| 极化散射特性保持的改进迭代Wishart分类算法 |
| |
| 引用本文： | 巫兆聪,欧阳群东,孙轩,邹滨.极化散射特性保持的改进迭代Wishart分类算法[J].测绘科学,2011,36(6):161-163. |
| |
| 作者姓名： | 巫兆聪 欧阳群东 孙轩 邹滨 |
| |
| 作者单位： | 1. 武汉大学遥感信息工程学院,武汉,430079
2. 中南大学信息物理工程学院,长沙,410083 |
| |
| 基金项目： | 国家863计划资助项目(2007AA12Z143);国家自然科学基金资助项目(40201039,40771157) |
| |
| 摘 要： | 为克服基于极化散射特性保持的迭代Wishart分类算法不适用于城区及对混合散射像素分类欠理想等不足,本文提出一种改进方法.其基本思想是先应用四分量分解算法将像素分成4种基本散射类型和混合散射类型,接着以平均合并度为指导对基本散射类型中的像素自适应聚类,最后对所有像素进行散射特性保持的迭代Wishart分类.试验结果表明...
|
| 关 键 词： | 极化SAR 散射特性保持 非监督分类 四分量散射模型 |
Improved iterative Wishart classification algorithm based on polarimetric scattering characteristics preservation |
| |
| WU Zhuo-cong,OUYANG Qun-dong,Sun Xuan,Zou Bin.Improved iterative Wishart classification algorithm based on polarimetric scattering characteristics preservation[J].Science of Surveying and Mapping,2011,36(6):161-163. |
| |
| Authors: | WU Zhuo-cong OUYANG Qun-dong Sun Xuan Zou Bin |
| |
| Institution: | ②(①School of Remote Sensing and Information Engineering,Wuhan University,Wuhan 430079,China;②School of Info-Physics and Geomatics Engineering,Central South University,Changsha 410083,China) |
| |
| Abstract: | To overcome the deficiencies of traditional iterative Wishart classification algorithm based on polarimetric scattering characteristics preservation,especially its inadaptability in urban area and insufficiency for classifying mixed scattering pixels,a novel improved approach was proposed in this paper.This improved classification algorithm could be implemented through three following steps.First,all pixels were divided into four fundamental scattering categories and a mixed scattering category using the four-component decomposition algorithm.Then,pixels in fundamental scattering categories were clustered adaptively in terms of the average merging measurement.Finally,Wishart classification operation which can preserve polarimetric scattering characteristics was utilized iteratively on all pixels.Experimental results showed that the proposed method could perform better in adaptability and classification than traditional one. |
| |
| Keywords: | polarimetric Synthetic Aperture Radar scattering characteristics preservation unsupervised classification four-component scattering model |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
