This paper is committed to explore object-oriented methods for the classification of Quickbird images, aiming to support future urban population estimates. The study area concerns the southern sector of São José dos Campos city, located in the State of São Paulo, Brazil. By means of a multi-resolution segmentation approach and a six-layer hierarchical classification network, homogeneous residential areas were identified in terms of density of occupation and building standards (single dwelling units or high-rise buildings). The classification network was built upon spectral, geometrical and topological features of the objects in each level of segmentation as well as upon their contextual and semantic interrelationships in-between the hierarchical levels. The final classification of homogeneous residential units was subject to validation, using an object-based Kappa statistics.
Este estudo tem como objetivo o desenvolvimento de uma metodologia de classificação automática de áreas urbanizadas contínuas e dispersas que seja replicável em diferentes regiões do Brasil. Com essa metodologia busca-se o aumento da exatidão do mapeamento bem como reduzir a subjetividade e o tempo empregado no procedimento. Para este fim, aplicou-se, usando o software Definiens, a classificação baseada em Objeto em imagem LANDSAT da região de Piracicaba, Limeira e Rio Claro, do estado de São Paulo, obtida em 2007. Este procedimento consiste na segmentação multiresolução das imagens e na classificação baseada na lógica fuzzy. Na avaliação dos resultados foram utilizadas imagens de alta resolução, disponíveis no Google Earth. O bom desempenho obtido na classificação automática da área de estudo (índice global de 0,94 e Kappa de 0,72) indica a viabilidade do método aplicado para outras áreas urbanizadas.
The objective of this study is to measure the socio-spatial inequalities in residential areas of São José dos Campos, based on an intra-urban analysis, using QuickBird-2 satellite data. The residential space was considered to connect economic, social and political dimensions, in order to verify the corresponding stratification in space. The stratification can be observed at the configuration of different residential areas, showing the social division of space. Initially the following elements were considered in the concrete space: Size of lots, Vegetation cover, Population density, Materials of roofs, Swimming pools, Shadow, Bare Soil, Organization of blocks and lots. In this work, the establishment of a correlation among spatial differentiation elements and the living space allows to understand both the social contents and the variation of social phenomena in space. The measurement of differentiation elements was possible with QuickBird-2 image data. The Object-based Image Analysis (OBIA) concept was used for the classification of differentiation elements, while the visual analysis and data crossing was considered for others. The results obtained present the distribution of these elements and the identification of patterns in residential areas, corresponding to different spatial occupation forms by social classes. This study shows that high resolution satellite images contribute with new and relevant information about the concrete space, at intra-urban scale and it opens new possibilities for the analysis of the inequality dimension: the socio-spatial differentiation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.