摘要
为了提高多光谱遥感图像的分类正确,提出了一种基于主成分分析(K-L变换)的分类方法。该方法先应用K-L变换对多波段遥感图像进行降维,提取最主要的三个成分合成假彩色图,然后利用BP神经网络对假彩色图进行监督分类。由于主成分之间是不相关的,增强了图象信息,降低了神经网络的计算量,提高了分类精度。实验结果证明,该算法分类精度优于传统分类方法,总正确率为88.5%,Kappa系数为0.862,因而具有实用价值。
In order to improve the classification accuracy of multi-spectral remote sensing image, this paper puts forward a new classification method based on principal component analysis. The method is consisted of two steps： reducing the dimensions of multispectral remote sensing image with principle component analysis and generating a new image by the three main components of the remote sensing image; performing supervised classification on the new image with BP neural network. The result indicates that this method is superior to traditional algorithms, and its overall accuracy and Kappa coefficient reach 88.5% and 0. 862.
出处
《测绘科学》
CSCD
北大核心
2009年第3期137-139,共3页
Science of Surveying and Mapping
关键词
K—L变换
BP神经网络
遥感图像
监督分类
principle component analysis
BP neural network
remote sensing images
supervised classification
