首页 | 本学科首页   官方微博 | 高级检索  
     

基于内容图像检索中的一种动态多维索引方法
引用本文:徐焕,林坤辉,周昌乐. 基于内容图像检索中的一种动态多维索引方法[J]. 计算机工程与应用, 2006, 42(23): 161-164
作者姓名:徐焕  林坤辉  周昌乐
作者单位:厦门大学软件学院,福建,厦门,361005;厦门大学信息科学与技术学院,福建,厦门,361005
基金项目:教育部面向21世纪教育振兴行动计划(985计划)
摘    要:多维索引技术是基于内容检索的图像数据库的关键技术。SR-tree和X-tree是目前比较成熟有效的多维索引技术。为了提高多维索引的性能,我们在分析SR-tree和X-tree的结构和性能的基础上,针对SR-tree分裂算法的不足,引入X-tree中超级节点的思想,通过改进插入和分裂算法,设计了一种新的多维索引结构ESR-tree,即ExtendedSR-tree。实验表明,随着数据量和维数的增多,ESR-tree的性能明显优于SR-tree和X-tree。

关 键 词:基于内容检索  多维索引  超球体  超矩形
文章编号:1002-8331-(2006)23-0161-04
收稿时间:2005-11-01
修稿时间:2005-11-01

A Dynamic Multi-dimension Index Method in Content-based Image Retrieval
Xu Huan,Lin Kunhui,Zhou Changle. A Dynamic Multi-dimension Index Method in Content-based Image Retrieval[J]. Computer Engineering and Applications, 2006, 42(23): 161-164
Authors:Xu Huan  Lin Kunhui  Zhou Changle
Affiliation:1 Software Department of Xiamen University, Xiamen, Fujian 361005; 2 Information Science and Technology Department of Xiamen University, Xiamen, Fujian 361005
Abstract:Multidimensional indexing technology is the key technology of content-based retrieval in image database.SR-tree and X-tree are riper and more effective multidimensional indexing technology at present.In order to improve the performance of the indexing,we assay the structure and capability of SR-tree and X-tree.In view of the deficiencies of split algorithm in SR-tree,the paper designs a new multidimensional indexing structure ESR-Tree,that is Extended SR-tree,which introduces the idea of super node of X-tree and improves the insert and split algorithm.The results of the experimentation show that the capability of ESR-Tree is better than SR-tree and X-tree with the increasing data number and dimensions.
Keywords:content-based retrieval   multidimensional indexing   super-sphere   super-rectangle
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号