Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387259387> ?p ?o ?g. }
- W4387259387 endingPage "4801" @default.
- W4387259387 startingPage "4801" @default.
- W4387259387 abstract "Polarimetric synthetic aperture radar (PolSAR) image classification has been an important area of research due to its wide range of applications. Traditional machine learning methods were insufficient in achieving satisfactory results before the advent of deep learning. Results have significantly improved with the widespread use of deep learning in PolSAR image classification. However, the challenge of reconciling the complex-valued inputs of PolSAR images with the real-valued models of deep learning remains unsolved. Current complex-valued deep learning models treat complex numbers as two distinct real numbers, providing limited assistance in PolSAR image classification results. This paper proposes a novel, complex-valued deep learning approach for PolSAR image classification to address this issue. The approach includes amplitude-based max pooling, complex-valued nonlinear activation, and a cross-entropy loss function based on complex-valued probability. Amplitude-based max pooling reduces computational effort while preserving the most valuable complex-valued features. Complex-valued nonlinear activation maps feature into a high-dimensional complex-domain space, producing the most discriminative features. The complex-valued cross-entropy loss function computes the classification loss using the complex-valued model output and dataset labels, resulting in more accurate and robust classification results. The proposed method was applied to a shallow CNN, deep CNN, FCN, and SegNet, and its effectiveness was verified on three public datasets. The results showed that the method achieved optimal classification results on any model and dataset." @default.
- W4387259387 created "2023-10-03" @default.
- W4387259387 creator A5013719267 @default.
- W4387259387 creator A5020872982 @default.
- W4387259387 creator A5044139442 @default.
- W4387259387 date "2023-10-01" @default.
- W4387259387 modified "2023-10-09" @default.
- W4387259387 title "A New Architecture of a Complex-Valued Convolutional Neural Network for PolSAR Image Classification" @default.
- W4387259387 cites W1974024703 @default.
- W4387259387 cites W1985194020 @default.
- W4387259387 cites W1992824785 @default.
- W4387259387 cites W2008826820 @default.
- W4387259387 cites W2048578702 @default.
- W4387259387 cites W2077547750 @default.
- W4387259387 cites W2078985447 @default.
- W4387259387 cites W2097272115 @default.
- W4387259387 cites W2101690226 @default.
- W4387259387 cites W2105923215 @default.
- W4387259387 cites W2111275757 @default.
- W4387259387 cites W2126168665 @default.
- W4387259387 cites W2132012856 @default.
- W4387259387 cites W2133989913 @default.
- W4387259387 cites W2135858797 @default.
- W4387259387 cites W2140973073 @default.
- W4387259387 cites W2144554203 @default.
- W4387259387 cites W2170481611 @default.
- W4387259387 cites W2346618237 @default.
- W4387259387 cites W2559324447 @default.
- W4387259387 cites W2751433056 @default.
- W4387259387 cites W2754361766 @default.
- W4387259387 cites W2766299269 @default.
- W4387259387 cites W2767003110 @default.
- W4387259387 cites W2772082715 @default.
- W4387259387 cites W2804596375 @default.
- W4387259387 cites W2897760800 @default.
- W4387259387 cites W2904891876 @default.
- W4387259387 cites W2969258943 @default.
- W4387259387 cites W2979476850 @default.
- W4387259387 cites W2986743354 @default.
- W4387259387 cites W2989661148 @default.
- W4387259387 cites W2996553655 @default.
- W4387259387 cites W3000122775 @default.
- W4387259387 cites W3038084131 @default.
- W4387259387 cites W3094965608 @default.
- W4387259387 cites W3119928944 @default.
- W4387259387 cites W3124574862 @default.
- W4387259387 cites W3127159627 @default.
- W4387259387 cites W3203407718 @default.
- W4387259387 cites W4225145356 @default.
- W4387259387 cites W4285125492 @default.
- W4387259387 doi "https://doi.org/10.3390/rs15194801" @default.
- W4387259387 hasPublicationYear "2023" @default.
- W4387259387 type Work @default.
- W4387259387 citedByCount "0" @default.
- W4387259387 crossrefType "journal-article" @default.
- W4387259387 hasAuthorship W4387259387A5013719267 @default.
- W4387259387 hasAuthorship W4387259387A5020872982 @default.
- W4387259387 hasAuthorship W4387259387A5044139442 @default.
- W4387259387 hasBestOaLocation W43872593871 @default.
- W4387259387 hasConcept C106301342 @default.
- W4387259387 hasConcept C108583219 @default.
- W4387259387 hasConcept C115961682 @default.
- W4387259387 hasConcept C119857082 @default.
- W4387259387 hasConcept C121332964 @default.
- W4387259387 hasConcept C153180895 @default.
- W4387259387 hasConcept C154945302 @default.
- W4387259387 hasConcept C41008148 @default.
- W4387259387 hasConcept C62520636 @default.
- W4387259387 hasConcept C70437156 @default.
- W4387259387 hasConcept C75294576 @default.
- W4387259387 hasConcept C81363708 @default.
- W4387259387 hasConcept C83665646 @default.
- W4387259387 hasConcept C87360688 @default.
- W4387259387 hasConcept C97931131 @default.
- W4387259387 hasConceptScore W4387259387C106301342 @default.
- W4387259387 hasConceptScore W4387259387C108583219 @default.
- W4387259387 hasConceptScore W4387259387C115961682 @default.
- W4387259387 hasConceptScore W4387259387C119857082 @default.
- W4387259387 hasConceptScore W4387259387C121332964 @default.
- W4387259387 hasConceptScore W4387259387C153180895 @default.
- W4387259387 hasConceptScore W4387259387C154945302 @default.
- W4387259387 hasConceptScore W4387259387C41008148 @default.
- W4387259387 hasConceptScore W4387259387C62520636 @default.
- W4387259387 hasConceptScore W4387259387C70437156 @default.
- W4387259387 hasConceptScore W4387259387C75294576 @default.
- W4387259387 hasConceptScore W4387259387C81363708 @default.
- W4387259387 hasConceptScore W4387259387C83665646 @default.
- W4387259387 hasConceptScore W4387259387C87360688 @default.
- W4387259387 hasConceptScore W4387259387C97931131 @default.
- W4387259387 hasFunder F4320321001 @default.
- W4387259387 hasIssue "19" @default.
- W4387259387 hasLocation W43872593871 @default.
- W4387259387 hasOpenAccess W4387259387 @default.
- W4387259387 hasPrimaryLocation W43872593871 @default.
- W4387259387 hasRelatedWork W2010629964 @default.
- W4387259387 hasRelatedWork W2096743164 @default.
- W4387259387 hasRelatedWork W2156233651 @default.
- W4387259387 hasRelatedWork W2160692033 @default.