| 基于ReliefF＋mRMR特征降维算法的多特征遥感图像分类 |
| |
| 引用本文： | 王露,;龚光红. 基于ReliefF＋mRMR特征降维算法的多特征遥感图像分类[J]. 中国体视学与图像分析, 2014, 0(3): 250-257 |
| |
| 作者姓名： | 王露, 龚光红 |
| |
| 作者单位： | [1]北京航空航天大学先进仿真技术航空重点实验室,北京100191; [2]北京航空航天大学自动化科学与电气工程学院,北京100191 |
| |
| 摘 要： | 遥感图像分类对地面背景红外辐射特性仿真具有重要作用，提取的特征的性能直接影响分类精度。本文以高分辨率遥感图像为研究对象，提出了一种结合ReliefF和mRMR算法的特征降维算法，首先，通过ReliefF算法计算出各特征的权重系数，对特征集进行加权；然后利用mRMR算法选出与类别具有最大相关性且相互之间具有最小冗余性的特征。实验采用提出的算法对原特征空间进行优化，然后基于优化后的特征空间进行遥感图像自动分类，结果表明此方法能较好的提高分类精度。
|
| 关 键 词： | 遥感图像 特征选择 分类 |
Multiple features remote sensing image classification based on combining ReliefF and mRMR |
| |
| Affiliation: | WANG Lu, GONG Guanghong( 1. Advanced 2. Department Simulation Technology of Automation Science Aviation Key Laboratory, and Electrical Engineering, Beihang University, Beihang University, Beijing Beijing 100191 100191, China ; China) |
| |
| Abstract: | Remote sensing image classification plays an important role in the simulation of Infrared terrain scene. The performance of selected features has a direct effect on the accuracy of classification. Selecting the high-resolution remote sensing image as the research object, a method of feature selection by combining ReliefF and mRMR is proposed, which can optimize the original feature set. The experimental results show that the method can largely improve the accuracy of the classification based on the optimized feature set. |
| |
| Keywords: | remote sensing image features selection classification |
| 本文献已被 维普 等数据库收录！ |
