SciELO - Scientific Electronic Library Online

 
 número9Modelo de investigación, aplicado en el desarrollo de software. Caso de estudio en instituciones publicas de educación superior, Saltillo, Coahuila MéxicoUtilização de Algoritmos de Processamento e Análise de Imagem na Avaliação do Dano em Placas Compósitas índice de autoresíndice de assuntosPesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Journal

Artigo

Indicadores

Links relacionados

  • Não possue artigos similaresSimilares em SciELO

Compartilhar


Tékhne - Revista de Estudos Politécnicos

versão impressa ISSN 1645-9911

Resumo

CARVALHO, Fernando J. S.  e  TAVARES, João Manuel R. S.. Detecção de Faces em Imagens baseada na Identificação da Pele e dos Olhos. Tékhne [online]. 2008, n.9, pp.266-291. ISSN 1645-9911.

One of the main research areas in Computational Vision concerns the detection of faces in images. Among several existing methodologies, two are presented in this paper: one is based on skin detection and the other on deformable templates for eyes segmentation and their features extraction. The detection of skin regions allows the segmentation of faces in images through the identification of regions in which the pixels’ values are, in terms of probability, similar with the ones usually exhibited by the skin. However, since from this methodology does not result sufficient information to conclude the detection process with a high success rate, a deformable template for the eyes is used in a complementary way. This template, adequately positioned near small regions detected inside the skin segments previously identify, allows the detection of the eyes and consequently validates or not the existence of a face in the input image. This paper presents the description of the above mentioned methodologies as well as some experimental results obtained using Matlab.

Palavras-chave : Skin detection; Face detection; Template matching; Deformable template.

        · resumo em Português     · texto em Português     · Português ( pdf )