Please use this identifier to cite or link to this item:
http://acervodigital.unesp.br/handle/11449/135581
- Title:
- Swarm-based descriptor combination and its application for image classification
- Universidade Estadual Paulista (UNESP)
- 1577-5097
- Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
- Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
- Fundação para o Desenvolvimento da UNESP (FUNDUNESP)
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
- CNPq: 306580/2012-8
- CNPq: 484254/2012-0
- CNPq: 470571/2013-6
- CNPq: 303182/2011-3
- CNPq: 303673/2010-9
- CNPq: 479070/2013-0
- CNPq: 1260-12-0
- FAPESP: 2010/14910-0
- FAPESP: 2013/20387-7
- FAPESP: 2013/50169-1
- In this paper, we deal with the descriptor combination problem in image classification tasks. This problem refers to the definition of an appropriate combination of image content descriptors that characterize different visual properties, such as color, shape and texture. In this paper, we propose to model the descriptor combination as a swarm-based optimization problem, which finds out the set of parameters that maximizes the classification accuracy of the Optimum-Path Forest (OPF) classifier. In our model, a descriptor is seen as a pair composed of a feature extraction algorithm and a suitable distance function. Our strategy here is to combine distance scores defined by different descriptors, as well as to employ them to weight OPF edges, which connect samples in the feature space. An extensive evaluation of several swarm-based optimization techniques was performed. Experimental results have demonstrated the robustness of the proposed combination approach.
- 2014
- ELCVIA. Electronic letters on computer vision and image analysis, v. 13, n. 3, p. 13-27, 2014.
- 13-27
- Image classification
- Descriptor combination
- Swarm intelligence
- Optimum-path forest
- http://dx.doi.org/10.5565/rev/elcvia.566
- Acesso restrito
- outro
- http://repositorio.unesp.br/handle/11449/135581
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.