Matches in SemOpenAlex for { <https://semopenalex.org/work/W2983707810> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2983707810 abstract "Aiming at improving the classification performance with greatly reduced annotation cost, this paper presents an active deep learning approach for minimally-supervised PolSAR image classification, which integrates active learning and fine-tuning convolutional neural network (CNN) into a principled framework. Starting from a CNN trained using a very limited number of labeled pixels, we iteratively and actively select the most informative candidates for annotation, and incrementally fine-tune the CNN by incorporating the newly annotated pixels. Moreover, to boost the performance and robustness of the proposed method, we employ Markov random field to enforce label smoothness, and data augmentation technique to enlarge the training set. Extensive experiments demonstrated that our approach achieved state-of-the-art classification results with significantly reduced annotation cost." @default.
- W2983707810 created "2019-11-22" @default.
- W2983707810 creator A5000694997 @default.
- W2983707810 creator A5026635895 @default.
- W2983707810 creator A5039966032 @default.
- W2983707810 creator A5071461704 @default.
- W2983707810 creator A5074077175 @default.
- W2983707810 creator A5075938156 @default.
- W2983707810 creator A5087264354 @default.
- W2983707810 date "2019-07-01" @default.
- W2983707810 modified "2023-09-29" @default.
- W2983707810 title "An Active Deep Learning Approach for Minimally-Supervised Polsar Image Classification" @default.
- W2983707810 cites W2112130876 @default.
- W2983707810 cites W2345055757 @default.
- W2983707810 cites W2370317051 @default.
- W2983707810 cites W2559324447 @default.
- W2983707810 cites W2595144186 @default.
- W2983707810 cites W2696603875 @default.
- W2983707810 cites W2754361766 @default.
- W2983707810 cites W2897760800 @default.
- W2983707810 cites W3104795559 @default.
- W2983707810 doi "https://doi.org/10.1109/igarss.2019.8899214" @default.
- W2983707810 hasPublicationYear "2019" @default.
- W2983707810 type Work @default.
- W2983707810 sameAs 2983707810 @default.
- W2983707810 citedByCount "54" @default.
- W2983707810 countsByYear W29837078102019 @default.
- W2983707810 countsByYear W29837078102020 @default.
- W2983707810 countsByYear W29837078102021 @default.
- W2983707810 countsByYear W29837078102022 @default.
- W2983707810 countsByYear W29837078102023 @default.
- W2983707810 crossrefType "proceedings-article" @default.
- W2983707810 hasAuthorship W2983707810A5000694997 @default.
- W2983707810 hasAuthorship W2983707810A5026635895 @default.
- W2983707810 hasAuthorship W2983707810A5039966032 @default.
- W2983707810 hasAuthorship W2983707810A5071461704 @default.
- W2983707810 hasAuthorship W2983707810A5074077175 @default.
- W2983707810 hasAuthorship W2983707810A5075938156 @default.
- W2983707810 hasAuthorship W2983707810A5087264354 @default.
- W2983707810 hasConcept C104317684 @default.
- W2983707810 hasConcept C108583219 @default.
- W2983707810 hasConcept C115961682 @default.
- W2983707810 hasConcept C119857082 @default.
- W2983707810 hasConcept C124504099 @default.
- W2983707810 hasConcept C153180895 @default.
- W2983707810 hasConcept C154945302 @default.
- W2983707810 hasConcept C160633673 @default.
- W2983707810 hasConcept C185592680 @default.
- W2983707810 hasConcept C2776321320 @default.
- W2983707810 hasConcept C2778045648 @default.
- W2983707810 hasConcept C41008148 @default.
- W2983707810 hasConcept C55493867 @default.
- W2983707810 hasConcept C63479239 @default.
- W2983707810 hasConcept C75294576 @default.
- W2983707810 hasConcept C81363708 @default.
- W2983707810 hasConceptScore W2983707810C104317684 @default.
- W2983707810 hasConceptScore W2983707810C108583219 @default.
- W2983707810 hasConceptScore W2983707810C115961682 @default.
- W2983707810 hasConceptScore W2983707810C119857082 @default.
- W2983707810 hasConceptScore W2983707810C124504099 @default.
- W2983707810 hasConceptScore W2983707810C153180895 @default.
- W2983707810 hasConceptScore W2983707810C154945302 @default.
- W2983707810 hasConceptScore W2983707810C160633673 @default.
- W2983707810 hasConceptScore W2983707810C185592680 @default.
- W2983707810 hasConceptScore W2983707810C2776321320 @default.
- W2983707810 hasConceptScore W2983707810C2778045648 @default.
- W2983707810 hasConceptScore W2983707810C41008148 @default.
- W2983707810 hasConceptScore W2983707810C55493867 @default.
- W2983707810 hasConceptScore W2983707810C63479239 @default.
- W2983707810 hasConceptScore W2983707810C75294576 @default.
- W2983707810 hasConceptScore W2983707810C81363708 @default.
- W2983707810 hasLocation W29837078101 @default.
- W2983707810 hasOpenAccess W2983707810 @default.
- W2983707810 hasPrimaryLocation W29837078101 @default.
- W2983707810 hasRelatedWork W2470368200 @default.
- W2983707810 hasRelatedWork W2766604260 @default.
- W2983707810 hasRelatedWork W2912288872 @default.
- W2983707810 hasRelatedWork W2986507176 @default.
- W2983707810 hasRelatedWork W2996856019 @default.
- W2983707810 hasRelatedWork W3011074480 @default.
- W2983707810 hasRelatedWork W3018421652 @default.
- W2983707810 hasRelatedWork W3160224718 @default.
- W2983707810 hasRelatedWork W3160711233 @default.
- W2983707810 hasRelatedWork W4220996320 @default.
- W2983707810 isParatext "false" @default.
- W2983707810 isRetracted "false" @default.
- W2983707810 magId "2983707810" @default.
- W2983707810 workType "article" @default.