摘要
主成份分析是建立在统计特征基础上的多维(如多波段)正交线性变换。它是遥感图像处理中最常用也是最有用的变换算法之一。本文研究了主成份分析的原理、几何解释与计算过程,并用遥感影像和数据加以说明。
Principal Component Analysis(PCA) is multi-dimensional (such as muhispectral) orthogonal linear transformation based on multivariate statistical analysis. It is one of the most important and the most frequent using transformation algorithms in remote sensing digital images processing. The article studies the theory,geometric explanation and the process of calculation of PCA,and uses remote sensing images and concrete data to explain for being clear.
出处
《测绘与空间地理信息》
2006年第5期56-59,共4页
Geomatics & Spatial Information Technology
关键词
遥感图像处理
主成份分析
特征值
特征向量
remote sensing images processing
Principal Component Analysis
eigenvalue
eigenvector