An extraction method of urban ecological types based on object-oriented classification: A case study on Wuhan City

Abstract

Combining SPOT5 multispectral and panchromatic data of Wuhan city, object-oriented method of land cover classification is used to extract the urban ecological types, such as water, vegetation, bare land and impervious surface (including building, road, and other artificial surface) in this paper. The first fundamental step is image segmentation, and then object features that can distinguish each ecological type are selected to extract urban ecological types progressively in this study. After the classification, an accuracy assessment is carried out and the overall accuracy is 94.75%, which can meet the needs of subsequent analyses to Wuhan city. © 2013. The authors.