基于映射的复轮廓波变换纹理图像检索系统

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (12): 2982-2987.

基于映射的复轮廓波变换纹理图像检索系统

陈新武1,2 , 龚俊斌2, 刘玮2, 田金文2

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，湖北武汉 430074

出版日期:2009-12-24 发布日期:2010-01-03

Mapping based complex contourlet transform texture image retrieval system

CHEN Xin-wu1,2, GONG Jun-bin2, LIU Wei2, TIAN Jin-wen2

1. Coll. of Physics and Electronic Engineering, Xinyang Normal Univ., Xinyang 464000, China; 2. State Key Lab. for MultiSpectral Information Processing Technology, Huazhong Univ. of Science and Technology, Wuhan 430074, China

Online:2009-12-24 Published:2010-01-03

摘要/Abstract

摘要：

轮廓波变换纹理检索系统检索率比较低的根本原因在于轮廓波变换域系数的振荡性和移变敏感性。为了克服轮廓波变换的这些缺陷，提出了一种基于映射的复轮廓波变换。在该变换的基础上，采用变换域子带系数的能量和标准偏差序列作为特征向量，以Canberra距离为相似度度量标准，构造了一种纹理图像检索系统。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，基于映射的复轮廓波变换检索系统比轮廓波变换检索系统具有更高的检索率。

Abstract:

The ultimate reason of low retrieval rate of contourlet transform texture image retrieval systems lies in the coefficients oscillating and shift sensitive characters. In order to overcome the defects of contourlet transforms, a mapping based contourlet transform is proposed, and a texture image retrieval system based on the new transform is proposed in which the feature vectors are formed by cascading the energy and standard deviation of each subband in the mapping based contourlet domain, and a Canberra distance is used as similarity metric. Experimental results show that the image retrieval system is superior to that of the original contourlet transform with almost same length of feature vectors, retrieval time and memory.

陈新武1,2 , 龚俊斌2, 刘玮2, 田金文2. 基于映射的复轮廓波变换纹理图像检索系统[J]. Journal of Systems Engineering and Electronics, 2009, 31(12): 2982-2987.

CHEN Xin-wu1,2, GONG Jun-bin2, LIU Wei2, TIAN Jin-wen2. Mapping based complex contourlet transform texture image retrieval system[J]. Journal of Systems Engineering and Electronics, 2009, 31(12): 2982-2987.