Contourlet-1.3广义高斯模型纹理图像检索系统  

Contourlet-1.3 and Generalized Gaussian Model Texture Image Retrieval

作　　者：郭振[1] 陈新武[2] 

机构地区：[1]信阳师范学院计算机与信息技术学院,河南信阳464000 [2]信阳师范学院物理电子工程学院,河南信阳464000

出　　处：《信阳师范学院学报（自然科学版）》2014年第3期436-439,共4页Journal of Xinyang Normal University(Natural Science Edition)

基　　金：国家自然科学基金项目(61202194);河南省教育厅科学基金项目(12A510020)

摘　　要：为了提高基本轮廓波变换纹理图像检索系统的检索率,提出一种基于contourlet-1.3的纹理图像检索系统.该系统将广义高斯模型参数级联构造特征向量,采用Kullback-Leibler距离来衡量纹理图像之间的相似性.结果表明:在特征向量长度、检索时间、所需内存相同的情况下,contourlet-1.3广义高斯模型比同样架构的基本轮廓波变换检索系统具有较高的检索率.而且,在计算复杂度相当的情况下,广义高斯模型与KLD距离相结合,检索率高于能量特征和欧氏距离相结合的情况.In order to improve the retrieval rate of the original contourlet transform texture image retrieval system, a contourlet-1.3 transform based texture image retrieval method was proposed. Generalized Gaussian Density （GGD） model parameters were cascaded to form feature vectors and Kullback-Leibler distance （ KLD ） function was used for similarity measure. Experimental results indicated that contourlet-1.3-transform-based image retrieval system is superi- or to that of the original eontourlet transform under the same system structure with almost same length of feature vectors, retrieval time and memory needed. Furthermore, GGD combined with KLD method has higher retrieval rates than energy based features combined with Euclidean distance under comparable levels of computational complexity.

关 键 词：基于内容的图像检索 contourlet-1 3 轮廓波变换 纹理图像检索 广义高斯密度模型 K-L距离 检索率 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]