Linear Feature Simplification Using Wavelets in GIS

초록

Feature Simplification is an essential method for multiple representations of spatial features in GIS. However, spatial features are various, complex and of a large size. Among spatial features which describe spatial infomation, linear feature is the most common. Therefore, an dfficient linear feature simplification method is most critical for spatial feature simplification in GIS. This paper proposes an original method, by which the problem of linear feature simplification is mapped intdo hte signal processing field. This method avoids conventional geometric computing in exising methods and exploits hte advantageous prperties of wavelet transform. Experimental results are presented to show htat the proposed method outperforms the sxisting method and achieves hte time complexity of o(n), where n is hte number of points of a linear feature. Furthemore, this method is not bound to two-dimension but can be extended to high-dimension spaec.