DEMAE: Diffusion-Enhanced Masked Autoencoder for Hyperspectral Image Classification With Few Labeled Samples

Unlike other deep learning (DL) models, Transformer has the ability to extract long-range dependency features from hyperspectral image (HSI) data. Masked autoencoder (MAE), which is based on Transformer architecture, employs a "mask-reconstruction" strategy for training, allowing the model...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	IEEE transactions on geoscience and remote sensing Ročník 62; s. 1 - 16
Hlavní autori:	Li, Ziyu, Xue, Zhaohui, Jia, Mingming, Nie, Xiangyu, Wu, Hao, Zhang, Mengxue, Su, Hongjun
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	New York IEEE 2024 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:	Classification Coders Data models Deep learning Diffusion Diffusion barriers Diffusion models Feature extraction few labeled samples hyperspectral image (HSI) classification Hyperspectral imaging Image classification Image enhancement Image reconstruction Machine learning masked autoencoder (MAE) Signal classification Signal to noise ratio Simultaneous discrimination learning Source code Spatial discrimination learning State-of-the-art reviews Task analysis Training Transformer Transformers
ISSN:	0196-2892, 1558-0644
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	Unlike other deep learning (DL) models, Transformer has the ability to extract long-range dependency features from hyperspectral image (HSI) data. Masked autoencoder (MAE), which is based on Transformer architecture, employs a "mask-reconstruction" strategy for training, allowing the model to be effective for downstream tasks. However, existing MAE-based methods only apply spectral or spatial masking to HSI and reconstruct them for feature learning, which is too simplistic and insufficient for the model to learn robust features. Additionally, the issue of lacking labeled samples in HSI and the primary objective of MAE to reduce the reliance on labeled samples are often overlooked. To address these issues, we are inspired by diffusion-based representation learning and propose diffusion-enhanced MAE (DEMAE) for HSI classification with few labeled samples. First, an asymmetric encoder-decoder framework is constructed as the backbone by stacking both conditional and standard Transformer blocks. Second, we devise an auxiliary task aimed at simultaneous denoising and reconstruction, facilitating heuristic feature learning from HSI data. Third, the encoder of DEMAE is isolated for training with few labeled samples. Finally, the encoder is used for classification, and a novel signal-to-noise ratio enhanced (SNR-Enhanced) loss function is introduced to regularize the model training process. The performance of DEMAE is evaluated on four benchmark datasets, demonstrating its superiority in classification accuracy and mapping capabilities on unlabeled areas compared to existing state-of-the-art methods with few labeled samples. The source code will be available online at https://github.com/ZhaohuiXue/DEMAE .
AbstractList	Unlike other deep learning (DL) models, Transformer has the ability to extract long-range dependency features from hyperspectral image (HSI) data. Masked autoencoder (MAE), which is based on Transformer architecture, employs a “mask-reconstruction” strategy for training, allowing the model to be effective for downstream tasks. However, existing MAE-based methods only apply spectral or spatial masking to HSI and reconstruct them for feature learning, which is too simplistic and insufficient for the model to learn robust features. Additionally, the issue of lacking labeled samples in HSI and the primary objective of MAE to reduce the reliance on labeled samples are often overlooked. To address these issues, we are inspired by diffusion-based representation learning and propose diffusion-enhanced MAE (DEMAE) for HSI classification with few labeled samples. First, an asymmetric encoder–decoder framework is constructed as the backbone by stacking both conditional and standard Transformer blocks. Second, we devise an auxiliary task aimed at simultaneous denoising and reconstruction, facilitating heuristic feature learning from HSI data. Third, the encoder of DEMAE is isolated for training with few labeled samples. Finally, the encoder is used for classification, and a novel signal-to-noise ratio enhanced (SNR-Enhanced) loss function is introduced to regularize the model training process. The performance of DEMAE is evaluated on four benchmark datasets, demonstrating its superiority in classification accuracy and mapping capabilities on unlabeled areas compared to existing state-of-the-art methods with few labeled samples. The source code will be available online at https://github.com/ZhaohuiXue/DEMAE .
Author	Zhang, Mengxue Wu, Hao Su, Hongjun Li, Ziyu Xue, Zhaohui Nie, Xiangyu Jia, Mingming
Author_xml	– sequence: 1 givenname: Ziyu orcidid: 0000-0002-6520-2530 surname: Li fullname: Li, Ziyu organization: College of Geography and Remote Sensing, Hohai University, Nanjing, China – sequence: 2 givenname: Zhaohui orcidid: 0000-0001-6253-2967 surname: Xue fullname: Xue, Zhaohui organization: College of Geography and Remote Sensing, Hohai University, Nanjing, China – sequence: 3 givenname: Mingming orcidid: 0000-0002-4548-899X surname: Jia fullname: Jia, Mingming email: jiamingming@iga.ac.cn organization: Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun, China – sequence: 4 givenname: Xiangyu orcidid: 0009-0001-5095-6401 surname: Nie fullname: Nie, Xiangyu organization: College of Geography and Remote Sensing, Hohai University, Nanjing, China – sequence: 5 givenname: Hao orcidid: 0000-0001-7340-6536 surname: Wu fullname: Wu, Hao organization: College of Geography and Remote Sensing, Hohai University, Nanjing, China – sequence: 6 givenname: Mengxue orcidid: 0000-0002-8587-4334 surname: Zhang fullname: Zhang, Mengxue organization: Image and Signal Processing (ISP) Group, University of València, València, Spain – sequence: 7 givenname: Hongjun orcidid: 0000-0002-8991-8568 surname: Su fullname: Su, Hongjun organization: College of Geography and Remote Sensing, Hohai University, Nanjing, China
BookMark	eNp9kEtLAzEUhYMoWB8_QHARcD01mSTTxF2ptQoVwQcuh9vkRqPTmTGZIv57p7YLceHqbL7vXO45ILt1UyMhJ5wNOWfm_HF2_zDMWS6HQkplDN8hA66Uzlgh5S4ZMG6KLNcm3ycHKb0xxqXiowGpLqe34-kFvQzer1Jo6mxav0Jt0dFbSO99jFddg7VtHEbqm0ivv1qMqUXbRajozRJekE4qSCn4YKHrK-hz6F7pFX7SOSyw6jseYNlWmI7Inocq4fE2D8nT1fRxcp3N72Y3k_E8s7mRXQZKFIVnymkUKLnWjoHiduEsLoRbeF7AiIGDgmkBIyMFFE5L1WO55t45cUjONr1tbD5WmLryrVnFuj9ZCma01jmXsqdGG8rGJqWIvrSh-3mg_yxUJWfletpyPW25nrbcTtub_I_ZxrCE-PWvc7pxAiL-4gthpBLiG_jwh0Q
CODEN	IGRSD2
CitedBy_id	crossref_primary_10_1016_j_eswa_2024_124277 crossref_primary_10_1007_s11227_025_07607_x crossref_primary_10_1109_TGRS_2024_3509720 crossref_primary_10_1109_TGRS_2025_3584804 crossref_primary_10_1109_TGRS_2024_3495525 crossref_primary_10_1109_TGRS_2025_3586287
Cites_doi	10.1109/TGRS.2024.3365719 10.1109/TGRS.2024.3407967 10.1016/j.isprsjprs.2023.05.025 10.1109/TGRS.2023.3310023 10.1609/aaai.v38i6.28392 10.1109/ICCV51070.2023.01110 10.1109/JSTARS.2022.3174135 10.3390/rs13030498 10.1016/j.rse.2023.113856 10.1109/TGRS.2023.3324497 10.1109/TGRS.2023.3315678 10.1109/TGRS.2022.3181501 10.3390/rs13112216 10.1109/TGRS.2022.3144158 10.1109/TGRS.2021.3130716 10.3390/rs12060923 10.1109/ICCV51070.2023.01448 10.1109/TGRS.2023.3279834 10.1109/79.974718 10.1109/TGRS.2024.3408475 10.1109/JSTARS.2023.3294623 10.1109/TGRS.2018.2827407 10.1109/TGRS.2023.3344782 10.1109/TGRS.2019.2908756 10.1109/TGRS.2024.3407206 10.1109/JSTARS.2020.3004973 10.1109/TGRS.2021.3139099 10.1109/TGRS.2022.3207933 10.1109/TGRS.2023.3264235 10.1117/1.JRS.15.031501 10.1109/TGRS.2023.3310489 10.1109/TGRS.2022.3217892 10.1109/TGRS.2021.3057066 10.1109/MGRS.2019.2902525 10.1016/j.neucom.2023.03.025 10.1109/TPAMI.2024.3362475 10.1016/j.neucom.2021.03.091 10.1109/TGRS.2017.2755542 10.1109/TGRS.2021.3057768 10.59717/j.xinn-geo.2024.100055 10.1109/TIP.2023.3322046
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 RIA RIE AAYXX CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M
DOI	10.1109/TGRS.2024.3445991
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library (IEL) (UW System Shared) 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
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: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering Physics
EISSN	1558-0644
EndPage	16
ExternalDocumentID	10_1109_TGRS_2024_3445991 10639453
Genre	orig-research
GrantInformation_xml	– fundername: National Natural Science Foundation of China grantid: 42271324 funderid: 10.13039/501100001809 – fundername: Postgraduate Research and Practice Innovation Program of Jiangsu Province grantid: KYCX24_0885 – fundername: Project on Excellent Post-Graduate Dissertation of Hohai University funderid: 10.13039/501100015973 – fundername: Natural Science Foundation of Jiangsu Province grantid: BK20221506 funderid: 10.13039/501100004608 – fundername: Youth Innovation Promotion Association of the Chinese Academy of Sciences; Youth Innovation Promotion Association of Chinese Academy of Sciences grantid: 2021227 funderid: 10.13039/501100004739
GroupedDBID	-~X 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT AENEX AETIX AFRAH AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD F5P HZ~ H~9 IBMZZ ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 O9- OCL P2P RIA RIE RNS RXW TAE TN5 VH1 Y6R AAYXX CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M
ID	FETCH-LOGICAL-c294t-a5366f05d8e3e4188d0a51cbdceb3dbf16a70ada6083a7943a6d8458d0281fdd3
IEDL.DBID	RIE
ISICitedReferencesCount	6
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001303543600012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0196-2892
IngestDate	Mon Jun 30 10:07:13 EDT 2025 Sat Nov 29 03:32:40 EST 2025 Tue Nov 18 21:34:59 EST 2025 Wed Aug 27 02:03:43 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c294t-a5366f05d8e3e4188d0a51cbdceb3dbf16a70ada6083a7943a6d8458d0281fdd3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-8587-4334 0009-0001-5095-6401 0000-0002-4548-899X 0000-0001-6253-2967 0000-0002-8991-8568 0000-0001-7340-6536 0000-0002-6520-2530
PQID	3098882144
PQPubID	85465
PageCount	16
ParticipantIDs	ieee_primary_10639453 crossref_primary_10_1109_TGRS_2024_3445991 proquest_journals_3098882144 crossref_citationtrail_10_1109_TGRS_2024_3445991
PublicationCentury	2000
PublicationDate	20240000 2024-00-00 20240101
PublicationDateYYYYMMDD	2024-01-01
PublicationDate_xml	– year: 2024 text: 20240000
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on geoscience and remote sensing
PublicationTitleAbbrev	TGRS
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 ref12 ref15 ref14 ref53 ref52 ref11 Ho (ref27) 2020 ref10 ref54 Preechakul (ref28) 2021 ref17 ref19 ref18 Asiedu (ref37) 2022 ref51 ref50 ref46 ref45 ref48 ref47 ref42 Baranchuk (ref36) 2021 ref41 ref44 Gui (ref16) 2023 Zhang (ref43) 2023 ref49 ref8 ref7 ref9 ref4 ref3 ref6 ref5 Hudson (ref30) 2023 ref40 Bao (ref34) 2022 Peebles (ref33) 2022 ref2 ref1 ref39 ref38 Tao Hu (ref31) 2023 ref24 ref23 ref26 Chen (ref32) 2024 ref25 ref20 ref22 He (ref21) 2021 Bao (ref35) 2023 ref29
References_xml	– ident: ref51 doi: 10.1109/TGRS.2024.3365719 – year: 2023 ident: ref16 article-title: A survey on self-supervised learning: Algorithms, applications, and future trends publication-title: arXiv:2301.05712 – year: 2023 ident: ref31 article-title: Guided diffusion from self-supervised diffusion features publication-title: arXiv:2312.08825 – ident: ref52 doi: 10.1109/TGRS.2024.3407967 – ident: ref5 doi: 10.1016/j.isprsjprs.2023.05.025 – year: 2023 ident: ref30 article-title: SODA: Bottleneck diffusion models for representation learning publication-title: arXiv:2311.17901 – ident: ref44 doi: 10.1109/TGRS.2023.3310023 – ident: ref38 doi: 10.1609/aaai.v38i6.28392 – ident: ref39 doi: 10.1109/ICCV51070.2023.01110 – ident: ref14 doi: 10.1109/JSTARS.2022.3174135 – ident: ref10 doi: 10.3390/rs13030498 – ident: ref54 doi: 10.1016/j.rse.2023.113856 – ident: ref18 doi: 10.1109/TGRS.2023.3324497 – ident: ref24 doi: 10.1109/TGRS.2023.3315678 – ident: ref48 doi: 10.1109/TGRS.2022.3181501 – ident: ref11 doi: 10.3390/rs13112216 – ident: ref13 doi: 10.1109/TGRS.2022.3144158 – ident: ref12 doi: 10.1109/TGRS.2021.3130716 – ident: ref47 doi: 10.3390/rs12060923 – ident: ref29 doi: 10.1109/ICCV51070.2023.01448 – ident: ref50 doi: 10.1109/TGRS.2023.3279834 – year: 2022 ident: ref33 article-title: Scalable diffusion models with transformers publication-title: arXiv:2212.09748 – year: 2021 ident: ref28 article-title: Diffusion autoencoders: Toward a meaningful and decodable representation publication-title: arXiv:2111.15640 – ident: ref1 doi: 10.1109/79.974718 – year: 2021 ident: ref36 article-title: Label-efficient semantic segmentation with diffusion models publication-title: arXiv:2112.03126 – ident: ref42 doi: 10.1109/TGRS.2024.3408475 – ident: ref41 doi: 10.1109/JSTARS.2023.3294623 – ident: ref7 doi: 10.1109/TGRS.2018.2827407 – ident: ref25 doi: 10.1109/TGRS.2023.3344782 – year: 2023 ident: ref35 article-title: One transformer fits all distributions in multi-modal diffusion at scale publication-title: arXiv:2303.06555 – year: 2022 ident: ref37 article-title: Decoder denoising pretraining for semantic segmentation publication-title: arXiv:2205.11423 – ident: ref8 doi: 10.1109/TGRS.2019.2908756 – ident: ref45 doi: 10.1109/TGRS.2024.3407206 – ident: ref46 doi: 10.1109/JSTARS.2020.3004973 – ident: ref19 doi: 10.1109/TGRS.2021.3139099 – ident: ref15 doi: 10.1109/TGRS.2022.3207933 – ident: ref23 doi: 10.1109/TGRS.2023.3264235 – year: 2024 ident: ref32 article-title: Deconstructing denoising diffusion models for self-supervised learning publication-title: arXiv:2401.14404 – ident: ref3 doi: 10.1117/1.JRS.15.031501 – year: 2020 ident: ref27 article-title: Denoising diffusion probabilistic models publication-title: arXiv:2006.11239 – ident: ref17 doi: 10.1109/TGRS.2023.3310489 – ident: ref22 doi: 10.1109/TGRS.2022.3217892 – ident: ref49 doi: 10.1109/TGRS.2021.3057066 – ident: ref4 doi: 10.1109/MGRS.2019.2902525 – year: 2023 ident: ref43 article-title: DiffUCD: Unsupervised hyperspectral image change detection with semantic correlation diffusion model publication-title: arXiv:2305.12410 – ident: ref2 doi: 10.1016/j.neucom.2023.03.025 – year: 2021 ident: ref21 article-title: Masked autoencoders are scalable vision learners publication-title: arXiv:2111.06377 – ident: ref26 doi: 10.1109/TPAMI.2024.3362475 – ident: ref9 doi: 10.1016/j.neucom.2021.03.091 – year: 2022 ident: ref34 article-title: All are worth words: A ViT backbone for diffusion models publication-title: arXiv:2209.12152 – ident: ref6 doi: 10.1109/TGRS.2017.2755542 – ident: ref20 doi: 10.1109/TGRS.2021.3057768 – ident: ref53 doi: 10.59717/j.xinn-geo.2024.100055 – ident: ref40 doi: 10.1109/TIP.2023.3322046
SSID	ssj0014517
Score	2.4687054
Snippet	Unlike other deep learning (DL) models, Transformer has the ability to extract long-range dependency features from hyperspectral image (HSI) data. Masked...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1
SubjectTerms	Classification Coders Data models Deep learning Diffusion Diffusion barriers Diffusion models Feature extraction few labeled samples hyperspectral image (HSI) classification Hyperspectral imaging Image classification Image enhancement Image reconstruction Machine learning masked autoencoder (MAE) Signal classification Signal to noise ratio Simultaneous discrimination learning Source code Spatial discrimination learning State-of-the-art reviews Task analysis Training Transformer Transformers
Title	DEMAE: Diffusion-Enhanced Masked Autoencoder for Hyperspectral Image Classification With Few Labeled Samples
URI	https://ieeexplore.ieee.org/document/10639453 https://www.proquest.com/docview/3098882144
Volume	62
WOSCitedRecordID	wos001303543600012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 1558-0644 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014517 issn: 0196-2892 databaseCode: RIE dateStart: 19800101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8QwEB5UFPTgW1xf5OBJqKZtmjbeFt1VQUV8oLeS5sEu6q5su_r3naRVFFHw1B4mbcnXycxkMt8A7BYGvW6mZJAqpgKmGQsKw1HdrStjFJLzwpO4nqeXl9nDg7hqitV9LYwxxh8-M_vu1ufy9VCN3VYZajjaU5bEkzCZprwu1vpMGbAkbGqjeYBRRNSkMEMqDm5Prm8wFIzYfsxYIkT4zQj5rio_lmJvX7oL__yyRZhvHEnSrpFfggkzWIa5L_SCyzDjj3eqcgWecLrbnUNy3Ld27PbHgs6g53P_5EKWj3hpj6uh47TUZkTQjyWnGJ_WZZgjfMvZMy47xDfQdEeLPJrkvl_1SNe8kXNZoPHS5EY6quFyFe66nduj06DpsxCoSLAqkEnMuaWJzkxsWJhlmsokVIVWGGnrwoZcplRqydFdk45QTnKdsQTFoiy0WsdrMDUYDsw6EE21CamxTEjLUsmLDB2cOE1sZLQVhWgB_Zj4XDUk5K4XxlPugxEqcodV7rDKG6xasPc55KVm4PhLeNWB80WwxqUFWx_w5o2SlnlMBcb_jjNu45dhmzDrnl5vuWzBVDUam22YVq9Vvxzt-P_vHSzW2BA
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Nb9QwEB1BAZUe-GiL2FLAB05IaZ1k7MTcVrDLVmxXiC6it8jxh7qi7KJNFv4-YyetihBInJKDLUd-sWeex_MG4FXtyOtGo5PCoEnQIia1k7TcfUhjVFrKOoq4TovZrDw_Vx_7ZPWYC-Oci5fP3FF4jbF8uzKbcFRGK5zsKYr8NtwRiBnv0rWugwYo0j47WibEI7I-iJlydTx__-mMyGCGRzmiUCr9zQzFuip_bMbRwowf_ue3PYIHvSvJhh32j-GWW-7Czg2BwV24Fy94mmYPLmnCh6M37N3C-004IUtGy4sY_WenuvlKj-GmXQVVS-vWjDxZNiGG2iVirmmUk2-08bBYQjNcLop4si-L9oKN3U821TWZL8vOdBAbbvbh83g0fztJ-koLickUtokWuZSeC1u63GFalpZrkZraGuLatvap1AXXVkty2HSQlNPSliioWVam3tr8CWwtV0v3FJjl1qXceVTaY6FlXZKLkxfCZ856VasB8KuJr0wvQx6qYVxWkY5wVQWsqoBV1WM1gNfXXb53Ghz_arwfwLnRsMNlAIdX8Fb9Mm2qnKuSKAaRyoO_dHsJ25P56bSansw-PIP7YaTuAOYQttr1xj2Hu-ZHu2jWL-K_-AuRfdtX
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=DEMAE%3A+Diffusion-Enhanced+Masked+Autoencoder+for+Hyperspectral+Image+Classification+With+Few+Labeled+Samples&rft.jtitle=IEEE+transactions+on+geoscience+and+remote+sensing&rft.au=Li%2C+Ziyu&rft.au=Xue%2C+Zhaohui&rft.au=Jia%2C+Mingming&rft.au=Nie%2C+Xiangyu&rft.date=2024&rft.issn=0196-2892&rft.eissn=1558-0644&rft.volume=62&rft.spage=1&rft.epage=16&rft_id=info:doi/10.1109%2FTGRS.2024.3445991&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TGRS_2024_3445991
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0196-2892&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0196-2892&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0196-2892&client=summon