Implementation of Area Boundary Vectorization and Area Structure Information Extraction for Bianry Raster Data

This paper makes a study of the approaches to the area vectorization,basing on the raster data expressed by simple determinant and run-length coding.On the aspect of border vectorization,a set of approaches are presented,which can effectively determine the initial segment of the curve for tracking and the trend of the following tracking points,avoid the borderline crossing and keep borderlines smooth.The external or inner borderline of a complex area object is fixed on by computing the nesting relation among the closed borderlines.This technique can have application in raster-based spatial analysis,borderlines extracting for thematic features of remote sensing images and so on.It is of high computing efficiency and is easy to realize.