Font Size: a A A

The Research On Land Gradation And Updae Of Standard Land Price Based On Grid Cells

Posted on:2009-12-05Degree:MasterType:Thesis
Country:ChinaCandidate:H ZhangFull Text:PDF
GTID:2189360245985159Subject:Physical geography
Abstract/Summary:PDF Full Text Request
From the angle of view in grading based on land price, theoretically and methodologically, this paper probed into the issues of land gradation and the update of standard land price in the cities of developed land market, and took the residential land of Fuzhou city as an example for possitive research. The mainly purpose of this study is for improving the research on spatial interpolation by using the land price data of samples and the research on land classifying.Firstly, based on the theory of geostatistics, the GIS assisted spatial analysis (ESDA) technology was employed to analysis the spatial variance order of the floor price, furthermore, the model of spatial interpolation was established. Secondly, based on the results of Kriging interpolation, the spatial distribution of residential land price was described. In addition, the unsupervised classification method for RS image was employed to classify the results of spatial interpolation, and the patch engendering in the progress of clustering was clumped by mathematical morphology method.The conclusions are as follows:(1) There are apparent quadratic trend of the floor price; the spatial variance of the price is isotropy; the approximation comparison between the semivariance functions and the validation results show that the Spherical model is the only one, with relative minor errors.(2) The region in which the floor price of residential land going beyond 3500 yuan distripute to the extent of Fuzhou city's downtown; and the region between 2000 to 3500 yuan in the Nantai island has a zonal distribution along the Mingjiang River. Basically, the price descends from the downtown to around.(3) By altering the parameter of "Maximum class stdv" from 250 to 750, the unsupervised classification result under ISODATA method show that the suitable number of class turn out to be 5. The results of K-means method confirm the conclusion.(4) By using the "Clump" fuction based on mathematical morphology in the Envi 4.2, the patch engendering from the progress of clustering was merged. It was shown that the approving results can be attained when the size of transform kernel was setting to 30.
Keywords/Search Tags:land gradation, update of standard land price, kriging interpolation, unsupervised classification
PDF Full Text Request
Related items