一种基于关键维的图像索引方法  

A Key Dimension Based Indexing for Image Retrieval

作　　者：蔡捷飞[1] 袁华[1] 

机构地区：[1]华南理工大学计算机科学与工程学院,广东广州510640

出　　处：《广西师范大学学报（自然科学版）》2009年第1期209-212,共4页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家863计划资助项目(2006AA01Z196);国家自然科学基金资助项目(60603022)

摘　　要：基于内容的图像检索(CB IR)技术中,表示图像内容的特征向量通常是高维向量,且待检索图像量相当大,所以索引问题一直是CB IR的一个研究难点和热点。提出一种基于关键维的图像索引方法,首先提取图像向量中的关键维,再在每一维上创建关键维的升序链表,最后,基于关键维的升序链表进行检索。实验结果表明,该种方法可以带来检索效率的提高,且与图像库的大小无关。One of the most challenging tasks in implement of content based image retrieval system is to overcome the degradation of performance when the number of both dimension and date is large. A new index structure which is called key dimension list （KDL） for high dimensional indexing is introduced. KDL method identifies the key dimensions of feature vector,then creates a dimension list for each dimension in feature space,and finally,calculates the similarity via the KDL with a low cost of CPU resource. Calculation of the similarity via the KDL can reduce the cost of CPU resource. Experimental result demonstrates that KDL method improves the performance in querying large image database and its efficiency is irrelevant with the size of database.

关 键 词：关键维 图像索引 高维 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]