Marcador fiducial colorido e recursivo para realidade aumentada: algoritmo de detecção e projeção

• Ano de defesa: 2018
• Autor(a) principal: Tybusch, Douglas
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Ciência da Computação; UFSM; Programa de Pós-Graduação em Ciência da Computação; Centro de Tecnologia
• Palavras-chave em Português: Marcadores fiduciais; Realidade aumentada; Detecção de cores; Visão computacional; Processamento de imagens; Marcadores fiduciais coloridos; Fiducial markers; Augmented reality; Color detection; Computer vision; Image processing; Colored fiducial markers; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.ufsm.br/handle/1/14065

Augmented Reality is a technological concept that has existed for several years, but it has been showing increasing popularity in recent years, due to the evolution of devices such as smartphones, where its mobile processors and cameras are constantly evolving. One of the main areas of Augmented Reality research is the detection of objects and elements, such as binary fiducial markers. The detection of these elements allows the overlay of virtual objects and the creation of virtual environments, allowing the interaction between the real and virtual scene for the user. It is presented in this work, a Colored and Recursive Fiducial Marker for Augmented Reality, it is shown the elaboration of its structure and layout, and also the development of its detection algorithm. As a result, the use of color in the marker has been successfully implemented, as well as the identification and generation of unique identifiers, based on its color structure and sequences. We also implemented a hierarchical layout that allows the detection of markers in situations of partial occlusion. The proposed algorithm presented satisfactory performance and detection accuracy, offering precision similar to classic binary algorithms, however, having performance increased by up to 37% faster.