Contourlet-1.3纹理图像检索系统  

Contourlet-1.3 texture image retrieval system

作　　者：陈新武[1,2] 李湘英[1] 龚俊斌[2] 田金文[2] 

机构地区：[1]信阳师范学院物理电子工程学院,河南信阳464000 [2]华中科技大学多谱信息处理技术国家重点实验室,武汉430074

出　　处：《计算机工程与应用》2011年第26期177-180,共4页Computer Engineering and Applications

基　　金：国家自然科学基金No.60572048/F010204~~

摘　　要：针对轮廓波变换纹理图像检索系统检索率较低的问题,提出了一种基于较强抗混叠能力的Contourlet-1.3纹理图像检索系统。该系统采用的Contourlet-1.3变换由抗混叠临界下采样拉普拉斯金字塔级联临界下采样方向滤波器构成,特征向量采用变换域子带系数的能量和标准偏差连接而成,以Canberra距离为相似度度量标准。实验结果表明:在特征向量长度、检索时间、所需存储空间基本相同的情况下,基于抗混叠Contourlet-1.3的检索系统比相同架构的轮廓波变换、无下采样轮廓波变换以及Contourlet-2.3检索系统具有更高的检索率;该检索系统中抗混叠Contourlet变换分解结构参数的选择对于检索率也有较大影响。Focusing on the low retrieval rate of contourlet texture image retrieval system,an anti-aliasing Contourlet-1.3 transform based texture image retrieval system is proposed.In the system,the Contourlet-1.3 transform is constructed by anti-aliasing critical subsampled Laplacian pyramid cascaded by critical subsampled directional filter banks,sub-bands energy and standard deviations in contourlet domain are cascaded to form feature vectors,and the similarity metric is Canberra distance.Experimental results show that anti-aliasing contourlet transform based image retrieval system is superior to those of the original contourlet transform,nun-subsampled contourlet system and Contourlet-2.3 under the same system structure with almost same length of feature vectors,retrieval time and memory needed;and anti-aliasing contourlet decomposition structure parameter can make significant effects on retrieval rates.

关 键 词：纹理图像 检索系统 Contourlet-1.3变换 检索率 

分 类 号：TN911.7[电子电信—通信与信息系统]