Self-Supervised Masked Graph Autoencoder for Hyperspectral Anomaly Detection
Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder(AE)-based methods are widely used due to their excellent image reconstruction capability. However, traditional grid-based image representation methods struggle to capture long-range d...
Uloženo v:
| Vydáno v: | IEEE transactions on image processing Ročník 34; s. 6714 - 6729 |
|---|---|
| Hlavní autoři: | , , , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
United States
IEEE
01.01.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Témata: | |
| ISSN: | 1057-7149, 1941-0042, 1941-0042 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder(AE)-based methods are widely used due to their excellent image reconstruction capability. However, traditional grid-based image representation methods struggle to capture long-range dependencies and model non-Euclidean structures. To address these issues, this paper proposes a self-supervised Masked Graph AutoEncoder (MGAE) for hyperspectral anomaly detection. MGAE utilizes a Graph Attention Network (GAT) autoencoder to reconstruct the background of hyperspectral images and identifies anomalies by comparing the reconstructed features with the original features. Specifically, we constructs a topological graph structure of the hyperspectral image, which is then input into the GAT autoencoder for reconstruction, leveraging the multi-head attention mechanism to learn spatial and spectral features. To prevent the decoder from learning trivial solutions, we introduce a re-masking strategy that randomly masks both the input features and hidden representations during training, forcing the model to learn and reconstruct features under limited information, thereby improving detection performance. Additionally, the proposed loss function with graph Laplacian regularization (Twice Loss) minimizes variations in feature representations, leading to more consistent background reconstruction. Experimental results on several real-world hyperspectral datasets demonstrate that MGAE outperforms existing methods. |
|---|---|
| AbstractList | Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder(AE)-based methods are widely used due to their excellent image reconstruction capability. However, traditional grid-based image representation methods struggle to capture long-range dependencies and model non-Euclidean structures. To address these issues, this paper proposes a self-supervised Masked Graph AutoEncoder (MGAE) for hyperspectral anomaly detection. MGAE utilizes a Graph Attention Network (GAT) autoencoder to reconstruct the background of hyperspectral images and identifies anomalies by comparing the reconstructed features with the original features. Specifically, we constructs a topological graph structure of the hyperspectral image, which is then input into the GAT autoencoder for reconstruction, leveraging the multi-head attention mechanism to learn spatial and spectral features. To prevent the decoder from learning trivial solutions, we introduce a re-masking strategy that randomly masks both the input features and hidden representations during training, forcing the model to learn and reconstruct features under limited information, thereby improving detection performance. Additionally, the proposed loss function with graph Laplacian regularization (Twice Loss) minimizes variations in feature representations, leading to more consistent background reconstruction. Experimental results on several real-world hyperspectral datasets demonstrate that MGAE outperforms existing methods. Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder (AE)-based methods are widely used due to their excellent image reconstruction capability. However, traditional grid-based image representation methods struggle to capture long-range dependencies and model non-Euclidean structures. To address these issues, this paper proposes a self-supervised Masked Graph AutoEncoder (MGAE) for hyperspectral anomaly detection. MGAE utilizes a Graph Attention Network (GAT) autoencoder to reconstruct the background of hyperspectral images and identifies anomalies by comparing the reconstructed features with the original features. Specifically, we constructs a topological graph structure of the hyperspectral image, which is then input into the GAT autoencoder for reconstruction, leveraging the multi-head attention mechanism to learn spatial and spectral features. To prevent the decoder from learning trivial solutions, we introduce a re-masking strategy that randomly masks both the input features and hidden representations during training, forcing the model to learn and reconstruct features under limited information, thereby improving detection performance. Additionally, the proposed loss function with graph Laplacian regularization (Twice Loss) minimizes variations in feature representations, leading to more consistent background reconstruction. Experimental results on several real-world hyperspectral datasets demonstrate that MGAE outperforms existing methods.Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder (AE)-based methods are widely used due to their excellent image reconstruction capability. However, traditional grid-based image representation methods struggle to capture long-range dependencies and model non-Euclidean structures. To address these issues, this paper proposes a self-supervised Masked Graph AutoEncoder (MGAE) for hyperspectral anomaly detection. MGAE utilizes a Graph Attention Network (GAT) autoencoder to reconstruct the background of hyperspectral images and identifies anomalies by comparing the reconstructed features with the original features. Specifically, we constructs a topological graph structure of the hyperspectral image, which is then input into the GAT autoencoder for reconstruction, leveraging the multi-head attention mechanism to learn spatial and spectral features. To prevent the decoder from learning trivial solutions, we introduce a re-masking strategy that randomly masks both the input features and hidden representations during training, forcing the model to learn and reconstruct features under limited information, thereby improving detection performance. Additionally, the proposed loss function with graph Laplacian regularization (Twice Loss) minimizes variations in feature representations, leading to more consistent background reconstruction. Experimental results on several real-world hyperspectral datasets demonstrate that MGAE outperforms existing methods. |
| Author | Li, Jun Liu, Bo He, Yan Zhou, Tao Tu, Bing He, Baoliang Plaza, Antonio |
| Author_xml | – sequence: 1 givenname: Bing orcidid: 0000-0001-5802-9496 surname: Tu fullname: Tu, Bing email: tubing@nuist.edu.cn organization: Institute of Optics and Electronics, the State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion, Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection, and Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere, Nanjing University of Information Science and Technology, Nanjing, China – sequence: 2 givenname: Baoliang surname: He fullname: He, Baoliang email: baoliang_he@163.com organization: Institute of Optics and Electronics, the State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion, Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection, and Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere, Nanjing University of Information Science and Technology, Nanjing, China – sequence: 3 givenname: Yan surname: He fullname: He, Yan email: 975861884@qq.com organization: Institute of Optics and Electronics, the State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion, Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection, and Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere, Nanjing University of Information Science and Technology, Nanjing, China – sequence: 4 givenname: Tao orcidid: 0009-0000-4967-3175 surname: Zhou fullname: Zhou, Tao email: zhoutao@stu.ahjzu.edu.cn organization: School of Electronic Information and Electrical Engineering, Anhui Jianzhu University, Hefei, China – sequence: 5 givenname: Bo orcidid: 0000-0002-9603-9975 surname: Liu fullname: Liu, Bo email: bo@nuist.edu.cn organization: Institute of Optics and Electronics, the State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion, Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection, and Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere, Nanjing University of Information Science and Technology, Nanjing, China – sequence: 6 givenname: Jun orcidid: 0000-0003-1613-9448 surname: Li fullname: Li, Jun email: lijuncug@cug.edu.cn organization: Faculty of Computer Science, China University of Geosciences, Wuhan, China – sequence: 7 givenname: Antonio orcidid: 0000-0002-9613-1659 surname: Plaza fullname: Plaza, Antonio email: aplaza@unex.es organization: Department of Technology of Computers and Communications, Escuela Politécnica, Hyperspectral Computing Laboratory, University of Extremadura, Cáceres, Spain |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/41100244$$D View this record in MEDLINE/PubMed |
| BookMark | eNpd0c1LwzAYBvAgitPp3YNIwYuXzjcfbdLjmJ8wUdg8l7R9g9WuqUkr7L83Y1PB0xvC7wnhfY7JfmtbJOSMwoRSyK6Xjy8TBiyZ8JQBZHSPHNFM0BhAsP1whkTGkopsRI69fwegIqHpIRmJkAYmxBGZL7Ax8WLo0H3VHqvoSfuPMO6d7t6i6dBbbEtboYuMddHDOjjfYdk73UTT1q50s45usA83tW1PyIHRjcfT3RyT17vb5ewhnj_fP86m87jkwPuYFxmqVDGlFDLkABVqikkitSlNkgrBTVZII5QqEi4KDqbSWkpAnqJUleBjcrV9t3P2c0Df56val9g0ukU7-JyzlDGmGNBAL__Rdzu4Nvxuo9JESqk26mKnhmKFVd65eqXdOv_ZUwCwBaWz3js0v4RCvqkiD1XkmyryXRUhcr6N1Ij4xymDFCjwb-Pigo4 |
| CODEN | IIPRE4 |
| Cites_doi | 10.1109/TGRS.2024.3351179 10.1109/TGRS.2023.3276175 10.1109/TGRS.2021.3069716 10.3390/rs17142438 10.2307/1403797 10.1109/TIT.1967.1053964 10.1109/TPAMI.2025.3557581 10.1126/science.290.5500.2323 10.1109/18.857796 10.1117/12.850741 10.1109/ICCV.2019.00533 10.11834/jrs.20210283 10.1109/TGRS.2023.3335484 10.1109/ACCESS.2025.3534981 10.1109/TIP.2021.3078329 10.1049/cit2.12154 10.1109/lgrs.2024.3449635 10.1109/TGRS.2024.3456548 10.1109/TGRS.2004.841487 10.1109/29.60107 10.1109/JSTARS.2025.3542457 10.1109/TCYB.2020.2968750 10.1117/12.745429 10.3390/rs14081784 10.1109/JSTARS.2019.2940278 10.1109/JSTARS.2022.3167830 10.1109/TGRS.2025.3593019 10.1109/JSTARS.2025.3580751 10.1109/TIP.2024.3385295 10.1109/TGRS.2023.3262928 10.1109/TIP.2023.3306916 10.1016/j.eswa.2025.127366 10.1109/ICCWAMTIP51612.2020.9317527 10.1109/ACCESS.2025.3530437 10.1109/TGRS.2006.873019 10.1109/TIP.2021.3071557 10.1109/TGRS.2025.3534185 10.1109/TGRS.2019.2936308 10.1109/JSTARS.2013.2238609 10.1109/IGARSS.2017.8127159 10.1109/TIP.2022.3144017 10.1016/j.inffus.2022.08.011 10.1109/MSP.2022.3217936 10.1109/MSP.2013.2278992 10.1109/TGRS.2025.3547145 10.1109/TCYB.2022.3175771 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025 |
| DBID | 97E RIA RIE AAYXX CITATION NPM 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 |
| DOI | 10.1109/TIP.2025.3620091 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef PubMed Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional MEDLINE - Academic |
| DatabaseTitle | CrossRef PubMed Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional MEDLINE - Academic |
| DatabaseTitleList | Technology Research Database MEDLINE - Academic PubMed |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Applied Sciences Engineering |
| EISSN | 1941-0042 |
| EndPage | 6729 |
| ExternalDocumentID | 41100244 10_1109_TIP_2025_3620091 11206010 |
| Genre | orig-research Journal Article |
| GrantInformation_xml | – fundername: National Natural Science Foundation of China grantid: 62535010; 62271200; 62375083 funderid: 10.13039/501100001809 – fundername: Start-Up Foundation for Introducing Talent of Nanjing University of Information Science and Technology (NUIST) grantid: 2023r091 funderid: 10.13039/501100013156 |
| GroupedDBID | --- -~X .DC 0R~ 29I 4.4 53G 5GY 5VS 6IK 97E AAJGR AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD F5P HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 AAYXX CITATION NPM 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 |
| ID | FETCH-LOGICAL-c303t-3b9e8682888e2e300dea1e557afcf56443f9b7f488b534b30fdaa770e36e78d43 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001604932300002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1057-7149 1941-0042 |
| IngestDate | Sat Oct 18 23:54:48 EDT 2025 Thu Oct 30 04:57:35 EDT 2025 Sat Nov 01 10:37:13 EDT 2025 Sat Nov 29 07:01:37 EST 2025 Wed Nov 05 07:07:56 EST 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-c303t-3b9e8682888e2e300dea1e557afcf56443f9b7f488b534b30fdaa770e36e78d43 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0003-1613-9448 0000-0001-5802-9496 0009-0000-4967-3175 0000-0002-9603-9975 0000-0002-9613-1659 |
| PMID | 41100244 |
| PQID | 3266577781 |
| PQPubID | 85429 |
| PageCount | 16 |
| ParticipantIDs | proquest_miscellaneous_3262228201 proquest_journals_3266577781 ieee_primary_11206010 pubmed_primary_41100244 crossref_primary_10_1109_TIP_2025_3620091 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-01-01 |
| PublicationDateYYYYMMDD | 2025-01-01 |
| PublicationDate_xml | – month: 01 year: 2025 text: 2025-01-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States – name: New York |
| PublicationTitle | IEEE transactions on image processing |
| PublicationTitleAbbrev | TIP |
| PublicationTitleAlternate | IEEE Trans Image Process |
| PublicationYear | 2025 |
| 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 ref11 ref10 ref17 ref16 ref19 ref18 Brody (ref36) 2021 ref46 ref45 ref48 ref47 ref41 ref44 ref43 ref49 ref8 ref7 ref9 ref4 ref3 ref6 ref5 Baldi (ref25) ref34 ref37 ref31 ref30 ref33 ref32 ref2 ref1 ref39 ref38 ref24 ref23 ref26 ref20 ref22 ref21 Roweis (ref42) 2000; 290 ref28 ref27 ref29 Fix (ref40) 1989; 57 Veličković (ref35) 2017 |
| References_xml | – ident: ref23 doi: 10.1109/TGRS.2024.3351179 – ident: ref24 doi: 10.1109/TGRS.2023.3276175 – year: 2021 ident: ref36 article-title: How attentive are graph attention networks? publication-title: arXiv:2105.14491 – ident: ref5 doi: 10.1109/TGRS.2021.3069716 – ident: ref28 doi: 10.3390/rs17142438 – volume: 57 start-page: 238 year: 1989 ident: ref40 article-title: Discriminatory analysis—Nonparametric discrimination: Consistency properties publication-title: Int. Stat. Rev. doi: 10.2307/1403797 – ident: ref41 doi: 10.1109/TIT.1967.1053964 – ident: ref34 doi: 10.1109/TPAMI.2025.3557581 – volume: 290 start-page: 2323 issue: 5500 year: 2000 ident: ref42 article-title: Nonlinear dimensionality reduction by locally linear embedding publication-title: Science doi: 10.1126/science.290.5500.2323 – ident: ref13 doi: 10.1109/18.857796 – ident: ref10 doi: 10.1117/12.850741 – year: 2017 ident: ref35 article-title: Graph attention networks publication-title: arXiv:1710.10903 – ident: ref43 doi: 10.1109/ICCV.2019.00533 – ident: ref1 doi: 10.11834/jrs.20210283 – ident: ref48 doi: 10.1109/TGRS.2023.3335484 – ident: ref26 doi: 10.1109/ACCESS.2025.3534981 – ident: ref15 doi: 10.1109/TIP.2021.3078329 – ident: ref27 doi: 10.1049/cit2.12154 – ident: ref49 doi: 10.1109/lgrs.2024.3449635 – ident: ref3 doi: 10.1109/TGRS.2024.3456548 – ident: ref11 doi: 10.1109/TGRS.2004.841487 – ident: ref8 doi: 10.1109/29.60107 – ident: ref32 doi: 10.1109/JSTARS.2025.3542457 – ident: ref46 doi: 10.1109/TCYB.2020.2968750 – ident: ref19 doi: 10.1117/12.745429 – ident: ref12 doi: 10.3390/rs14081784 – ident: ref47 doi: 10.1109/JSTARS.2019.2940278 – ident: ref20 doi: 10.1109/JSTARS.2022.3167830 – ident: ref16 doi: 10.1109/TGRS.2025.3593019 – ident: ref33 doi: 10.1109/JSTARS.2025.3580751 – ident: ref39 doi: 10.1109/TIP.2024.3385295 – ident: ref4 doi: 10.1109/TGRS.2023.3262928 – ident: ref38 doi: 10.1109/TIP.2023.3306916 – ident: ref29 doi: 10.1016/j.eswa.2025.127366 – ident: ref45 doi: 10.1109/ICCWAMTIP51612.2020.9317527 – ident: ref22 doi: 10.1109/ACCESS.2025.3530437 – start-page: 37 volume-title: Proc. Int. Conf. Unsupervised Transf. Learn. Workshop ident: ref25 article-title: Autoencoders, unsupervised learning and deep architectures – ident: ref18 doi: 10.1109/TGRS.2006.873019 – ident: ref2 doi: 10.1109/TIP.2021.3071557 – ident: ref30 doi: 10.1109/TGRS.2025.3534185 – ident: ref21 doi: 10.1109/TGRS.2019.2936308 – ident: ref9 doi: 10.1109/JSTARS.2013.2238609 – ident: ref7 doi: 10.1109/IGARSS.2017.8127159 – ident: ref37 doi: 10.1109/TIP.2022.3144017 – ident: ref44 doi: 10.1016/j.inffus.2022.08.011 – ident: ref14 doi: 10.1109/MSP.2022.3217936 – ident: ref6 doi: 10.1109/MSP.2013.2278992 – ident: ref31 doi: 10.1109/TGRS.2025.3547145 – ident: ref17 doi: 10.1109/TCYB.2022.3175771 |
| SSID | ssj0014516 |
| Score | 2.4676416 |
| Snippet | Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder(AE)-based methods are widely used due to... Hyperspectral image anomaly detection faces the challenge of difficulty in annotating anomalous targets. Autoencoder (AE)-based methods are widely used due to... |
| SourceID | proquest pubmed crossref ieee |
| SourceType | Aggregation Database Index Database Publisher |
| StartPage | 6714 |
| SubjectTerms | Anomalies Anomaly detection autoencoder Autoencoders Decoding Detectors Feature extraction Gaussian distribution graph attention network Hyperspectral imaging Image reconstruction Kernel Laplace equations re-masking strategy Regularization Representations twice loss |
| Title | Self-Supervised Masked Graph Autoencoder for Hyperspectral Anomaly Detection |
| URI | https://ieeexplore.ieee.org/document/11206010 https://www.ncbi.nlm.nih.gov/pubmed/41100244 https://www.proquest.com/docview/3266577781 https://www.proquest.com/docview/3262228201 |
| Volume | 34 |
| WOSCitedRecordID | wos001604932300002&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: 1941-0042 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014516 issn: 1057-7149 databaseCode: RIE dateStart: 19920101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Bb9UwDLZg2gEOjI0BD8bUSbtwyNbXNnVynBhjSDBN2ia9W5UmjoQY7fTaIvHvcdK-N3bYgVMr1W2j2I4_x7ENcGgkY9qMUkHoSlH4IhOGdCFsljlLc0cK69hsAi8u1GKhL6dk9ZgLQ0Tx8BkdhdsYy3etHcJW2TFjg1A-hD30p4jlmKy1DhmEjrMxtClRIOP-VUwy1cfXXy_ZE8zkEa_WjClCd5gilEpj0_bAHMX-Ko9DzWhyzrb-c7Av4cWELZOTURi24Qk1O7A14cxk0uJuB57_U4TwFXy7olsvroa7sGp0TPjddD_58iWUsk5Ohr4NtS4dLRPGt8k5-61jeuYy_Kppf5nbP8kp9fFIV7MLN2efrz-di6nHgrBsvHqR15pUyW6XUpRRnqaOzJykROOtlwyWcq9r9KzmtcyLOk-9MwYxpbwkVK7IX8NG0zb0FhJHNiu8L-dWBXqpLRVI3mhX5kjazuDjaqqru7GURhVdkFRXzKEqcKiaODSD3TCj93TTZM5gb8WcalK2rmIEWkpEVPzawfoxq0mIfZiG2iHShL0uhjszeDMydf3xlSy8e-Sn7-FZGNu48bIHG_1yoA-waX_3P7rlPsviQu1HWfwLYAfYtg |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dT9wwDLcQQ9r2MBiD7RhsnbSXPQR6bdM0j2h8HNpxQuIm8ValiSMhWIuu7aT99zhp7jYeeNhTK9VpojiOf45jG-Cr4oRpE4wZCpOzzGYJUygzppPEaBwbLETli02I2ay4uZFXIVjdx8Igor98hofu1fvyTaN7d1R2RNjApQ8hC_2FK50VwrVWTgNXc9Y7N7lggpD_0isZy6P5xRXZggk_pP2aUIWrD5O5ZGmk3J4oJF9h5Xmw6ZXO2eZ_DncL3gR0GR0Py-EtrGG9DZsBaUZBjttteP1PGsJ3ML3Ge8uu-we3b7REeKnaO3qcu2TW0XHfNS7bpcFFRAg3mpDlOgRoLlxXdfNL3f-JTrDzl7rqHfh5djr_PmGhygLTpL46llYSi5wMr6LABNM4NqjGyLlQVltOcCm1shKWBL3iaValsTVKCRFjmqMoTJbuwnrd1PgBIoM6yazNx7pw9FxqzARaJU2eCpR6BN-WU10-DMk0Sm-ExLIkDpWOQ2Xg0Ah23Iz-pQuTOYL9JXPKIG5tSRg050KIgpp9WX0mQXHeD1Vj03sad9pFgGcE7wemrn6-XAt7z3T6GV5O5pfTcnox-_ERXrlxDscw-7DeLXo8gA39u7ttF5_8inwEuK_bFw |
| 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=Self-Supervised+Masked+Graph+Autoencoder+for+Hyperspectral+Anomaly+Detection&rft.jtitle=IEEE+transactions+on+image+processing&rft.au=Tu%2C+Bing&rft.au=He%2C+Baoliang&rft.au=He%2C+Yan&rft.au=Zhou%2C+Tao&rft.date=2025-01-01&rft.issn=1057-7149&rft.eissn=1941-0042&rft.volume=34&rft.spage=6714&rft.epage=6729&rft_id=info:doi/10.1109%2FTIP.2025.3620091&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TIP_2025_3620091 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1057-7149&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1057-7149&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1057-7149&client=summon |