Study On Estimation Of Urban Impervious Surface Percentage Based On Medium And High Spatial Resolution Remote Sensing Images

Remote sensing image interpretation and information extraction have been a hot and difficult problem in the field of international remote sensing research.The study of urban land cover classification has been the focus and core of modern geoscientific,this study can not only provide an important reference for urban land planning,sustainable development policies,but also provide a data base for the rational use of urban land.The impervious surface is one of the most important components of urban land cover,the impervious surface percentage is an important indicator of the degree of urbanization and the changes of ecological environment in the region.At the same time,the change of impervious surface can objectively reflect the urbanization and urban expansion of a city.Since twenty-first century,the rapid development of remote sensing technology,more and more researchers at home and abroad research using remote sensing technology to carry out the research and application of impervious surface,such as the development of applied research in the field of city thematic mapping,city ecological environment monitoring and so on.Obtaining an accurate and reliable impervious surface percentage information of a city can provide accurate input parameters for these studies.This dissertation mainly focuses on the object-oriented urban impervious surface extraction based on fused digital orthophoto map and nDSM data and the sub-pixel impervious surface percentage estimation based on medium spatial resolution remote sensing images.The main research contents and conclusions are as follows:(1)The estimation method of impervious surface percentage based on medium resolution satellite images.Based on the medium spatial resolution images(Landsat 5)in Ludwigsburg Germany in 2010,the mixed pixel decomposition method of fully constrained least squares was used to estimate the sub-pixel level impervious surface percentage.By the estimation above,we got the impervious surface percentage in Ludwigsburg,Germany.Then high resolution remote sensing images were used to verify the accuracy of experiment's results,the results showed that,the average relative error between the estimated value and the true value of the impervious surface percentage was 12.00% and the correlation coefficient was 0.81,which proved the reliability of the estimation method above.In the meanwhile,the problem that the data in abundance map is usually less than zero or greater than one was solved in this research,and the problem that the high resolution remote sensing images is difficult to cover all the research area was solved.(2)The fine extraction method of urban impervious surface based on object-oriented high resolution remote sensing images.In this paper,Ludwigsburg Germany was chosen as the study area,its land cover was classified by the object-oriented image classification method based on the digital orthophoto map and nDSM data with the resolution of 0.09 m.During the process of classifying,based on spectral characteristics,texture features and nDSM elevation characteristics,support vector machine,random forest,rule classification and fuzzy membership function were used to classify the land cover,finally the classification accuracy was evaluated by the overall classification accuracy,Kappa coefficient and other criteria,the results showed that the overall classification accuracy of support vector machine,random forest,fuzzy membership function and rule classification was respectively 100.00%,99.05%,99.05% and 91.43%,Kappa coefficient was respectively 1.0000,0.9871,0.9871 and 0.8840.