Few-Shot Hyperspectral Image Classification Based on Convolutional Residuals and SAM Siamese Networks
With the development of few-shot learning, significant progress has been achieved in hyperspectral image classification using related networks, leading to improved classification outcomes. However, practical few-shot hyperspectral image classification encounters challenges such as network overfittin...
Saved in:
Published in | Electronics (Basel) Vol. 12; no. 16; p. 3415 |
---|---|
Main Authors | , , , , , , |
Format | Journal Article |
Language | English |
Published |
Basel
MDPI AG
01.08.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | With the development of few-shot learning, significant progress has been achieved in hyperspectral image classification using related networks, leading to improved classification outcomes. However, practical few-shot hyperspectral image classification encounters challenges such as network overfitting and insufficient feature extraction during the model training process. To address these issues, we propose a model called CRSSNet (Convolutional Residuals and SAM Siamese Networks) for few-shot hyperspectral image classification. In this model, we deepen the network depth and employ the convolutional residual technique to enhance the feature extraction capabilities and alleviate the problem of network gradient degradation. Additionally, we introduce the Spatial Attention Mechanism (SAM) to effectively leverage spatial information features in hyperspectral images. Lastly, metric learning is employed by comparing the distance between two output feature vectors to determine the label category. Experimental results demonstrate that our method achieves superior classification performance compared to other methods. |
---|---|
AbstractList | With the development of few-shot learning, significant progress has been achieved in hyperspectral image classification using related networks, leading to improved classification outcomes. However, practical few-shot hyperspectral image classification encounters challenges such as network overfitting and insufficient feature extraction during the model training process. To address these issues, we propose a model called CRSSNet (Convolutional Residuals and SAM Siamese Networks) for few-shot hyperspectral image classification. In this model, we deepen the network depth and employ the convolutional residual technique to enhance the feature extraction capabilities and alleviate the problem of network gradient degradation. Additionally, we introduce the Spatial Attention Mechanism (SAM) to effectively leverage spatial information features in hyperspectral images. Lastly, metric learning is employed by comparing the distance between two output feature vectors to determine the label category. Experimental results demonstrate that our method achieves superior classification performance compared to other methods. |
Audience | Academic |
Author | Xia, Mengen Zhou, Hao Xia, Kunming Shi, Zhiliang Ren, Ying Yuan, Guowu Yang, Lingyu |
Author_xml | – sequence: 1 givenname: Mengen surname: Xia fullname: Xia, Mengen – sequence: 2 givenname: Guowu orcidid: 0000-0002-8449-6861 surname: Yuan fullname: Yuan, Guowu – sequence: 3 givenname: Lingyu surname: Yang fullname: Yang, Lingyu – sequence: 4 givenname: Kunming surname: Xia fullname: Xia, Kunming – sequence: 5 givenname: Ying surname: Ren fullname: Ren, Ying – sequence: 6 givenname: Zhiliang surname: Shi fullname: Shi, Zhiliang – sequence: 7 givenname: Hao orcidid: 0000-0002-3150-7029 surname: Zhou fullname: Zhou, Hao |
BookMark | eNptUctOwzAQtFCRKKVfwMUS5xQ_8vKxVJRWKiBROEfGWReXJA52QtW_x1E5cGAvO9qdGe3jEo0a2wBC15TMOBfkFipQnbONUZ4ymvKYJmdozEgmIsEEG_3BF2jq_Z6EEJTnnIwRLOEQbT9sh1fHFpxvBy9Z4XUtd4AXlfTeaKNkZ2yD76SHEgewsM23rfqhGLgv4E3Zy8pj2ZR4O3_EWyNr8ICfoDtY9-mv0LkOfZj-5gl6W96_LlbR5vlhvZhvIsVT2kVKEapoHBMRi4xlgiSMvXOtS5lBmVAKOQihZAkqozRVmuepIowQTgiUJZd8gm5Ovq2zXz34rtjb3oUZfcHyJCM0YTwNrNmJtZMVFKbRNqys5GBcGxWOq02oz7OUxWmWsyQI-EmgnPXegS5aZ2rpjgUlxfCD4p8f8B9KI37k |
CitedBy_id | crossref_primary_10_1016_j_ophoto_2024_100062 crossref_primary_10_3390_rs15225306 crossref_primary_10_3390_rs16122103 crossref_primary_10_1007_s11082_024_06753_5 crossref_primary_10_1117_1_JRS_18_016514 crossref_primary_10_3390_electronics12173641 crossref_primary_10_3390_electronics12244956 |
Cites_doi | 10.1109/TGRS.2017.2755542 10.1109/JSTARS.2018.2872969 10.1109/TNNLS.2022.3185795 10.3390/rs12060923 10.1109/MGRS.2017.2762087 10.1109/WHISPERS56178.2022.9955116 10.1109/CVPR.2018.00131 10.1109/LGRS.2020.2979604 10.1109/TGRS.2011.2129595 10.1109/ICIP46576.2022.9897181 10.3390/rs9010067 10.1109/CVPR.2016.90 10.1109/WHISPERS52202.2021.9484047 10.1109/IGARSS46834.2022.9884537 10.1109/TGRS.2020.2963848 10.1126/science.1127647 10.1109/WHISPERS.2019.8920973 10.1109/JSTARS.2020.3017544 10.1109/TGRS.2004.831865 10.1162/neco.2006.18.7.1527 10.1109/TGRS.2018.2872830 10.1186/s13634-022-00926-8 |
ContentType | Journal Article |
Copyright | COPYRIGHT 2023 MDPI AG 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: COPYRIGHT 2023 MDPI AG – notice: 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | AAYXX CITATION 7SP 8FD 8FE 8FG ABUWG AFKRA ARAPS AZQEC BENPR BGLVJ CCPQU DWQXO HCIFZ L7M P5Z P62 PIMPY PQEST PQQKQ PQUKI PRINS |
DOI | 10.3390/electronics12163415 |
DatabaseName | CrossRef Electronics & Communications Abstracts Technology Research Database ProQuest SciTech Collection ProQuest Technology Collection ProQuest Central (Alumni Edition) ProQuest Central UK/Ireland Advanced Technologies & Aerospace Database (1962 - current) ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central Technology Collection ProQuest One Community College ProQuest Central SciTech Premium Collection Advanced Technologies Database with Aerospace Advanced Technologies & Aerospace Database ProQuest Advanced Technologies & Aerospace Collection Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China |
DatabaseTitle | CrossRef Publicly Available Content Database Advanced Technologies & Aerospace Collection Technology Collection Technology Research Database ProQuest Advanced Technologies & Aerospace Collection ProQuest Central Essentials ProQuest One Academic Eastern Edition Electronics & Communications Abstracts ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Technology Collection ProQuest SciTech Collection ProQuest Central China ProQuest Central Advanced Technologies & Aerospace Database ProQuest One Academic UKI Edition ProQuest Central Korea ProQuest One Academic Advanced Technologies Database with Aerospace |
DatabaseTitleList | CrossRef Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering |
EISSN | 2079-9292 |
ExternalDocumentID | A762467825 10_3390_electronics12163415 |
GeographicLocations | China |
GeographicLocations_xml | – name: China |
GroupedDBID | 5VS 8FE 8FG AAYXX AFKRA ALMA_UNASSIGNED_HOLDINGS ARAPS BENPR BGLVJ CCPQU CITATION GROUPED_DOAJ HCIFZ IAO ITC KQ8 MODMG M~E OK1 P62 PIMPY PROAC 7SP 8FD ABUWG AZQEC DWQXO L7M PQEST PQQKQ PQUKI PRINS |
ID | FETCH-LOGICAL-c361t-cc01c144094972790522b3ffda7ed511e8e99cadec7116cf386c0200300edd3a3 |
IEDL.DBID | BENPR |
ISSN | 2079-9292 |
IngestDate | Thu Oct 10 19:30:50 EDT 2024 Tue Mar 12 05:15:52 EDT 2024 Fri Aug 23 02:37:07 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 16 |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c361t-cc01c144094972790522b3ffda7ed511e8e99cadec7116cf386c0200300edd3a3 |
ORCID | 0000-0002-8449-6861 0000-0002-3150-7029 |
OpenAccessLink | https://www.proquest.com/docview/2857015236?pq-origsite=%requestingapplication% |
PQID | 2857015236 |
PQPubID | 2032404 |
ParticipantIDs | proquest_journals_2857015236 gale_infotracacademiconefile_A762467825 crossref_primary_10_3390_electronics12163415 |
PublicationCentury | 2000 |
PublicationDate | 2023-08-01 |
PublicationDateYYYYMMDD | 2023-08-01 |
PublicationDate_xml | – month: 08 year: 2023 text: 2023-08-01 day: 01 |
PublicationDecade | 2020 |
PublicationPlace | Basel |
PublicationPlace_xml | – name: Basel |
PublicationTitle | Electronics (Basel) |
PublicationYear | 2023 |
Publisher | MDPI AG |
Publisher_xml | – name: MDPI AG |
References | Xue (ref_23) 2022; 60 Zhong (ref_34) 2018; 56 Luo (ref_38) 2020; 58 Tu (ref_9) 2018; 11 ref_14 ref_36 ref_11 ref_33 ref_32 ref_31 ref_30 Zhang (ref_25) 2020; 13 ref_18 Zhao (ref_21) 2021; 32 ref_17 Huang (ref_29) 2021; 18 ref_16 ref_15 Hinton (ref_13) 2006; 18 Chen (ref_8) 2011; 49 Zhang (ref_10) 2018; 44 Zhang (ref_39) 2022; 2022 Li (ref_37) 2022; 60 ref_22 ref_1 Liu (ref_35) 2019; 57 Liao (ref_19) 2023; 61 ref_3 ref_2 Hinton (ref_12) 2006; 313 ref_27 ref_26 Wang (ref_28) 2022; 19 Melgani (ref_6) 2004; 42 ref_5 Ghamisi (ref_7) 2017; 5 ref_4 Tang (ref_24) 2022; 19 Cui (ref_20) 2017; 21 |
References_xml | – volume: 56 start-page: 847 year: 2018 ident: ref_34 article-title: Spectral-spatial residual network for hyperspectral image classification: A 3-D deep learning framework publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2017.2755542 contributor: fullname: Zhong – volume: 11 start-page: 4032 year: 2018 ident: ref_9 article-title: KNN-based representation of super pixels for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2018.2872969 contributor: fullname: Tu – ident: ref_30 – ident: ref_36 doi: 10.1109/TNNLS.2022.3185795 – ident: ref_32 – volume: 44 start-page: 961 year: 2018 ident: ref_10 article-title: Deep Learning for Hyperspectral Imagery Classification: The State of the Art and Prospects publication-title: Acta Autom. Sin. contributor: fullname: Zhang – ident: ref_27 doi: 10.3390/rs12060923 – volume: 5 start-page: 37 year: 2017 ident: ref_7 article-title: Advances in hyperspectral image and signal processing a comprehensive overview of the state of the art publication-title: IEEE Geosci. Remote Sens. doi: 10.1109/MGRS.2017.2762087 contributor: fullname: Ghamisi – ident: ref_11 – ident: ref_15 doi: 10.1109/WHISPERS56178.2022.9955116 – ident: ref_16 – ident: ref_26 doi: 10.1109/CVPR.2018.00131 – volume: 18 start-page: 518 year: 2021 ident: ref_29 article-title: Dual-Path Siamese CNN for Hyperspectral Image Classification With Limited Training Samples publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2020.2979604 contributor: fullname: Huang – ident: ref_14 – ident: ref_1 – volume: 21 start-page: 728 year: 2017 ident: ref_20 article-title: Hyperspectral image de-noising and classification with small training samples publication-title: J. Remote Sens. contributor: fullname: Cui – volume: 49 start-page: 3973 year: 2011 ident: ref_8 article-title: Hyperspectral image classification using dictionary-based sparse representation publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2011.2129595 contributor: fullname: Chen – volume: 19 start-page: 1 year: 2022 ident: ref_28 article-title: Soft Augmentation-Based Siamese CNN for Hyperspectral Image Classification With Limited Training Samples publication-title: IEEE Geosci. Remote Sens. Lett. contributor: fullname: Wang – volume: 32 start-page: 349 year: 2021 ident: ref_21 article-title: Survey on few-shot learning publication-title: Ruan Jian Xue Bao/J. Softw. contributor: fullname: Zhao – ident: ref_5 doi: 10.1109/ICIP46576.2022.9897181 – ident: ref_33 doi: 10.3390/rs9010067 – ident: ref_17 doi: 10.1109/CVPR.2016.90 – volume: 61 start-page: 1 year: 2023 ident: ref_19 article-title: A Spectral–Spatial Fusion Transformer Network for Hyperspectral Image Classification publication-title: IEEE Trans. Geosci. Remote Sens. contributor: fullname: Liao – volume: 60 start-page: 1 year: 2022 ident: ref_37 article-title: Deep Cross Domain Few-Shot Learning For Hyperspectral Image Classification publication-title: IEEE Trans. Geosci. Remote Sens. contributor: fullname: Li – ident: ref_31 – ident: ref_2 – ident: ref_4 doi: 10.1109/WHISPERS52202.2021.9484047 – ident: ref_18 doi: 10.1109/IGARSS46834.2022.9884537 – volume: 58 start-page: 5336 year: 2020 ident: ref_38 article-title: Dimensionality reduction with enhanced hybrid-graph discriminant learning for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2020.2963848 contributor: fullname: Luo – volume: 19 start-page: 1 year: 2022 ident: ref_24 article-title: A Multiscale Spatial–Spectral Prototypical Network for Hyperspectral Image Few-Shot Classification publication-title: IEEE Geosci. Remote Sens. Lett. contributor: fullname: Tang – volume: 313 start-page: 504 year: 2006 ident: ref_12 article-title: Reducing the dimensionality of data with neural networks publication-title: Science doi: 10.1126/science.1127647 contributor: fullname: Hinton – ident: ref_3 doi: 10.1109/WHISPERS.2019.8920973 – volume: 13 start-page: 4748 year: 2020 ident: ref_25 article-title: Global Prototypical Network for Few-Shot Hyperspectral Image Classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2020.3017544 contributor: fullname: Zhang – volume: 42 start-page: 1778 year: 2004 ident: ref_6 article-title: Classification of hyperspectral remote sensing images with support vector machines publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2004.831865 contributor: fullname: Melgani – volume: 18 start-page: 1527 year: 2006 ident: ref_13 article-title: A fast learning algorithm for deep belief nets publication-title: Neural Comput. doi: 10.1162/neco.2006.18.7.1527 contributor: fullname: Hinton – volume: 60 start-page: 1 year: 2022 ident: ref_23 article-title: S3Net: Spectral–Spatial Siamese Network for Few-Shot Hyperspectral Image Classification publication-title: IEEE Trans. Geosci. Remote Sens. contributor: fullname: Xue – volume: 57 start-page: 2290 year: 2019 ident: ref_35 article-title: Deep few-shot learning for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2018.2872830 contributor: fullname: Liu – ident: ref_22 – volume: 2022 start-page: 92 year: 2022 ident: ref_39 article-title: Multilayer graph spectral analysis for hyperspectral images publication-title: EURASIP J. Adv. Signal Process. doi: 10.1186/s13634-022-00926-8 contributor: fullname: Zhang |
SSID | ssj0000913830 |
Score | 2.3607612 |
Snippet | With the development of few-shot learning, significant progress has been achieved in hyperspectral image classification using related networks, leading to... |
SourceID | proquest gale crossref |
SourceType | Aggregation Database |
StartPage | 3415 |
SubjectTerms | Accuracy Agricultural production Algorithms Artificial neural networks Automatic classification Classification Deep learning Design Feature extraction Hyperspectral imaging Image classification Image processing Learning Machine learning Methods Neural networks Spatial data Teaching methods |
Title | Few-Shot Hyperspectral Image Classification Based on Convolutional Residuals and SAM Siamese Networks |
URI | https://www.proquest.com/docview/2857015236 |
Volume | 12 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3PT8IwFG4ELnow_owokh5MvLiwrqNsJwOEiSYQA5JwW0bbRQ9uyKbe_Nt9byvCgXjbsi1Lvrbvfa_93nuE3PAFjCoTseXpNrdcWE5YA9KxGAeuLNoLTxdZ_KOxGM7cp3l7bjbcMiOrXNvEwlCrVOIeecvBSuzgbLi4X35Y2DUKT1dNC40KqTnMxWPaWm8wfp787bJg1UuP22W5IQ7xfWvTXSZjDpARFxvibrmk3Ya58DbBETk0NJF2y3E9Jns6OSEHW8UDT4kO9Lc1fU1zOoRYskyZXME3j-9gImjR7BJlQAXytAfOSlG46KfJl5lt8O5EZ0UyVkajRNFpd0Snbyib1XRcysOzMzILBi_9oWWaJliSC5ZbUtpM4omt7_rATXwbCNaCx7GKOloBu9KAvo_Se9lhTMiYe0LaqFCzba0Uj_g5qSZpoi8I1ZET-X4ndoRWbixdX0cwAhIYQAR-T_h1crfGLVyWtTFCiCkQ5nAHzHVyi9iGuHIADxmZBAD4GdagCrtgl8FsQ8haJ401_KFZUlm4mQCX_z--IvvYE75U6TVINV996mtgDvmiSSpe8NA0kwTuRj-DX3ruxsg |
link.rule.ids | 315,786,790,12792,21416,27955,27956,33406,33777,43633,43838,74390,74657 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3PT4MwGG10HtSD8WecTu3BxIuNQFkHJzMXkem2g9uS3ZqulOhBmAP13_f7gLkdFm8kQEhe2-9HeX2PkGs-hVG1Rcw80-TMheWEGpAOsznUyqI59Uxxir8_EOHYfZ40J9WGW1bRKhcxsQjUUapxj_zOQSV2SDZc3M8-GbpG4d_VykJjk2y5XHCc517w9LfHgpqXHrdKsSEO3f3d0lsmsx0oRVy0w11JSOvDcpFrgn2yVxWJtF2O6gHZMMkh2V2RDjwiJjA_bPiW5jSETrI8MDmHd7ofECBoYXWJJKACd_oAqSqicNFJk-9qrsGzryYrjmJlVCURHbb7dPiOpFlDByU5PDsm4-Bx1AlZZZnANBd2zrS2bI3_a33Xh8rEt6C8mvI4jlTLRFBbGcDeR-K9btm20DH3hLaQn2ZZJoq44ieklqSJOSXUKEf5fit2hIncWLu-UYC_hvyvIOsJv05uF7jJWamMIaGjQJjlGpjr5AaxlbhuAA-tKvo_fAwVqGQbojIEbWhY66SxgF9WCyqTy-E_-__2FdkOR_2e7HUHL-dkB93hS75eg9Ty-Ze5gBoin14WE-UXQMjGOA |
linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3PS8MwFA46QfQg_sTp1BwEL4a1TZe1J5nTuqkb4hx4K22SogfbuVb9932vzdwOw1shLYX3Xt77knx5HyHnPAav2iJhnm5x5sJ0wh6QDrM5YGXRij1d3uIfDEVv7N6_tl4N_yk3tMpZTiwTtcok7pE3HezEDsWGi2ZiaBFPN8HV5JOhghSetBo5jVWyhiAbZRy84O5vvwX7X3rcqhoPcRhvznVmctsBWOKiNO5CcVqeosu6E2yTLQMYaafy8A5Z0eku2VxoI7hHdKB_2OgtK2gPVpXV5ckpfNP_gGRBS9lLJASVPqDXULYUhYduln6buIN3n3VeXsvKaZQqOuoM6OgdCbSaDiuieL5PxsHtS7fHjHwCk1zYBZPSsiWe3fquDyjFtwBqxTxJVNTWCnCWBj_4SMKXbdsWMuGekBZy1SxLK8UjfkBqaZbqQ0J15ES-304coZWbSNfXEfhCAhaIoAIKv04uZ3YLJ1WXjBBWF2jmcImZ6-QCbRviHAJ7yMhcBYCfYTeqsAMZGhI4LF7rpDEzf2gmVx7OQ-Ho_-Ezsg4xEj72hw_HZAOF4ivqXoPUiumXPgE4UcSnZZz8Aj6rym0 |
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=Few-Shot+Hyperspectral+Image+Classification+Based+on+Convolutional+Residuals+and+SAM+Siamese+Networks&rft.jtitle=Electronics+%28Basel%29&rft.au=Xia%2C+Mengen&rft.au=Yuan%2C+Guowu&rft.au=Yang%2C+Lingyu&rft.au=Xia%2C+Kunming&rft.date=2023-08-01&rft.pub=MDPI+AG&rft.issn=2079-9292&rft.eissn=2079-9292&rft.volume=12&rft.issue=16&rft_id=info:doi/10.3390%2Felectronics12163415&rft.externalDocID=A762467825 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2079-9292&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2079-9292&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2079-9292&client=summon |