Matches in SemOpenAlex for { <https://semopenalex.org/work/W4295290404> ?p ?o ?g. }
- W4295290404 endingPage "118797" @default.
- W4295290404 startingPage "118797" @default.
- W4295290404 abstract "The band selection (BS) is an essential task in hyperspectral images (HSIs), consisting of huge spectral bands with noises and redundancies. The attention mechanism can be used for BS. However, the existing attention-based BS schemes are failed to capture the cross-dimension interaction between the input spectral and spatial dimensions during computation of attention weights and may produce poor feature representations. Again, the used reconstruction network in the existing BS methods unable to detect the HSIs features in multiple scales by blindly increasing the depth of the network. To deal with these problems, a novel end-to-end unsupervised triplet-attention multiscale reconstruction network for BS ( TAttMSRecNet ) has been proposed. The proposed network utilizes a triplet-attention mechanism having three parallel branches responsible to aggregate interactive cross-dimensional features between the spatial and spectral dimensions. After that, the network restores the original HSIs by using a 3D multiscale reconstruction network that applies multiple size convolution kernels to capture the discriminative HSIs features over the multiple scales where these features also communicate themselves to find the most efficacious HSIs information. In this way, the rich features are captured at a little computation cost, and the most informative bands can be effectively chosen for classification. Three standard data sets - Indian Pines (IP), Salinas (SA), and University of Pavia (UP) have been taken to conduct the experiments. The presented TAttMSRecNet can efficiently suppress the redundant or useless bands and selects more informative bands for better classification performance and also outperforms the other existing BS methods. • We propose an end-to-end network for band selection in hyperspectral images. • It applies a triplet-attention with a multiscale reconstruction network. • Captures the robust feature representations at a low computation overhead. • Finds the most informative subset of spectral bands. • Yields significant classification performance improvements." @default.
- W4295290404 created "2022-09-12" @default.
- W4295290404 creator A5027312639 @default.
- W4295290404 creator A5035508615 @default.
- W4295290404 creator A5075013625 @default.
- W4295290404 creator A5079443694 @default.
- W4295290404 creator A5087427076 @default.
- W4295290404 date "2023-02-01" @default.
- W4295290404 modified "2023-10-10" @default.
- W4295290404 title "TAttMSRecNet:Triplet-attention and multiscale reconstruction network for band selection in hyperspectral images" @default.
- W4295290404 cites W1902936532 @default.
- W4295290404 cites W1932531222 @default.
- W4295290404 cites W1967275758 @default.
- W4295290404 cites W1977791453 @default.
- W4295290404 cites W1980007140 @default.
- W4295290404 cites W2028436154 @default.
- W4295290404 cites W2036833357 @default.
- W4295290404 cites W2039409148 @default.
- W4295290404 cites W2071185414 @default.
- W4295290404 cites W2071821878 @default.
- W4295290404 cites W2133125644 @default.
- W4295290404 cites W2133665775 @default.
- W4295290404 cites W2152786555 @default.
- W4295290404 cites W2288723698 @default.
- W4295290404 cites W2343745409 @default.
- W4295290404 cites W2743111138 @default.
- W4295290404 cites W2754507318 @default.
- W4295290404 cites W2764276316 @default.
- W4295290404 cites W2776449634 @default.
- W4295290404 cites W2789249105 @default.
- W4295290404 cites W2793357412 @default.
- W4295290404 cites W2808682285 @default.
- W4295290404 cites W2897121118 @default.
- W4295290404 cites W2904335384 @default.
- W4295290404 cites W2909721765 @default.
- W4295290404 cites W2919520811 @default.
- W4295290404 cites W2928182459 @default.
- W4295290404 cites W2937638900 @default.
- W4295290404 cites W2950325582 @default.
- W4295290404 cites W2953423780 @default.
- W4295290404 cites W2958108982 @default.
- W4295290404 cites W2997272341 @default.
- W4295290404 cites W3011660349 @default.
- W4295290404 cites W3047443805 @default.
- W4295290404 cites W3048317942 @default.
- W4295290404 cites W3048631361 @default.
- W4295290404 cites W3049737467 @default.
- W4295290404 cites W3100714546 @default.
- W4295290404 cites W3113649568 @default.
- W4295290404 cites W3114720220 @default.
- W4295290404 cites W3130821687 @default.
- W4295290404 cites W3132983279 @default.
- W4295290404 cites W3140885850 @default.
- W4295290404 cites W3167109952 @default.
- W4295290404 cites W3172608860 @default.
- W4295290404 cites W3189718454 @default.
- W4295290404 cites W4313229413 @default.
- W4295290404 doi "https://doi.org/10.1016/j.eswa.2022.118797" @default.
- W4295290404 hasPublicationYear "2023" @default.
- W4295290404 type Work @default.
- W4295290404 citedByCount "5" @default.
- W4295290404 countsByYear W42952904042023 @default.
- W4295290404 crossrefType "journal-article" @default.
- W4295290404 hasAuthorship W4295290404A5027312639 @default.
- W4295290404 hasAuthorship W4295290404A5035508615 @default.
- W4295290404 hasAuthorship W4295290404A5075013625 @default.
- W4295290404 hasAuthorship W4295290404A5079443694 @default.
- W4295290404 hasAuthorship W4295290404A5087427076 @default.
- W4295290404 hasConcept C153180895 @default.
- W4295290404 hasConcept C154945302 @default.
- W4295290404 hasConcept C159078339 @default.
- W4295290404 hasConcept C31972630 @default.
- W4295290404 hasConcept C41008148 @default.
- W4295290404 hasConcept C81917197 @default.
- W4295290404 hasConceptScore W4295290404C153180895 @default.
- W4295290404 hasConceptScore W4295290404C154945302 @default.
- W4295290404 hasConceptScore W4295290404C159078339 @default.
- W4295290404 hasConceptScore W4295290404C31972630 @default.
- W4295290404 hasConceptScore W4295290404C41008148 @default.
- W4295290404 hasConceptScore W4295290404C81917197 @default.
- W4295290404 hasFunder F4320318041 @default.
- W4295290404 hasFunder F4320337120 @default.
- W4295290404 hasLocation W42952904041 @default.
- W4295290404 hasOpenAccess W4295290404 @default.
- W4295290404 hasPrimaryLocation W42952904041 @default.
- W4295290404 hasRelatedWork W1491778359 @default.
- W4295290404 hasRelatedWork W1869808405 @default.
- W4295290404 hasRelatedWork W2028628118 @default.
- W4295290404 hasRelatedWork W2031007444 @default.
- W4295290404 hasRelatedWork W2775464024 @default.
- W4295290404 hasRelatedWork W2783789044 @default.
- W4295290404 hasRelatedWork W2972973180 @default.
- W4295290404 hasRelatedWork W3211035526 @default.
- W4295290404 hasRelatedWork W4291701050 @default.
- W4295290404 hasRelatedWork W4293272463 @default.
- W4295290404 hasVolume "212" @default.
- W4295290404 isParatext "false" @default.
- W4295290404 isRetracted "false" @default.