BDBComp
Parceria:
SBC
MeSegHi: Método de Segmentação por Processamento Linear e Não-Linear de Imagens.

Luciana V. LouregaDaniela M. UshizimaGabrielle D. FreitasMarcos C. D'Ornellas

This article describes the development of the hybrid segmentation method, which uses two segmentation techniques: Mean-Shift (linear) and non-linear Watersheds for Image Foresting Transform (IFT). To implement the hybrid method it makes use of the grayscale reduced set. This set was achieved by the application of the Mean-Shift procedure as markers for IFT algorithm. Thus, the user can decide for the automatic or manual segmentation. Preliminary results present a potential application of this method in the pathology area: screening for cervical carcinoma cells using Papanicolau (Pap) smears.

http://www.lbd.dcc.ufmg.br/colecoes/wvc/2006/0054.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