计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (13): 166-168.

• 数据库、信号与信息处理 • 上一篇    下一篇

一种面向扩展空间对象的密度聚类算法

黎韶光1,周巨锁2,3,谢玉波4,5,陈 琦6   

  1. 1.国土资源部 信息中心,北京 100812
    2.中国海洋大学 海洋地球科学学院,山东 青岛 266003
    3.国家海洋局 南海工程勘察中心,广州 510300
    4.武汉大学 测绘遥感信息工程国家重点实验室,武汉 430079
    5.华北计算技术研究所,北京 100083
    6.广州市国土资源和房屋管理局海珠区分局,广州 510245

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-05-01 发布日期:2011-05-01

Density-based clustering algorithm for extended spatial objects

LI Shaoguang1,ZHOU Jusuo2,3,XIE Yubo4,5,CHEN Qi6   

  1. 1.Information Center,Ministry of Land and Resources,Beijing 100812,China
    2.College of Marine Geosciences,Ocean University of China,Qingdao,Shandong 266003,China
    3.South China Sea Marine Engineering Surveying Center,State Oceanic Administration,Guangzhou 510300,China
    4.State Key Lab of Information Engineering in Surveying Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China
    5.North China Institute of Computing Technology,Beijing 100083,China
    6.Haizhu Sub-bureau of Land Resources and Housing Management of Guangzhou,Guangzhou 510245,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-01 Published:2011-05-01

摘要: 提出一个面向扩展空间对象的基于密度的空间聚类算法,对点、线和多边形等扩展空间对象进行聚类。在该算法中,通过空间对象的缓冲区统一计算各对象在其附近空间的密度值,并根据参数区分两类不同的空间聚类应用场景,从而实现对空间对象的分类。实验表明,算法能够较好实现对空间对象分类。

关键词: 空间聚类, 扩展空间对象, 缓冲区

Abstract: A density-based clustering algorithm for extended spatial objects is proposed to cluster extended objects such as points,polylines and polygons.In the algorithm,buffers of extended objects such as points,polylines and polygons are created to compute the density around spatial objects in an efficient way.Two scenarios which can be managed by parameters are presented according to the requirements of the applications.In the case study,the proposed method demonstrates good quality.

Key words: spatial clustering, extended spatial objects, buffer