FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3089820912> ?p ?o ?g. }

• W3089820912 endingPage "5952" @default.
• W3089820912 startingPage "5938" @default.
• W3089820912 abstract "Convolutional neural networks (CNNs) have proven to be a powerful tool for the classification of hyperspectral images (HSIs). The CNN kernels are able to naturally include spatial information to smooth out the spectral variability and the noise present in HSI data. However, these kernels are composed of a large number of learning parameters that must be correctly adjusted to achieve good performance. This forces the model to consume a large amount of training data, being prone to overfitting when limited labeled samples are available. In addition, the execution of kernels is computationally very expensive, increasing quadratically with respect to the size of the convolution filter. This significantly reduces the performance of the model. To overcome the aforementioned limitations, this work presents a new few-parameter CNN (based on shift operations) for HSI classification that dramatically reduces both the number of parameters and the computational complexity of the model in terms of floating-point operations (FLOPs). The operational module combines a shift kernel (which adjusts the input data in particular directions without involving any parameters nor FLOPs) with pointwise convolutions that perform the feature extraction stage. The newly developed shift-based CNN has been employed to conduct HSI classification over five widely used and challenging data sets, achieving very promising results in terms of computational performance and classification accuracy." @default.
• W3089820912 created "2020-10-08" @default.
• W3089820912 creator A5010624980 @default.
• W3089820912 creator A5033046838 @default.
• W3089820912 creator A5039673511 @default.
• W3089820912 creator A5046123228 @default.
• W3089820912 creator A5054292278 @default.
• W3089820912 date "2021-07-01" @default.
• W3089820912 modified "2023-10-02" @default.
• W3089820912 title "FLOP-Reduction Through Memory Allocations Within CNN for Hyperspectral Image Classification" @default.
• W3089820912 cites W1840106123 @default.
• W3089820912 cites W1932847118 @default.
• W3089820912 cites W1969733464 @default.
• W3089820912 cites W2001068000 @default.
• W3089820912 cites W2001298023 @default.
• W3089820912 cites W2006175358 @default.
• W3089820912 cites W2010797000 @default.
• W3089820912 cites W2012273471 @default.
• W3089820912 cites W2067532478 @default.
• W3089820912 cites W2081562328 @default.
• W3089820912 cites W2085625911 @default.
• W3089820912 cites W2087263574 @default.
• W3089820912 cites W2092071303 @default.
• W3089820912 cites W2095687521 @default.
• W3089820912 cites W2098057602 @default.
• W3089820912 cites W2101711129 @default.
• W3089820912 cites W2103734061 @default.
• W3089820912 cites W2118020653 @default.
• W3089820912 cites W2121494034 @default.
• W3089820912 cites W2136251662 @default.
• W3089820912 cites W2136656942 @default.
• W3089820912 cites W2161742217 @default.
• W3089820912 cites W2167583754 @default.
• W3089820912 cites W2218047931 @default.
• W3089820912 cites W2417303407 @default.
• W3089820912 cites W2519653196 @default.
• W3089820912 cites W2538244214 @default.
• W3089820912 cites W2548776929 @default.
• W3089820912 cites W2549139847 @default.
• W3089820912 cites W2550848904 @default.
• W3089820912 cites W2764276316 @default.
• W3089820912 cites W2767805377 @default.
• W3089820912 cites W2772452219 @default.
• W3089820912 cites W2782522152 @default.
• W3089820912 cites W2792827505 @default.
• W3089820912 cites W2793927960 @default.
• W3089820912 cites W2794284562 @default.
• W3089820912 cites W2809113079 @default.
• W3089820912 cites W2888119354 @default.
• W3089820912 cites W2892075618 @default.
• W3089820912 cites W2896847173 @default.
• W3089820912 cites W2898381489 @default.
• W3089820912 cites W2914331134 @default.
• W3089820912 cites W2919115771 @default.
• W3089820912 cites W2941141441 @default.
• W3089820912 cites W2963844898 @default.
• W3089820912 cites W2976477742 @default.
• W3089820912 cites W2977002487 @default.
• W3089820912 cites W2979785627 @default.
• W3089820912 cites W2982619380 @default.
• W3089820912 cites W2983255812 @default.
• W3089820912 cites W2991616716 @default.
• W3089820912 cites W3003552243 @default.
• W3089820912 cites W3103753223 @default.
• W3089820912 cites W4301109526 @default.
• W3089820912 cites W4375905827 @default.
• W3089820912 doi "https://doi.org/10.1109/tgrs.2020.3024730" @default.
• W3089820912 hasPublicationYear "2021" @default.
• W3089820912 type Work @default.
• W3089820912 sameAs 3089820912 @default.
• W3089820912 citedByCount "23" @default.
• W3089820912 countsByYear W30898209122021 @default.
• W3089820912 countsByYear W30898209122022 @default.
• W3089820912 countsByYear W30898209122023 @default.
• W3089820912 crossrefType "journal-article" @default.
• W3089820912 hasAuthorship W3089820912A5010624980 @default.
• W3089820912 hasAuthorship W3089820912A5033046838 @default.
• W3089820912 hasAuthorship W3089820912A5039673511 @default.
• W3089820912 hasAuthorship W3089820912A5046123228 @default.
• W3089820912 hasAuthorship W3089820912A5054292278 @default.
• W3089820912 hasBestOaLocation W30898209122 @default.
• W3089820912 hasConcept C111335779 @default.
• W3089820912 hasConcept C11413529 @default.
• W3089820912 hasConcept C114614502 @default.
• W3089820912 hasConcept C115961682 @default.
• W3089820912 hasConcept C134306372 @default.
• W3089820912 hasConcept C138885662 @default.
• W3089820912 hasConcept C153180895 @default.
• W3089820912 hasConcept C154945302 @default.
• W3089820912 hasConcept C159078339 @default.
• W3089820912 hasConcept C173608175 @default.
• W3089820912 hasConcept C179799912 @default.
• W3089820912 hasConcept C22019652 @default.
• W3089820912 hasConcept C2524010 @default.
• W3089820912 hasConcept C2776401178 @default.
• W3089820912 hasConcept C2777984123 @default.
• W3089820912 hasConcept C33923547 @default.
• W3089820912 hasConcept C3826847 @default.

• next