Identificação de pontos robustos em marcadores naturais e aplicação de metodologia baseada em aprendizagem situada no desenvolvimento de sistemas de realidade aumentada

• Ano de defesa: 2015
• Autor(a) principal: Forte, Cleberson Eugenio
• Orientador(a): Marengoni, Maurício
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Presbiteriana Mackenzie
• Palavras-chave em Português: visão computacional; pontos de interesse; realidade aumentada
• Palavras-chave em Inglês: computer vision; feature points; augmented reality
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: http://dspace.mackenzie.br/handle/10899/24264

In the past, the Augmented Reality (AR) required advanced technologies in special devices for interaction and visualization. Nowadays, with the emergence of the mobile devices it has become common the usage of these tools in the development of AR systems applied to various purposes, including education, using natural markers. As the quality of images captured by mobile devices has increased the number of detected feature points has increased also, which ultimately hampers, or even prevents the technique to be used in applications, which run in real time. In addition, it becomes clear the necessity of proposing methodologies to be used in the development of educational applications using AR systems in order to improve the user s experience as well as the longevity of these applications, adding elements based on educational theories. The technique presented in this work determine illumination robust feature points, in order to reduce the time required to match high-resolution images. Additionally, the research also provides a conceptual framework methodology that, based on situated learning theory, combines the educational and technological aspects related to the context of developing mobile AR applications. Based on the experiments, it is possible to say that the technique using robust feature points saves about 70% in the processing time for matching high resolution images.