一种基于线面包含关系的GML空间聚类算法

山东大学学报(工学版)

一种基于线面包含关系的GML空间聚类算法

张丽1, 吉根林1,2*

1. 南京师范大学计算机系, 南京 210097;
2. 南京师范大学虚拟地理环境教育部重点实验室, 南京 210046

收稿日期:2009-03-20 修回日期:1900-01-01 出版日期:2009-04-16 发布日期:2009-04-16
通讯作者: 吉根林

ZHANG Li1, JI Genlin1,2*

1. Department of Computer, Nanjing Normal University, Nanjing 210097, China;
2. Key Laboratory of Virtual Geographic Environment, Ministry of Education,
Nanjing Normal University, Nanjing 210046， China

Received:2009-03-20 Revised:1900-01-01 Online:2009-04-16 Published:2009-04-16

摘要/Abstract

摘要：

针对目前大多数空间聚类算法主要是针对关系数据且没有考虑空间拓扑关系相似性的问题,对基于空间拓扑关系的空间聚类方法进行研究.提出了一种基于线面包含关系的GML(geography markup language)空间聚类算法SCGML-LRI(spatial clustering in GML data based on lineregion inclusion relations).算法将GML文档中线面空间对象的包含关系作为空间对象相似性度量准则,并用CLOPE算法对空间对象进行聚类.实验结果表明:算法SCGML-LRI能实现GML数据的空间聚类,并具有较高的效率.

关键词: 空间聚类, 拓扑关系, 线面空间包含, GML

Abstract:

For solving the problem most spatial clustering algorithms deal with the relational data without consideration of the similarity of spatial topological relations. A method for spatial clustering based on spatial topological relations was put forth, and the algorithm SCGMLLRI for spatial clustering in GML data based on lineregion inclusion relations was proposed. This algorithm considered the inclusion relations between line and region spatial objects as the similarity measurement criteria. The CLOPE algorithm was used for clustering of spatial objects. The experimental results showed that algorithm SCGMLLRI was effective and efficient.

Key words: spatial clustering, topological relation, lineregion spatial inclusion, GML

中图分类号:

TP391

张丽1, 吉根林1,2.

一种基于线面包含关系的GML空间聚类算法

[J]. 山东大学学报(工学版), 2009, 39(2): 21-25.

ZHANG Li1, JI Genlin1,2*. [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(2): 21-25.

参考文献

多维度评价

Viewed

Full text

159

HTML			PDF

Just accepted	Online first	Issue	Just accepted	Online first	Issue
0	0	0	0	0	159

From	Others	local

Times	14	145
Rate	9%	91%

Abstract

451

Just accepted	Online first	Issue

0	0	451

	From	Others

	Times	451
	Rate	100%

Cited

Web of Science	Crossref	ScienceDirect	Search for Citations in Google Scholar >>


This page requires you have already subscribed to WoS.

Shared

Discussed