Matches in SemOpenAlex for { <https://semopenalex.org/work/W3047443805> ?p ?o ?g. }
- W3047443805 endingPage "5978" @default.
- W3047443805 startingPage "5966" @default.
- W3047443805 abstract "Convolutional neural networks (CNNs) have been attracting increasing attention in hyperspectral (HS) image classification due to their ability to capture spatial-spectral feature representations. Nevertheless, their ability in modeling relations between the samples remains limited. Beyond the limitations of grid sampling, graph convolutional networks (GCNs) have been recently proposed and successfully applied in irregular (or nongrid) data representation and analysis. In this article, we thoroughly investigate CNNs and GCNs (qualitatively and quantitatively) in terms of HS image classification. Due to the construction of the adjacency matrix on all the data, traditional GCNs usually suffer from a huge computational cost, particularly in large-scale remote sensing (RS) problems. To this end, we develop a new minibatch GCN (called miniGCN hereinafter), which allows to train large-scale GCNs in a minibatch fashion. More significantly, our miniGCN is capable of inferring out-of-sample data without retraining networks and improving classification performance. Furthermore, as CNNs and GCNs can extract different types of HS features, an intuitive solution to break the performance bottleneck of a single model is to fuse them. Since miniGCNs can perform batchwise network training (enabling the combination of CNNs and GCNs), we explore three fusion strategies: additive fusion, elementwise multiplicative fusion, and concatenation fusion to measure the obtained performance gain. Extensive experiments, conducted on three HS data sets, demonstrate the advantages of miniGCNs over GCNs and the superiority of the tested fusion strategies with regard to the single CNN or GCN models. The codes of this work will be available at https://github.com/danfenghong/IEEE_TGRS_GCN for the sake of reproducibility." @default.
- W3047443805 created "2020-08-10" @default.
- W3047443805 creator A5013885739 @default.
- W3047443805 creator A5035508615 @default.
- W3047443805 creator A5045473616 @default.
- W3047443805 creator A5054292278 @default.
- W3047443805 creator A5066378186 @default.
- W3047443805 creator A5075013625 @default.
- W3047443805 date "2021-07-01" @default.
- W3047443805 modified "2023-10-18" @default.
- W3047443805 title "Graph Convolutional Networks for Hyperspectral Image Classification" @default.
- W3047443805 cites W2029316659 @default.
- W3047443805 cites W2043665634 @default.
- W3047443805 cites W2097915756 @default.
- W3047443805 cites W2114819256 @default.
- W3047443805 cites W2127199143 @default.
- W3047443805 cites W2158787690 @default.
- W3047443805 cites W2163886442 @default.
- W3047443805 cites W2500751094 @default.
- W3047443805 cites W2588117332 @default.
- W3047443805 cites W2600746131 @default.
- W3047443805 cites W2602024454 @default.
- W3047443805 cites W2753248899 @default.
- W3047443805 cites W2782517596 @default.
- W3047443805 cites W2803057685 @default.
- W3047443805 cites W2889773939 @default.
- W3047443805 cites W2892621946 @default.
- W3047443805 cites W2902746003 @default.
- W3047443805 cites W2920405132 @default.
- W3047443805 cites W2937675449 @default.
- W3047443805 cites W2952565170 @default.
- W3047443805 cites W2953308875 @default.
- W3047443805 cites W2965945478 @default.
- W3047443805 cites W2969881582 @default.
- W3047443805 cites W2977355106 @default.
- W3047443805 cites W2994639710 @default.
- W3047443805 cites W3009883650 @default.
- W3047443805 cites W3037458146 @default.
- W3047443805 cites W3040988483 @default.
- W3047443805 cites W3097170121 @default.
- W3047443805 cites W3100011500 @default.
- W3047443805 cites W3101012758 @default.
- W3047443805 cites W3101640299 @default.
- W3047443805 cites W3104313739 @default.
- W3047443805 cites W3105021316 @default.
- W3047443805 cites W3122774149 @default.
- W3047443805 doi "https://doi.org/10.1109/tgrs.2020.3015157" @default.
- W3047443805 hasPublicationYear "2021" @default.
- W3047443805 type Work @default.
- W3047443805 sameAs 3047443805 @default.
- W3047443805 citedByCount "758" @default.
- W3047443805 countsByYear W30474438052020 @default.
- W3047443805 countsByYear W30474438052021 @default.
- W3047443805 countsByYear W30474438052022 @default.
- W3047443805 countsByYear W30474438052023 @default.
- W3047443805 crossrefType "journal-article" @default.
- W3047443805 hasAuthorship W3047443805A5013885739 @default.
- W3047443805 hasAuthorship W3047443805A5035508615 @default.
- W3047443805 hasAuthorship W3047443805A5045473616 @default.
- W3047443805 hasAuthorship W3047443805A5054292278 @default.
- W3047443805 hasAuthorship W3047443805A5066378186 @default.
- W3047443805 hasAuthorship W3047443805A5075013625 @default.
- W3047443805 hasBestOaLocation W30474438053 @default.
- W3047443805 hasConcept C114614502 @default.
- W3047443805 hasConcept C119857082 @default.
- W3047443805 hasConcept C124101348 @default.
- W3047443805 hasConcept C132525143 @default.
- W3047443805 hasConcept C149635348 @default.
- W3047443805 hasConcept C153180895 @default.
- W3047443805 hasConcept C154945302 @default.
- W3047443805 hasConcept C159078339 @default.
- W3047443805 hasConcept C180356752 @default.
- W3047443805 hasConcept C2780513914 @default.
- W3047443805 hasConcept C33923547 @default.
- W3047443805 hasConcept C41008148 @default.
- W3047443805 hasConcept C80444323 @default.
- W3047443805 hasConcept C81363708 @default.
- W3047443805 hasConcept C87619178 @default.
- W3047443805 hasConceptScore W3047443805C114614502 @default.
- W3047443805 hasConceptScore W3047443805C119857082 @default.
- W3047443805 hasConceptScore W3047443805C124101348 @default.
- W3047443805 hasConceptScore W3047443805C132525143 @default.
- W3047443805 hasConceptScore W3047443805C149635348 @default.
- W3047443805 hasConceptScore W3047443805C153180895 @default.
- W3047443805 hasConceptScore W3047443805C154945302 @default.
- W3047443805 hasConceptScore W3047443805C159078339 @default.
- W3047443805 hasConceptScore W3047443805C180356752 @default.
- W3047443805 hasConceptScore W3047443805C2780513914 @default.
- W3047443805 hasConceptScore W3047443805C33923547 @default.
- W3047443805 hasConceptScore W3047443805C41008148 @default.
- W3047443805 hasConceptScore W3047443805C80444323 @default.
- W3047443805 hasConceptScore W3047443805C81363708 @default.
- W3047443805 hasConceptScore W3047443805C87619178 @default.
- W3047443805 hasFunder F4320321001 @default.
- W3047443805 hasFunder F4320321048 @default.
- W3047443805 hasIssue "7" @default.
- W3047443805 hasLocation W30474438051 @default.
- W3047443805 hasLocation W30474438052 @default.