BDBComp
Parceria:
SBC
Proposta de um Histograma Perceptual de cores como característica para recuperação de imagens baseada em conteúdo

Kátia Veloso SilvaAdilson Gonzaga

This work was developed with the intention of establishing a methodology for the classificationof the colors of digital images in perception colors to generate a feature vector that allows to retrieve images by their content from a database. In related works and studies analyzed the methodologies of grouping the diverse possible colors of an image do not allow making an association between the digital colors perceived for human beings. Studies show that the in majority of the cultures human beings associate with the color perception only eleven terms: red, yellow, blue, green, pink, brow,black, white, purple, orange and gray.Therefore, this work considers a methodology based on rules of the Fuzzy Logic, that allows to associate with all the possible colors of digital images one of the eleven defined cultural colors, thus creating a Perceptual Color Histogram. This will allow the generation of a feature vector for the Content-Based Images Retrieval in a Database.

http://www.lbd.dcc.ufmg.br/colecoes/wvc/2005/0021.pdf

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD