顾及权重的点群目标自动综合算法

An automatic generalization algorithm of point clusters considering weight

针对已有算法对点的权重的确定缺乏充足的理论依据的问题,该文提出了一种顾及权重的点群目标自动综合算法。其原理是:首先,引入影响范围与影响人群两种因素作为点的权重指标,获取相关数据并对其进行表达;其次,在遵循地图综合的基本原则基础上,以影响范围多边形面积及影响人群数量为依据,利用归一化及“同心圆”方法实现了点的取舍操作;最后,采用3种典型的点群数据,对该方法进行了实验与对比。实验结果表明,该算法不但增加了点群的语义信息,同时消除了局部密度对点群综合的影响,在点群权重的现势性方面有较为明显的优势,可适用于各种类型的点群综合。

Aiming at the problem that existing algorithms have poor theoretical foundation in evaluation of points'weight,an algorithm of point cluster generalization was proposed based on points' weight in this paper.The main idea of the algorithm is:first,two factors which are influence scope and influence population were proposed into the generalization as the weight indexes,and the corresponding data was obtained and visualized; second,the corresponding points were reserved or deleted by means of normalization and "concentric circles"based on the area of the influenced region and the number of influenced population on the premise of the basic principles which should be taken into account in map generalization;at last,three types of point cluster were selected for experiments to verify the algorithm.Experiments were done to verify the advantages of the method by comparing it with multi-weighted Voronoi diagram based algorithm using three kinds of point clusters.Results showed that the algorithm added semantic data to point cluster,and could eliminate the influence of local density in the process of generalization,and was suitable to meet with the selection of various kinds of point clusters.