基于相关关系的图像分类和图像检索  

Image classification and retrieval based on correlation

作　　者：贾淑华[1] 李星野[1] 马琳琳[1] 姜兴乾[1] 

机构地区：[1]上海理工大学管理学院,上海200093

出　　处：《微计算机信息》2009年第15期294-296,共3页Control & Automation

摘　　要：提出了一种基于图像分类的图像检索算法。算法首先以图像间的相关系数作为距离进行聚类,并确定每类图像的中心图像。图像检索过程分为两步:首先寻找匹配程度最高的中心图像,然后在该中心图像所在类中寻找最佳匹配图像。由于分类和确定中心均可离线操作,所以该算法显著加快了图像检索速度。此外,本文还运用信息熵分析了分类的有效性。实验结果表明:算法有效。An efficient method which based on the image classification is being used to the image retrieval . The cross correlation among the images is being used as the distance to cluster ,and the central image of per classification is found . The image retrieval method consists of two phases ： （1） finding the central image which has the best matching in all the central images （2） finding the perfect matching image from the classification which has the central image . The image classification and the finding for the central image are the off-line works , so they accelerate the image retrieval . And the information entropy is also being used to prove the validity of the image classification . The experiment indicates that the method has advantage over other methods .

关 键 词：图像分类 图像检索 聚类分析 相关系数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]