基于direct LDA的高光谱遥感影像地物分类

Hyperspectral Remote Sensing Image Terrain Classification Based on Direct LDA

针对高光谱遥感影像的降维问题,提出一种高光谱影像地物分类方法:direct LDA子空间法。先采用直接线性判别分析(direct linear discriminant analysis,direct LDA)进行特征提取,然后在特征子空间中采用最短距离分类器进行地物分类。机载可见光/红外成像光谱仪(airborne visible/infrared imaging spectrometer,AVIRIS)的高光谱影像识别结果表明,该方法相比LDA子空间法和原空间法,可显著降低数据维数,提高识别率。

Hyperspectral remote sensing image has the problem of high dimensionality. A new hyperspectral image ter rain classification method, i. e. ,direct LDA subspace method, was presented. Firstly, direct linear discriminant analysis （direct LDA） was used to extract features in original high dimensional hyperspectral space, and then shortest distance classifier was used to perform terrain classification in the feature subspace. Recognition results based on airborne visi hie/infrared imaging spectrometer（AVIRIS） hyperspectral image show that the presented method can remarkably re- duce dat^t dimensionality and improve recognition efficiency.