基于小波多尺度分析的综合特征图像检索  

Image Retrieval Based on Wavelet Using Multiple Features

作　　者：高燕[1] 胡学龙[2] 李玲[1] 丁宁[1] 

机构地区：[1]南京信息职业技术学院电子信息学院,江苏南京210046 [2]扬州大学信息工程学院,江苏扬州225009

出　　处：《微电子学与计算机》2010年第4期21-24,28,共5页Microelectronics & Computer

基　　金：江苏省图像处理与图像通信重点实验室资助项目(ZK206008)

摘　　要：以整数小波变换为基础,提出了一种综合颜色和空间特征的图像检索算法.该算法首先对图像做整数小波变换,提取其低频系数矩阵中环形区域的F-范数作为颜色特征;然后对低频分量分块并编码,利用马尔可夫链特性提取一步转移概率矩阵作为空间特征;再综合利用上述两个特征计算图像间的相似度,并进行彩色图像检索.通过对不同类型图像的检索对比实验结果表明,这种图像检索方法是行之有效的.According to the integer wavelet transform, a new method of content-based image retrieval is presented, which integrates color and spatial features. Firstly, the circular region F-norms in low frequency band of integer wavelet transform are extracted as color feature. Secondly, low frequency band of integer wavelet transform is segmented into blocks and encoded, the transition probability matrix utilizing characteristic of Markov chain is as spatial information. Finally, the similarities between color images are computed by using a combined index based on color feature and spatial feature. Through the comparative experiment on the image retrieval over various images, the results show that this representation of image is feasible and valid in image retrieval.

关 键 词：图像检索 整数小波变换 F-范数 一步转移概率矩阵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]