453 / 2019-02-27 22:36:16

A multi-index approach for classification of urban land use intensity based on remote sensing

Urban land use intensity,Spectral trait variation,Remote Sensing,NDVI,NDBI,GLCM

全体主题 > Remote Sensing

The diversities of land use types and living processes, particularly in urban areas, are strongly influenced by the types and intensities of urban inhabitants' activities, it will definitely reflect on the grounds/land in the end, resulting in varieties of urban land use intensity (U-LUI). In order to measure and quantify the U-LUI, we proposed a new method that using the spectral trait (ST) and spectral trait variation (STV) to characterize the degree of land using. We have carried out a study on RapidEye and Sentinel-2 images in Leipzig, Germany, and compared the results of different resolution data. Specifically, two categories of indicators have been proposed in the study: (1) spectral traits: elements that can be extracted from remote sensing images (eg. Vegetation, built-up area, water), including the normalized difference vegetation index (NDVI) and the normalized difference built-up index (NDBI); (2) spectral trait variation: Indicators that can reflect the spatial relationships and dynamic changes in a region, such as the gray-level co-occurrence matrix (GLCM), which considers the relation and separation between two pixels at a time in an image. Finally, the classification of LUI is carried out by the method of multi-index fusion. The results show that the accuracy of LUI quantification depends on the completeness of land use classification. GLCM not only provides rich spatial information, but also verifies the spatial relationship between patches and within patches within a period of time. Therefore, it is used to discuss the LUI of different land use types. In addition, since the study area is relatively stable in the process of urbanization, and the main human activities include vegetation management and urban planning, this method needs to be considered from multiple perspectives when applied to other types of cities. However, studying on STV is an important step to quantify and evaluate the urban ecosystem and functions.