基于GIS的空间位置关系聚类研究与应用

聚类分析是空间数据挖掘的一种方法,聚类算法能从空间数据库中直接发现一些有用的聚类结构。为实现空间复杂地理对象的聚类分析,文中给出了GIS及空间聚类分析技术．介绍了一种基于GIS的空间位置关系聚类分析算法。该算法按照空间相邻关系,将空间相邻的空间目标聚类成一类。在具体应用实例中利用本聚类算法将空间上相邻的大片阵地地域连成整块地域,并去掉了不符合条件的小块地域,形成阵地聚类,取得了满意的效果．实现了能够发现任意形状、并满足特定约束条件的聚类。

The Research and Application of Spatial Position Relation Clustering Algorithm in GIS

Cluster analysis is a method of spatial data mining. Clustering algorithm can find some useful clustering structures directly from spatial data base. In order to realize the clustering of complicated spatial geographical object,the paper introduces GIS and spatial clustering analyzing,and expores a clustering algorithm in GIS.The algorithm clusters neighboring spatial objects to a cluster on the neighborhood relation in spatial.In the insatance of actual application,with the algorithm,large position terrain on the neighborhood relation in spatial is connected to a whole terrain,the small one which is not qualified is deleted,realizes position clusters,and gains content effect.The algorighm realizes the clusters which have arbitrary shapes and satisfy special constraints.