Active learning paradigms for CBIR systems based on optimum-path forest classification

dc.creatorda Silva, AT
dc.creatorFalcao, AX
dc.creatorMagalhaes, LP
dc.date2011
dc.dateDEC
dc.date2014-07-30T13:43:31Z
dc.date2015-11-26T16:33:32Z
dc.date2014-07-30T13:43:31Z
dc.date2015-11-26T16:33:32Z
dc.date.accessioned2018-03-28T23:15:26Z
dc.date.available2018-03-28T23:15:26Z
dc.identifierPattern Recognition. Elsevier Sci Ltd, v. 44, n. 12, n. 2971, n. 2978, 2011.
dc.identifier0031-3203
dc.identifierWOS:000292947000013
dc.identifier10.1016/j.patcog.2011.04.026
dc.identifierhttp://www.repositorio.unicamp.br/jspui/handle/REPOSIP/54161
dc.identifierhttp://repositorio.unicamp.br/jspui/handle/REPOSIP/54161
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/1270888
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionThis paper discusses methods for content-based image retrieval (CBIR) systems based on relevance feedback according to two active learning paradigms, named greedy and planned. In greedy methods, the system aims to return the most relevant images for a query at each iteration. in planned methods, the most informative images are returned during a few iterations and the most relevant ones are only presented afterward. In the past, we proposed a greedy approach based on optimum-path forest classification (OPF) and demonstrated its gain in effectiveness with respect to a planned method based on support-vector machines and another greedy approach based on multi-point query. In this work, we introduce a planned approach based on the OFF classifier and demonstrate its gain in effectiveness over all methods above using more image databases. In our tests, the most informative images are better obtained from images that are classified as relevant, which differs from the original definition. The results also indicate that both OPF-based methods require less user involvement (efficiency) to satisfy the user's expectation (effectiveness), and provide interactive response times. (C) 2011 Elsevier Ltd. All rights reserved.
dc.description44
dc.description12
dc.description2971
dc.description2978
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionCNPq [140968/2007-5]
dc.languageen
dc.publisherElsevier Sci Ltd
dc.publisherOxford
dc.publisherInglaterra
dc.relationPattern Recognition
dc.relationPattern Recognit.
dc.rightsfechado
dc.rightshttp://www.elsevier.com/about/open-access/open-access-policies/article-posting-policy
dc.sourceWeb of Science
dc.subjectContent-based image retrieval
dc.subjectRelevance feedback
dc.subjectOptimum-path forest classifiers
dc.subjectActive learning
dc.subjectImage pattern analysis
dc.titleActive learning paradigms for CBIR systems based on optimum-path forest classification
dc.typeArtículos de revistas


Este ítem pertenece a la siguiente institución