Matches in SemOpenAlex for { <https://semopenalex.org/work/W2970085166> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W2970085166 endingPage "122795" @default.
- W2970085166 startingPage "122784" @default.
- W2970085166 abstract "In recent years, convolutional neural networks (CNNs) have made great achievements in object extraction from very high-resolution (VHR) images. However, most existing approaches require large quantities of clean and accurate training data to achieve impressive classification results. The presence of inaccurate labels in training datasets is known to deteriorate the performance of CNNs. In this paper, we introduce a novel efficient method for improving the robustness when training CNN on the dataset with relatively noisy labels. First, we propose a feature and label noise model (FLNM) to model the noisy label distribution in the training dataset. Then, we use a multitask deep learning framework (MDLF) to integrate the FLNM into the training process of CNN. Finally, a novel loss function concerning the high-level features is introduced to efficiently train the MDLF. We evaluate our method on datasets from Massachusetts and compare this method with other state-of-the-art methods. The experimental results demonstrate the effectiveness of the proposed method in improving the classification performance of CNNs trained with noisy training dataset." @default.
- W2970085166 created "2019-09-05" @default.
- W2970085166 creator A5022469287 @default.
- W2970085166 creator A5038832233 @default.
- W2970085166 creator A5040320394 @default.
- W2970085166 creator A5044795752 @default.
- W2970085166 creator A5066321568 @default.
- W2970085166 creator A5069328331 @default.
- W2970085166 creator A5072511448 @default.
- W2970085166 creator A5074461919 @default.
- W2970085166 creator A5082859009 @default.
- W2970085166 date "2019-01-01" @default.
- W2970085166 modified "2023-10-18" @default.
- W2970085166 title "Object Extraction From Very High-Resolution Images Using a Convolutional Neural Network Based on a Noisy Large-Scale Dataset" @default.
- W2970085166 cites W2124219775 @default.
- W2970085166 cites W2287418003 @default.
- W2970085166 cites W2403681572 @default.
- W2970085166 cites W2548390752 @default.
- W2970085166 cites W2577784528 @default.
- W2970085166 cites W2589453516 @default.
- W2970085166 cites W2595964094 @default.
- W2970085166 cites W2615237590 @default.
- W2970085166 cites W2621060313 @default.
- W2970085166 cites W2747638294 @default.
- W2970085166 cites W2755226765 @default.
- W2970085166 cites W2771079624 @default.
- W2970085166 cites W2774320778 @default.
- W2970085166 cites W2889816407 @default.
- W2970085166 cites W2905332579 @default.
- W2970085166 cites W2964274690 @default.
- W2970085166 cites W2964292098 @default.
- W2970085166 cites W4247589477 @default.
- W2970085166 doi "https://doi.org/10.1109/access.2019.2938215" @default.
- W2970085166 hasPublicationYear "2019" @default.
- W2970085166 type Work @default.
- W2970085166 sameAs 2970085166 @default.
- W2970085166 citedByCount "8" @default.
- W2970085166 countsByYear W29700851662020 @default.
- W2970085166 countsByYear W29700851662021 @default.
- W2970085166 countsByYear W29700851662022 @default.
- W2970085166 crossrefType "journal-article" @default.
- W2970085166 hasAuthorship W2970085166A5022469287 @default.
- W2970085166 hasAuthorship W2970085166A5038832233 @default.
- W2970085166 hasAuthorship W2970085166A5040320394 @default.
- W2970085166 hasAuthorship W2970085166A5044795752 @default.
- W2970085166 hasAuthorship W2970085166A5066321568 @default.
- W2970085166 hasAuthorship W2970085166A5069328331 @default.
- W2970085166 hasAuthorship W2970085166A5072511448 @default.
- W2970085166 hasAuthorship W2970085166A5074461919 @default.
- W2970085166 hasAuthorship W2970085166A5082859009 @default.
- W2970085166 hasBestOaLocation W29700851661 @default.
- W2970085166 hasConcept C104317684 @default.
- W2970085166 hasConcept C108583219 @default.
- W2970085166 hasConcept C119857082 @default.
- W2970085166 hasConcept C124101348 @default.
- W2970085166 hasConcept C153180895 @default.
- W2970085166 hasConcept C154945302 @default.
- W2970085166 hasConcept C185592680 @default.
- W2970085166 hasConcept C41008148 @default.
- W2970085166 hasConcept C51632099 @default.
- W2970085166 hasConcept C52622490 @default.
- W2970085166 hasConcept C55493867 @default.
- W2970085166 hasConcept C63479239 @default.
- W2970085166 hasConcept C81363708 @default.
- W2970085166 hasConceptScore W2970085166C104317684 @default.
- W2970085166 hasConceptScore W2970085166C108583219 @default.
- W2970085166 hasConceptScore W2970085166C119857082 @default.
- W2970085166 hasConceptScore W2970085166C124101348 @default.
- W2970085166 hasConceptScore W2970085166C153180895 @default.
- W2970085166 hasConceptScore W2970085166C154945302 @default.
- W2970085166 hasConceptScore W2970085166C185592680 @default.
- W2970085166 hasConceptScore W2970085166C41008148 @default.
- W2970085166 hasConceptScore W2970085166C51632099 @default.
- W2970085166 hasConceptScore W2970085166C52622490 @default.
- W2970085166 hasConceptScore W2970085166C55493867 @default.
- W2970085166 hasConceptScore W2970085166C63479239 @default.
- W2970085166 hasConceptScore W2970085166C81363708 @default.
- W2970085166 hasLocation W29700851661 @default.
- W2970085166 hasOpenAccess W2970085166 @default.
- W2970085166 hasPrimaryLocation W29700851661 @default.
- W2970085166 hasRelatedWork W2059299633 @default.
- W2970085166 hasRelatedWork W2279398222 @default.
- W2970085166 hasRelatedWork W2732542196 @default.
- W2970085166 hasRelatedWork W2738221750 @default.
- W2970085166 hasRelatedWork W2773120646 @default.
- W2970085166 hasRelatedWork W3011074480 @default.
- W2970085166 hasRelatedWork W3156786002 @default.
- W2970085166 hasRelatedWork W4299822940 @default.
- W2970085166 hasRelatedWork W4311257506 @default.
- W2970085166 hasRelatedWork W4366492315 @default.
- W2970085166 hasVolume "7" @default.
- W2970085166 isParatext "false" @default.
- W2970085166 isRetracted "false" @default.
- W2970085166 magId "2970085166" @default.
- W2970085166 workType "article" @default.