limited. |
---|---|
AbstractList | Recently, hybrid networks, combining graph convolutional networks (GCNs) and convolutional neural networks into a unified framework, have garnered significant attention in hyperspectral image (HSI) classification. However, existing hybrid networks have the following limitations. 1) Existing methods primarily utilize simple fusion strategies such as concatenation or direct addition, resulting in the ineffective utilization of advantageous features. 2) Traditional GCNs only consider the relationship between pairs of vertices, limiting their ability to capture complex high-order and long-range correlations. In this work, a novel differential feature fusion network (DF2Net) is proposed for HSI classification. Specifically, DF2Net utilizes two subnetworks to learn features at different abstraction levels: 1) the spectral–spatial hypergraph convolutional network for capturing complex high-order and long-range correlations, and the spectral–spatial convolution network for pixel-level local information extraction. Subsequently, we introduce an advantageous feature differential enhancement fusion module, in which mutual enhancement of advantageous features from different network structures is performed, thereby improving the classification robustness of different regions in HSI. The experiments on four HSI benchmark datasets demonstrate that our DF2Net exhibits superior advantages over state-of-the-art models, particularly when the training samples are limited. |
Author | Shen, Tao Wang, Qingwang Meng, Yuanqin Huang, Jiangbo |
Author_xml | – sequence: 1 givenname: Qingwang orcidid: 0000-0001-5820-5357 surname: Wang fullname: Wang, Qingwang email: wangqingwang@kust.edu.cn organization: Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China – sequence: 2 givenname: Jiangbo orcidid: 0000-0002-6103-7769 surname: Huang fullname: Huang, Jiangbo email: jiangbohuang@stu.kust.edu.cn organization: Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China – sequence: 3 givenname: Yuanqin orcidid: 0009-0007-3920-814X surname: Meng fullname: Meng, Yuanqin email: mengyuanqin@stu.kust.edu.cn organization: Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China – sequence: 4 givenname: Tao orcidid: 0000-0003-1273-7950 surname: Shen fullname: Shen, Tao email: shentao@kust.edu.cn organization: Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China |
BookMark | eNpNkc1KAzEYRYMoWH-eQBcDrqcm-TI_cVdaayuiUHUdMumXklonNZkivr2pI-ImgXDuzYVzQg5b3yIhF4wOGaPy-v75ZbR4HnLKxRAEhbqEAzLgrGA5K6A4JAMmQeZMUHFMTmJcU1rySsKALCZT_ojdTTZx1mLAtnN6k01Rd7uA2XQXnW-zBHz68JZZH7LZ1xZD3KLpQgLn73qF2XijY3TWGd0l_IwcWb2JeP57n5LX6e3LeJY_PN3Nx6OH3EAhu9xaoJWoq6ZstGTSVhx0BWkYskYveSNRUwYowTCdMFqDRlsLy2wJZkkBTsm87116vVbb4N51-FJeO_Xz4MNK6dA5s0FVNFLIpWUAaATyUpbSWGOLupayrESZuq76rm3wHzuMnVr7XWjTfJVWcuDplImCnjLBxxjQ_v3KqNqLUL0ItRehfkWk1GWfcoj4L5HEVKyGb6rmhZY |
CODEN | IJSTHZ |
Cites_doi | 10.1109/LGRS.2019.2918719 10.1109/TIM.2021.3056750 10.1007/s11431-021-1989-9 10.1109/TGRS.2004.842481 10.1109/TGRS.2024.3362471 10.1109/TGRS.2016.2584107 10.1109/TGRS.2016.2530807 10.1109/CVPR42600.2020.00297 10.1109/LGRS.2010.2047711 10.1109/JSTARS.2021.3123371 10.1155/2015/258619 10.1109/JSTARS.2021.3074469 10.3390/rs12162659 10.1109/TGRS.2016.2636241 10.1109/LGRS.2024.3379232 10.1016/j.isprsjprs.2022.07.001 10.1109/IGARSS.2015.7326945 10.1109/JSTARS.2022.3223423 10.1109/LGRS.2018.2869563 10.1109/TGRS.2020.3037361 10.1109/TGRS.2021.3123423 10.1109/TGRS.2019.2907932 10.1007/978-3-030-58523-5_46 10.1109/TIP.2022.3144017 10.1109/TPAMI.2022.3182052 10.1109/TGRS.2020.3031928 10.1109/LGRS.2021.3108883 10.1109/JSTARS.2022.3177235 10.1109/TIP.2021.3098246 10.1109/TGRS.2019.2949180 10.1109/JSTARS.2019.2892975 10.1109/JSTARS.2021.3112158 10.1609/aaai.v33i01.33013558 10.1109/TGRS.2008.2005729 10.1109/TGRS.2023.3268944 10.1109/TGRS.2023.3265879 10.1109/LGRS.2019.2915315 10.1109/TGRS.2014.2360672 10.1109/ICIP.2017.8297014 10.1109/JSTARS.2023.3328389 |
ContentType | Journal Article |
Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
DBID | 97E ESBDL RIA RIE AAYXX CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M DOA |
DOI | 10.1109/JSTARS.2024.3403863 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Electronic Library Online CrossRef 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 Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Advanced Technologies Database with Aerospace Directory of Open Access Journals |
DatabaseTitle | CrossRef Aerospace Database Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Technology Research Database ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Water Resources Abstracts Environmental Sciences and Pollution Management |
DatabaseTitleList | Aerospace Database |
Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: RIE name: IEEE Electronic Library Online url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Geology |
EISSN | 2151-1535 |
EndPage | 10673 |
ExternalDocumentID | oai_doaj_org_article_5b949df133ec4e26969cfcf588996746 10_1109_JSTARS_2024_3403863 10535718 |
Genre | orig-research |
GrantInformation_xml | – fundername: Major Science and Technology Projects in Yunnan Province grantid: 202202AD080013; 202302AG050009 funderid: 10.13039/501100018531 – fundername: Xingdian Talent Support Plan of Yunnan Province grantid: KKRD202203068 – fundername: Yunnan Fundamental Research grantid: 202101BE070001-008; 202401AW070019; 202301AV070003 – fundername: National Natural Science Foundation of China grantid: 62201237 funderid: 10.13039/501100001809 |
GroupedDBID | 0R~ 29I 4.4 5GY 5VS 6IK 97E AAFWJ AAJGR ABVLG ACIWK AENEX AETIX AFPKN AFRAH ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ DU5 EBS EJD ESBDL GROUPED_DOAJ HZ~ IFIPE IPLJI JAVBF M43 O9- OCL OK1 RIA RIE RIG RNS AAYXX AGSQL CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M |
ID | FETCH-LOGICAL-c359t-ff307487b6ba919f723a73006e1bad2b9ea013e93c1a487083aef84f1f63cd033 |
IEDL.DBID | DOA |
ISSN | 1939-1404 |
IngestDate | Tue Oct 22 14:49:03 EDT 2024 Thu Oct 10 15:24:20 EDT 2024 Fri Dec 06 09:50:04 EST 2024 Wed Jul 03 05:40:34 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
License | https://creativecommons.org/licenses/by-nc-nd/4.0 |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c359t-ff307487b6ba919f723a73006e1bad2b9ea013e93c1a487083aef84f1f63cd033 |
ORCID | 0009-0007-3920-814X 0000-0002-6103-7769 0000-0003-1273-7950 0000-0001-5820-5357 |
OpenAccessLink | https://doaj.org/article/5b949df133ec4e26969cfcf588996746 |
PQID | 3072323079 |
PQPubID | 75722 |
PageCount | 14 |
ParticipantIDs | proquest_journals_3072323079 ieee_primary_10535718 doaj_primary_oai_doaj_org_article_5b949df133ec4e26969cfcf588996746 crossref_primary_10_1109_JSTARS_2024_3403863 |
PublicationCentury | 2000 |
PublicationDate | 20240000 2024-00-00 20240101 2024-01-01 |
PublicationDateYYYYMMDD | 2024-01-01 |
PublicationDate_xml | – year: 2024 text: 20240000 |
PublicationDecade | 2020 |
PublicationPlace | Piscataway |
PublicationPlace_xml | – name: Piscataway |
PublicationTitle | IEEE journal of selected topics in applied earth observations and remote sensing |
PublicationTitleAbbrev | JSTARS |
PublicationYear | 2024 |
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 | ref13 ref35 ref12 ref34 ref15 ref37 ref14 ref36 ref31 ref30 ref11 ref33 ref10 ref32 ref2 ref1 ref17 ref39 ref16 ref19 ref18 ref23 ref26 ref25 ref20 ref42 ref41 ref22 ref21 Kipf (ref24) 2017 Liu (ref38) 2020; 48 ref28 ref27 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5 ref40 |
References_xml | – ident: ref22 doi: 10.1109/LGRS.2019.2918719 – ident: ref23 doi: 10.1109/TIM.2021.3056750 – ident: ref3 doi: 10.1007/s11431-021-1989-9 – ident: ref11 doi: 10.1109/TGRS.2004.842481 – ident: ref31 doi: 10.1109/TGRS.2024.3362471 – ident: ref40 doi: 10.1109/TGRS.2016.2584107 – ident: ref8 doi: 10.1109/TGRS.2016.2530807 – ident: ref37 doi: 10.1109/CVPR42600.2020.00297 – ident: ref9 doi: 10.1109/LGRS.2010.2047711 – ident: ref14 doi: 10.1109/JSTARS.2021.3123371 – ident: ref13 doi: 10.1155/2015/258619 – start-page: 1 volume-title: Proc. Int. Conf. Learn. Represent. year: 2017 ident: ref24 article-title: Semi-supervised classification with graph convolutional networks contributor: fullname: Kipf – ident: ref16 doi: 10.1109/JSTARS.2021.3074469 – ident: ref2 doi: 10.3390/rs12162659 – ident: ref18 doi: 10.1109/TGRS.2016.2636241 – ident: ref36 doi: 10.1109/LGRS.2024.3379232 – ident: ref34 doi: 10.1016/j.isprsjprs.2022.07.001 – ident: ref19 doi: 10.1109/IGARSS.2015.7326945 – ident: ref21 doi: 10.1109/JSTARS.2022.3223423 – ident: ref25 doi: 10.1109/LGRS.2018.2869563 – ident: ref27 doi: 10.1109/TGRS.2020.3037361 – ident: ref35 doi: 10.1109/TGRS.2021.3123423 – ident: ref12 doi: 10.1109/TGRS.2019.2907932 – ident: ref39 doi: 10.1007/978-3-030-58523-5_46 – ident: ref28 doi: 10.1109/TIP.2022.3144017 – ident: ref33 doi: 10.1109/TPAMI.2022.3182052 – ident: ref10 doi: 10.1109/TGRS.2020.3031928 – ident: ref17 doi: 10.1109/LGRS.2021.3108883 – ident: ref30 doi: 10.1109/JSTARS.2022.3177235 – volume: 48 start-page: 751 issue: 4 year: 2020 ident: ref38 article-title: SSCDenseNet: A spectral-spatial convolutional dense network for hyperspectral image classification publication-title: Acta Electronica Sinica contributor: fullname: Liu – ident: ref41 doi: 10.1109/TIP.2021.3098246 – ident: ref26 doi: 10.1109/TGRS.2019.2949180 – ident: ref42 doi: 10.1109/JSTARS.2019.2892975 – ident: ref1 doi: 10.1109/JSTARS.2021.3112158 – ident: ref32 doi: 10.1609/aaai.v33i01.33013558 – ident: ref6 doi: 10.1109/TGRS.2008.2005729 – ident: ref4 doi: 10.1109/TGRS.2023.3268944 – ident: ref29 doi: 10.1109/TGRS.2023.3265879 – ident: ref5 doi: 10.1109/LGRS.2019.2915315 – ident: ref7 doi: 10.1109/TGRS.2014.2360672 – ident: ref20 doi: 10.1109/ICIP.2017.8297014 – ident: ref15 doi: 10.1109/JSTARS.2023.3328389 |
SSID | ssj0062793 |
Score | 2.4280183 |
Snippet | Recently, hybrid networks, combining graph convolutional networks (GCNs) and convolutional neural networks into a unified framework, have garnered significant... |
SourceID | doaj proquest crossref ieee |
SourceType | Open Website Aggregation Database Publisher |
StartPage | 10660 |
SubjectTerms | Advantageous feature differential enhancement fusion (AFDEF) Apexes Artificial neural networks Classification Convolution Convolutional neural networks convolutional neural networks (CNNs) Correlation Data mining Feature extraction hypergraph convolutional network hyperspectral image (HSI) classification Hyperspectral imaging Image classification Information retrieval Neural networks Training Transformers |
SummonAdditionalLinks | – databaseName: IEEE Electronic Library Online dbid: RIE link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB7RSkhcKI8itrTIB45kSfzKmltpWRYOeyhU6s2ynfEF2KJtcqC_nvFjEQ8hcYmiyFESf57xzGTmG4AXA-ndYeFdo5WUjUT0jeOhbzqytckjcoMPOdtirVeX8sOVuqrF6rkWBhFz8hnO02n-lz9chymFykjClVCkTPdgrze6FGvt1K7mfWbYJYPENIkzplIMda15RWv89OIjOYNczoVsxUKL37ahzNZf26v8pZPzRrM8gPXuFUt-yef5NPp5uP2DvfG_v-EB3K8mJzsta-Qh3MHNI7j7Lrf0_f4YLs6XfI3ja3Zee6WQzH9hyTSctsiWUwqnsXXJFmdk4rIVua6lQnNLA99_JY3Ecm_NlHWUgT6Ey-XbT2erpnZaaIJQZmxiJFEn18Vr70xnYs-FS0T2GjvvBu4NuhQuNSJ0joaR2eYwLmTsohZhaIV4Avub6w0-BRb7aAK5cS22KFU7eCmV7yJ3LfaeiziDl7uJt98KoYbNjkhrbMHJJpxsxWkGbxI4P4cmNux8gSbVVuGyyhtphkjuNgaJXBttQgxRLciZ1L3UMzhMQPzyvILBDI53WNsqujeWZoKsTDqao3_c9gzupVcsgZhj2B-3E56QaTL653lJ_gBnct7r priority: 102 providerName: IEEE |
Title | DF2Net: Differential Feature Fusion Network for Hyperspectral Image Classification |
URI | https://ieeexplore.ieee.org/document/10535718 https://www.proquest.com/docview/3072323079 https://doaj.org/article/5b949df133ec4e26969cfcf588996746 |
Volume | 17 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07T8MwELYQEhIL4ikKpfLASKjjV2q28iiFoQNQic2yHXuCgko68O852ykqYmBhyRBZcnLnO3-fdf4OodMa8m49sKaQgvOCe28LQ11VlIC1gRGZ2rpUbTGR4ym_fxbPK62-Yk1YlgfOhusLq7iqA1Ap77inUknlggtiAERBVjyLbRO6JFM5B0taJbldQCeqiAIyrd5QSVQfFvzw4RGYIeXnjBM2kOzHnpSk-9teK78SdNp1Rttoq4WLeJg_cwet-dku2rhN7Xg_99DD9YhOfHOBr9s-JxCvLzjCusXc49EiHoXhSa70xgBP8RhoZ75dOYeBd6-QTXDqixkrhpKT9tF0dPN0NS7aLgmFY0I1RQgQpkA7rLRGlSpUlJkoQi99aU1NrfImHnUq5koDwwByGR8GPJRBMlcTxg7Q-uxt5g8RDlVQDigY8cRzQWrLubBloIb4ylIWOuhsaSf9nsUwdCIRROlsVh3NqluzdtBltOX30KhknV6Af3XrX_2XfztoP3piZT7BBGyoHdRduka3YfehwRKAEOGpjv5j7mO0Gf8nn7h00XozX_gTwCCN7aXl1kvXBb8Aqq7VjA |
link.rule.ids | 314,780,784,796,864,2102,4024,27923,27924,27925,54758 |
linkProvider | Directory of Open Access Journals |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB5BEaIXnkVdKOADR7IkfmXNrVCWLZQcSiv1ZtnO-AJs0ZIc4NczdryIh5C4RFHkKInHM_6-yTwAnvZkd_uFd5VWUlYS0VeOh7ZqCGsTI3K9DznaotOrc_n2Ql2UZPWcC4OIOfgM5-k0_8vvL8OYXGWk4UooMqZX4ZqSBHSndK2t4dW8zTV2CZKYKlWNKUWGmto8p1V-ePqB6CCXcyFrsdDit40o1-svDVb-ssp5q1negm77klOEycf5OPh5-P5H_cb__orbcLOATnY4rZI7cAXXd-H6m9zU99s9OD1a8g6HF-yodEshrf_EEjgcN8iWY3KosW6KF2cEctmKyOuUo7mhgcefySax3F0zxR1lUe_B-fL12atVVXotVEEoM1QxkrITefHaO9OY2HLhUil7jY13PfcGXXKYGhEaR8MIuDmMCxmbqEXoayHuw876co37wGIbTSAiV2ONUtW9l1L5JnJXY-u5iDN4tp14-2UqqWEzFamNneRkk5xskdMMXibh_Bya6mHnCzSptqiXVd5I00ci3Bgkcm20CTFEtSA6qVupZ7CXBPHL8yYZzOBgK2tblPerpZkgnElH8-Aftz2BG6uz9yf25Lh79xB20-tObpkD2Bk2Iz4ioDL4x3l5_gBNiuI- |
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=DF2Net%3A+Differential+Feature+Fusion+Network+for+Hyperspectral+Image+Classification&rft.jtitle=IEEE+journal+of+selected+topics+in+applied+earth+observations+and+remote+sensing&rft.au=Wang%2C+Qingwang&rft.au=Huang%2C+Jiangbo&rft.au=Meng%2C+Yuanqin&rft.au=Shen%2C+Tao&rft.date=2024&rft.issn=1939-1404&rft.eissn=2151-1535&rft.volume=17&rft.spage=10660&rft.epage=10673&rft_id=info:doi/10.1109%2FJSTARS.2024.3403863&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_JSTARS_2024_3403863 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1939-1404&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1939-1404&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1939-1404&client=summon |