Informações Principais
     Resumo
     Abstract
     Introdução
     Conclusão
     Download
  
  
  
 
Abstract
 
 
Acadêmico(a): Márcio Koch
Título: Visão Computacional para Reconhecimento de Faces aplicado na Identificação e Autenticação de Usuários na Web
 
Abstract:
This paper presents the specification and implementation of how to perform identification and authentication of users through facial recognition applications, specifically web. The user faces are detected and captured from a Java applet loaded in the browser. Images of faces are processed and transferred via socket to a server where your information is extracted. In a new account information is stored in an Oracle database. In the case of identification, are compared against the data presented with data of known faces already stored in the database, the user is recognized, your data is returned to the applet that redirects to a welcome page. Virtually all processing, detection and identification of images of faces was performed using the API JavaCV which is an extension to the Java API OpenCV. The technique used for feature extraction and recognition of faces was the PCA.