面向分类的高光谱遥感影像数据特性的研究

Classification Oriented Research on the Characters of Hyperspectral Remote Sensing Image Data

分析了高光谱遥感的特点、影像数据的应用现状,针对高光谱遥感影像处理、分类的现状及遇到的困难,对高光谱遥感影像数据的分类特性进行了深入的分析总结,对高光谱遥感影像分类的支持向量机方法进行了系统研究,并分析指出了支持向量机分类法的优越性,最后提出了需要进一步研究的问题。

This paper discusses the characters of hyperspectral remote sensing, the application of image data and the bottle problems in the area of hyperspectral image processing and classification methods. Then it presents a thorough investigation into the hyperspectral remote sensing image data and summarizes the high dimensional spectral dam＇s performance in terms of classification. In the third section, it makes a systematic research on the support vector machine （SVM） classification method and shows the advantages as well. Lastly, the paper puts forward several problems to be researched in future.