利用Contourlet变换和谱直方图进行多源遥感影像纹理特征检索  

Texture Retrieval for Multi-source Remote Sensing Image Based on Contourlet Transform and Spectral Histogram

作　　者：邵振峰[1] 朱先强[1] 张斯斯[2] 

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,武汉市珞喻路129号430079 [2]华中科技大学电子与信息工程系,武汉市珞喻路1037号430074

出　　处：《武汉大学学报（信息科学版）》2010年第6期723-726,共4页Geomatics and Information Science of Wuhan University

基　　金：国家973计划资助项目(2010CB731800);国家863计划资助项目(2009AA121404);国家自然科学基金资助项目(40801165);测绘遥感信息工程国家重点实验室专项科研经费资助项目;武汉大学优秀博士学位论文培育基金资助项目;武汉大学2008年博士研究生(含1+4)自主科研项目

摘　　要：将Contourlet变换应用于多源遥感影像多尺度多方向纹理特征提取,分别采用Contourlet变换低频子带谱直方图和多尺度多方向高频子带谱直方图实现多源遥感影像库的初步过滤和精确检索。通过USC和QuickBird影像组成的纹理检索实验表明,所提出的检索算法在检索精度和效率方面优于小波变换。A two-step progressive texture retrieval algorithm for multi-source remotely sensed imagery based on Contourlet transform and spectral histogram is proposed.Firstly,the Contourlet transform is applied to decompose texture features of remotely sensed imagery at different scales and different directions.Then,low-pass signal and high-pass signals are adopted to realize elementary retrieval and exact retrieval relatively.The experiments on texture database obtained from USC and QucikBird images show that the proposed algorithm not only utilizes the advantages of Contourlet on multi-scale and multi-direction texture feature extraction,but also makes full use of the efficiency of spectral histogram on distributed statistical feature representation.On the aspect of precision and efficiency,Contourlet transform outperforms wavelet transform.The proposed method provides a powerful tool for texture retrieval of remote sensing image.

关 键 词：CONTOURLET变换 谱直方图 遥感影像 纹理检索 

分 类 号：P237.4[天文地球—摄影测量与遥感] TP753[天文地球—测绘科学与技术]