|
|
| Acesso ao texto completo restrito à biblioteca da Embrapa Agricultura Digital. Para informações adicionais entre em contato com cnptia.biblioteca@embrapa.br. |
Registro Completo |
Biblioteca(s): |
Embrapa Agricultura Digital. |
Data corrente: |
04/12/2014 |
Data da última atualização: |
08/01/2020 |
Tipo da produção científica: |
Artigo em Periódico Indexado |
Autoria: |
CORDEIRO, R. L. F.; GUO, F.; HAVERKAMP, D. S.; HORNE, J. H.; HUGHES, E. K.; KIM, G.; ROMANI, L. A. S.; COLTRI, P. P.; SOUZA, T. T.; TRAINA, A. J. M.; TRAINA JÚNIOR, C.; FALOUTSOS, C. |
Afiliação: |
ROBSON L. F. CORDEIRO, USP; FAN GUO, Carnegie Mellon University; DONNA S. HAVERKAMP, Science Applications International Corporation, McLean; JAMES H. HORNE, Science Applications International Corporation, McLean; ELLEN K. HUGHES, Science Applications International Corporation, McLean; GUNHEE KIM, Carnegie Mellon University; LUCIANA ALVIM SANTOS ROMANI, CNPTIA; PRISCILA P. COLTRI, Unicamp; TAMIRES T. SOUZA, USP; AGMA J. M. TRAINA, USP; CAETANO TRAINA JÚNIOR, USP; CHRISTOS FALOUTSOS, Carnegie Mellon University. |
Título: |
QuMinS: fast and scalable querying, mining and summarizing multi-modal databases. |
Ano de publicação: |
2014 |
Fonte/Imprenta: |
Information Sciences, New York, v. 164, p. 211-229, Apr. 2014. |
DOI: |
10.1016/j.ins.2013.11.013 |
Idioma: |
Inglês |
Conteúdo: |
Given a large image set, in which very few images have labels, how to guess labels for the remaining majority? How to spot images that need brand new labels different from the predefined ones? How to summarize these data to route the user's attention to what really matters? Here we answer all these questions. Specifically, we propose QuMinS, a fast, scalable solution to two problems: (i) Low-labor labeling (LLL) - given an image set, very few images have labels, find the most appropriate labels for the rest; and (ii) Mining and attention routing - in the same setting, find clusters, the top-NO outlier images, and the NR images that best represent the data. Experiments on satellite images spanning up to 2.25 GB show that, contrasting to the state-of-the-art labeling techniques, QuMinS scales linearly on the data size, being up to 40 times faster than top competitors (GCap), still achieving better or equal accuracy, it spots images that potentially require unpredicted labels, and it works even with tiny initial label sets, i.e., nearly five examples. We also report a case study of our method?s practical usage to show that QuMinS is a viable tool for automatic coffee crop detection from remote sensing images. |
Palavras-Chave: |
Clusterização; Imagens de satélite; Low-labor labeling; Outlier detection; Query by example; Satellite imagery; Summarization. |
Thesaurus Nal: |
Cluster analysis. |
Categoria do assunto: |
X Pesquisa, Tecnologia e Engenharia |
Marc: |
LEADER 02265naa a2200361 a 4500 001 2001677 005 2020-01-08 008 2014 bl uuuu u00u1 u #d 024 7 $a10.1016/j.ins.2013.11.013$2DOI 100 1 $aCORDEIRO, R. L. F. 245 $aQuMinS$bfast and scalable querying, mining and summarizing multi-modal databases.$h[electronic resource] 260 $c2014 520 $aGiven a large image set, in which very few images have labels, how to guess labels for the remaining majority? How to spot images that need brand new labels different from the predefined ones? How to summarize these data to route the user's attention to what really matters? Here we answer all these questions. Specifically, we propose QuMinS, a fast, scalable solution to two problems: (i) Low-labor labeling (LLL) - given an image set, very few images have labels, find the most appropriate labels for the rest; and (ii) Mining and attention routing - in the same setting, find clusters, the top-NO outlier images, and the NR images that best represent the data. Experiments on satellite images spanning up to 2.25 GB show that, contrasting to the state-of-the-art labeling techniques, QuMinS scales linearly on the data size, being up to 40 times faster than top competitors (GCap), still achieving better or equal accuracy, it spots images that potentially require unpredicted labels, and it works even with tiny initial label sets, i.e., nearly five examples. We also report a case study of our method?s practical usage to show that QuMinS is a viable tool for automatic coffee crop detection from remote sensing images. 650 $aCluster analysis 653 $aClusterização 653 $aImagens de satélite 653 $aLow-labor labeling 653 $aOutlier detection 653 $aQuery by example 653 $aSatellite imagery 653 $aSummarization 700 1 $aGUO, F. 700 1 $aHAVERKAMP, D. S. 700 1 $aHORNE, J. H. 700 1 $aHUGHES, E. K. 700 1 $aKIM, G. 700 1 $aROMANI, L. A. S. 700 1 $aCOLTRI, P. P. 700 1 $aSOUZA, T. T. 700 1 $aTRAINA, A. J. M. 700 1 $aTRAINA JÚNIOR, C. 700 1 $aFALOUTSOS, C. 773 $tInformation Sciences, New York$gv. 164, p. 211-229, Apr. 2014.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Agricultura Digital (CNPTIA) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
Registros recuperados : 7 | |
4. | | PRADO, C.; PEREIRA, R.; DURRANT, L.; SCORZA JUNIOR, R. P.; PIUBELI, F.; BONFÁ, M. Fipronil Degradation in Soil by Enterobacter chengduensis Strain G2.8: Metabolic Perspective. Life, v. 13, n. 9, p. 1935, 2023.Tipo: Artigo em Periódico Indexado | Circulação/Nível: A - 2 |
Biblioteca(s): Embrapa Agropecuária Oeste. |
| |
5. | | MARTINEZ, C. O.; SILVA, C. M. M. de S.; FAY, E. F.; MAIA, A. de H. N.; ABAKERLI, R. B.; DURRANT, L. R. Degradation of the herbicide sulfentrazone in a Brazilian typic hapludox soil. Soil Biology and Biochemistry, v. 40, n. 4, p. 853-860, 2008.Tipo: Artigo em Periódico Indexado | Circulação/Nível: Internacional - A |
Biblioteca(s): Embrapa Meio Ambiente. |
| |
6. | | MARTINEZ, C. O.; SILVA, C. M. M. de S.; FAY, E. F.; ABAKERLI, R. B.; MAIA, A. de H. N.; DURRANT, L. R. The effects of moisture and temperature on the degradation of sulfentrazone. Geoderma, v. 147, n. 1, p. 56-62, 2008.Tipo: Artigo em Periódico Indexado | Circulação/Nível: Internacional - A |
Biblioteca(s): Embrapa Meio Ambiente. |
| |
Registros recuperados : 7 | |
|
Nenhum registro encontrado para a expressão de busca informada. |
|
|