相邻比例尺居民地匹配

A Multi-to-Multi Matching Algorithm Between Neighborhood Scale Settlement Data

首先以居民地要素为例,通过对多对多关系产生原因的分析,明确制图综合中典型化操作是多对多关系产生的主要根源。其次,从对多对多匹配的实例入手,根据邻近、轮廓规则、分布有序等特征,设计了多对多关系的探测和发现方法。最后,根据典型化算法整体性、结构化的原则,从群对象的对象结构、轮廓形状、面积和方位等特征入手,设计了相邻比例尺多对多匹配算法。实验表明,该方法符合人类的匹配认知习惯,能够有效发现并确认相邻比例尺中同名对象之间的多对多对应关系。

Detection of matching relationships in neighborhood scale spatial data is the keystone formulti-scale data propagation updates; however, this kind of multi-to-multi matching is the most com- plex and critical roadblock. Firstly, it is certain that most multi-to-multi relationships of neighbor- hood scale settlements come from typification in map generalization. Secondly, after analyzing settle- ment cluster characters such as adjacency, order, etc. , a new algorithm is proposed to solve the multi- to-multi matching problem. Finally, detailed steps based on a comparison of similarities in shape, structure, area and location of settlement cluster are put into effect. Tests illustrate that the proposed algorithm neatly accords with human cognition habits for settlement matching and is effective for de- tection and identification of multi-to-multi relationships between neighborhood scale settlement data.