一种改进的基于区域划分的综合索引方法

Improved generalization index method based on region partition

为解决现有基于区域划分的自动制图综合索引方法中层次划分简单固定和没有考虑综合数据的空间分布特点等问题,提出了一种改进的基于区域划分的综合索引方法。该方法的不同之处在于对数据进行了空间聚类和基于缓冲区的冲突探测,将需要综合的空间数据分为四个层次:第一层为整个综合区域;第二层为进行聚类和冲突探测后的区域划分;第三层是利用境界、水系和道路网分割所得的区域划分;第四层是单个目标。实验表明,该方法扩展了索引层次,有效地弥补了原方法的缺点。

To solve the problem that grade-partition of present automatic generalization index method based on region partition is simple and without considering spatial distribution, an improved generalization index method based on region partition is proposed. Such method, different from others, disposes the data to spatial clustering and conflict detecting based on buffer, and divides the spatial data which needs generalizing into four grades： the first is the whole generalization region; the second is region after clustering and conflict detecting; the third is region partition got by partitioning boundaries, water bank and road-network; the last is single object. Experiment results indicate that such method expands index grades and overcome the defect of the former method.