In order to obtain urban distribution, the decision tree classification technology is used to classify remote sensing image in the study area. Landsat TM image ,digital elevation model (DEM), and normalized water index, vegetation index and cultivated land index as well as image spectral statistical characteristics and space information characteristics are used.Through the threshold determination methods, decision tree is built to achieve a classification of six kinds of objects in the study area ,which are water body, artificial structure, bare land, cultivated land, forest land and grassland. Finally confusion matrix is used to evaluate classification results, the overall classification accuracy of the decision tree is 89.52%, the Kappa coefficient is 0.867.
To get the typical land cover information of urban area, present a method to get four typical land covers in study area by displaying three binary index images in RGB coordinate system without any algorithms. One scene Landsat TM image was used to calculate Normalized Difference Vegetation Index (NDVI), Negative Normalized Difference Vegetation Index (NNDVI) the paper present, and extract water information based on the spectral relationship of typical land objects. After compared the digital number of the typical land cover information of water, vegetation, impervious surface and soil in the three calculated layer, the four typical land cover information showed obvious differences. With carefully selecting appropriate threshold value for each index image, we obtained three binary images for water, vegetation and impervious surfaces. Then they were stacked to one image and assigned red to impervious surfaces, blue to water, and green to vegetation in a false color composite, and because the soil’s digital number was zero in the three binary images, it was shown black color automatically. Two hundred sample points were randomly selected for an accuracy assessment using high resolution ZY-3(China) image obtained at almost the same time as reference. The overall accuracy of the classification is 86% with the Kappa coefficient of 0.802. The result indicates that the method presented in this paper is feasible.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.