Estratégias para delimitação de regiões de interesse na visualização de consultas por similaridades

• Autor(a) principal: Paiva, Claudio Eduardo
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/10876
• topic: Consulta por similaridade; Dados complexos; Espaços métricos; Visualização; Mapeamento multidimensional; Similarity query; Complex data; Metric spaces; Visualization; Multidimensional mapping; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO::BANCO DE DADOS

The increase in data volume that has been occurring in recent time has generated the need for new methods of storing and retrieving information not only for the well known conventional data but also for the complex data such as images. In an image database, content-based searches can be done by comparing image characteristics such as color, texture, and shape to find objects similar to a particular search image. The fact, which people can do visual analysis more efficiently than in other ways, makes data visualization an important ally in creating knowledge in many different areas. Therefore, this study aims at analysing ways of visualizing the results of queries similarity in complex data, limiting itself only to the region that the query is being carried out, mapping the data in three-dimensional space and allowing users to interact with them, by changing the visualisation and being favored by the human perception to improve the analysis and the understanding of the data.