AREA ERROR ANALYSIS OF VECTOR TO RASTER CONVERSION OF AREAL FEATURE IN GIS

Abstract

The process of vector to raster conversion has information losses to a certain extent, and generates the error problems. The thesis discussed a few types of vector to raster conversion and some reasons of the errors, and systematically analyzed the progress that the error analysis methods have progressed in vector to raster conversion of areal feature, meanwhile analyzed the current existent key problems. The thesis put forward a new error analysis method, which made use of proportion component method to create structure raster data and carried on error analysis in vector to raster conversion of areal feature based structure raster data. Finally, the thesis took the vector to raster conversion of laud-use data as an example carried on the verification research. The result discovered that the normal analysis method underrated the error of vector to raster conversion, because none full considered "this reduce, simultaneity that increment" problem that errors were not distributed equably on space in the vector to raster conversion. The thesis suggested that we might adopt structure raster data to improve the error analysis in vector to raster conversion. The method can distinguish errors from the cause of "this reduce" and the cause of "that increment", and the error analysis is more overall, objective and accurate. In the meanwhile, we easily make visualization and create error map on the results of error analysis.