Efficient deep learning of image denoising using patch complexity local divide and deep conquer

Cited 1 time in webofscience Cited 0 time in scopus
  • Hit : 131
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorHong, Inpyoko
dc.contributor.authorHwang, Youngbaeko
dc.contributor.authorKim, Daeyoungko
dc.date.accessioned2019-10-22T06:20:11Z-
dc.date.available2019-10-22T06:20:11Z-
dc.date.created2019-10-22-
dc.date.created2019-10-22-
dc.date.issued2019-12-
dc.identifier.citationPATTERN RECOGNITION, v.96-
dc.identifier.issn0031-3203-
dc.identifier.urihttp://hdl.handle.net/10203/268028-
dc.description.abstractImage denoising is a fundamental task in computer vision and image processing domain. In recent years, the task has been tackled with deep neural networks by learning the patterns of noises and image patches. However, because of the high diversity of natural image patches and noise distributions, a huge network with a large amount of training data is necessary to obtain a state-of-the-art performance. In this paper, we propose a novel ensemble strategy of exploiting multiple deep neural networks for efficient deep learning of image denoising. We divide the task of image denoising into several local subtasks according to the complexity of clean image patches and conquer each subtask using a network trained on its local space. Then, we combine the local subtasks at test time by applying the set of networks to each noisy patch as a weighted mixture, where the mixture weights are determined by the likelihood of each network for each noisy patch. Our methodology of using locally-learned networks based on patch complexity effectively decreases the diversity of image patches at each single network, and their adaptively-weighted mixture to the input combines the local subtasks efficiently. Extensive experimental results on Berkeley segmentation dataset and standard test images demonstrate that our strategy significantly boosts denoising performance in comparison to using a single network of the same total capacity. Furthermore, our method outperforms previous methods with much smaller training samples and trainable parameters, and so with much reduced time complexity both in training and running. (C) 2019 Elsevier Ltd. All rights reserved.-
dc.languageEnglish-
dc.publisherELSEVIER SCI LTD-
dc.titleEfficient deep learning of image denoising using patch complexity local divide and deep conquer-
dc.typeArticle-
dc.identifier.wosid000487569700034-
dc.identifier.scopusid2-s2.0-85068508784-
dc.type.rimsART-
dc.citation.volume96-
dc.citation.publicationnamePATTERN RECOGNITION-
dc.identifier.doi10.1016/j.patcog.2019.06.011-
dc.contributor.localauthorKim, Daeyoung-
dc.contributor.nonIdAuthorHwang, Youngbae-
dc.description.isOpenAccessN-
dc.type.journalArticleArticle-
dc.subject.keywordAuthorImage denoising-
dc.subject.keywordAuthorDeep neural networks-
dc.subject.keywordAuthorStacked denoising autoencoders-
dc.subject.keywordAuthorDivide and conquer-
dc.subject.keywordAuthorLocal experts-
dc.subject.keywordAuthorImage patch complexity-
dc.subject.keywordAuthorEnsemble selection-
dc.subject.keywordAuthorImage patch classification-
dc.subject.keywordPlusSPARSE-
dc.subject.keywordPlusREPRESENTATIONS-
dc.subject.keywordPlusFRAMEWORK-
dc.subject.keywordPlusMIXTURES-
dc.subject.keywordPlusMODEL-
Appears in Collection
CS-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 1 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0