Study On The Method Of Urban Impervious Surface Extraction By Remote Sensing And Spatiotemporal Variation

In recent years,the process of urbanization has intensified the expansion of impervious surfaces,and a large amount of green space has been reduced in some cities,which has seriously affected the urban climate,water cycle and carbon cycle.The urban heat island effect has become increasingly apparent,and urban waterlogging,water pollution and ecological degradation have become prominent.Therefore,the impervious surface is considered as one of the important indicators for evaluating the health and environmental quality of urban ecosystems,and has attracted widespread attention from scholars.So,it is particularly important to quickly and accurately obtain the information of urban impervious surfaces,and remote sensing images are the most important data source for rapid extraction of impervious surfaces.Due to the different imaging physical models and spatial resolutions of different types of remote sensing sensors,their impervious surface extraction methods differ significantly.Based on a large amount of literature analysis and experiments,this paper proposes two new impervious surface extraction methods based on project requirements,which are aimed at low and medium spatial resolutions and high spatial resolutions,respectively.The main research contents of this thesis are:(1)A new method for impervious surface extraction based on Landsat 8-OLI image multifeature integrationAiming at the problem of insufficient selection of end elements in the extraction of impervious surface by linear spectral decomposition model,the paper first integrated the normalized water body index(NDWI),normalized vegetation index(NDVI)and adjusted soiladjusted vegetation index(MSAVI);Secondly,make a mask of water body,vegetation and soil,and superimpose the three information;again,use the MNF transform to concentrate the original image information into the first three bands;finally,combine the mask file with the MNF transformed image,select the threshold value to extract Permeable surface information.Experiments show that this method can extract the impervious surface information in Landsat 8 images with high accuracy.(2)A method for extracting impervious surface based on GF-2 imageExtracting impervious surfaces based on high spatial resolution satellite imagery is a new problem,and there are few methods that can be used for reference.To this end,this paper constructs an efficient extraction method based on domestic GF-2 images.First,based on the PISI model,the paper analyzes the problems of the model in the extraction of water,vegetation and soil,and constructs a targeted calculation model.Second,it performs integrated logical operations on the four sub-models constructed to initially extract impervious water.Surface information;again,morphological filtering,thinning and other algorithms were used to denoise and improve the accuracy of impervious surface extraction;finally,four experimental areas in Lanzhou were selected for verification and analysis.The experimental results show that this method can effectively extract the impervious surface information in high-scoring images.(3)Research and analysis of the impervious surface changes in the main urban area of Lanzhou City from 2013 to 2018Based on Landsat 8-OLI images,the information about the impervious surface of the main urban area of Lanzhou from 2013 to 2018 was analyzed and researched.It was found that with the advancement of urbanization,the rate of change of impervious surfaces has also increased.The largest is Qilihe District,followed by Chengguan District,followed by Xigu District,and finally Anning District.