You are in the accessibility menu

Please use this identifier to cite or link to this item: http://acervodigital.unesp.br/handle/11449/71961
Title: 
Optimizing optimum-path forest classification for huge datasets
Author(s): 
Institution: 
  • Universidade Estadual Paulista (UNESP)
  • Universidade Estadual de Campinas (UNICAMP)
ISSN: 
1051-4651
Abstract: 
Traditional pattern recognition techniques can not handle the classification of large datasets with both efficiency and effectiveness. In this context, the Optimum-Path Forest (OPF) classifier was recently introduced, trying to achieve high recognition rates and low computational cost. Although OPF was much faster than Support Vector Machines for training, it was slightly slower for classification. In this paper, we present the Efficient OPF (EOPF), which is an enhanced and faster version of the traditional OPF, and validate it for the automatic recognition of white matter and gray matter in magnetic resonance images of the human brain. © 2010 IEEE.
Issue Date: 
18-Nov-2010
Citation: 
Proceedings - International Conference on Pattern Recognition, p. 4162-4165.
Time Duration: 
4162-4165
Keywords: 
  • Brain image classification
  • Optimum-Path forest
  • Supervised classification
  • Support Vector machines
  • Automatic recognition
  • Brain images
  • Computational costs
  • Data sets
  • Forest classification
  • Gray matter
  • Human brain
  • Large datasets
  • Magnetic resonance images
  • Pattern recognition techniques
  • Recognition rates
  • Support vector
  • White matter
  • Image analysis
  • Image classification
  • Magnetic resonance
  • Magnetic resonance imaging
  • Support vector machines
  • Classification (of information)
Source: 
http://dx.doi.org/10.1109/ICPR.2010.1012
URI: 
Access Rights: 
Acesso restrito
Type: 
outro
Source:
http://repositorio.unesp.br/handle/11449/71961
Appears in Collections:Artigos, TCCs, Teses e Dissertações da Unesp

There are no files associated with this item.
 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.