BDBComp
Parceria:
SBC
Agrupamentos Nebulosos em Segmentação de Imagens Coloridas: Quantificação Histológica

Waldemar Bonventi Jr.

In this work, a method for segmentation of colored histological images and structures quantification, using fuzzy clustering, was developed. Since color distributions in colorspaces from real images form clusters with fuzzy limits, various sizes and orientations, the clustering technique was improved with a metric based upon a Mahalanobis distance. Image segmentation was performed by fuzzy sets built with obtained groups. Membership of their elements is used to quantify image structures. This method was successfully applied in segmentation of digital images of human liver tissues to quantify cirrhosis.

http://www.lbd.dcc.ufmg.br/colecoes/wvc/2008/0020.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