摘要
用非监督全约束最小二乘法对线性光谱混合模型进行了反演 ,通过获得各像元组分的面积比图像来达到对各像元分类的目的。将非监督全约束最小二乘法的分类结果与有限光谱混合分析法的分类结果进行对比 ,结果表明 ,无论从分类效果还是计算时间上看 ,前者都优于后者。
The abundance fractions of endmembers in an image pixel are estimated by unsupervised fully constrained least squares (UFCLS) based on the inversion of linear spectral mixture method. The results of the experiment show that the effects are good. Compared to CSMA method, UFCLS method is better in both the effects of classification and the consumption of computation time.
出处
《武汉大学学报(信息科学版)》
EI
CSCD
北大核心
2004年第7期615-618,共4页
Geomatics and Information Science of Wuhan University
基金
国家 973计划资助项目 ( 2 0 0 3CB415 2 0 5 )
关键词
遥感图像分类
线性光谱混合模型
最小二乘
classification of remote sensing image
linear spectral mixture model
least square