Urban Land-Cover Change Detection through Sub-Pixel Imperviousness Mapping Using Remotely Sensed Data

摘要: We developed a Sub-pixel Imperviousness Change Detection (SICD) approach to detect urban land-cover changes using Landsat and high-resolution imagery. The sub-pixel percent imperviousness was mapped for two dates (09 March 1993 11 2001) over western Georgia regression tree algorithm. accuracy of the predicted reasonable based on comparison independent reference data. average absolute error between data 16.4 15.3 2001. correlation coefficient (r) 0.73 0.78 2001, respectively. Areas with significant increase (greater than 20 percent) in impervious surface from 2001 were mostly related known land-cover/land-use that occurred this area, suggesting spatial change an is useful indicator identifying extent, intensity, and, potentially, type changes. Compared other pixel-based change-detection methods (band differencing, rationing, vector, post-classification), information allow users quantify interpret their own definition. Such considered complementary products generated methods. In addition, procedure mapping objective repeatable, hence, can be used monitoring large geographic area. Potential applications limitations through study environmental studies are also discussed.