摘要
利用线性解混合方法处理高光谱图像数据,需要获取存在于光谱图像中的纯光谱.目前的纯光谱提取方法都需要复杂的运算,并且都没有被证明具有普遍适用的特点.在特征空间对光谱图像中信息存在形式进行有效分析的基础上,提出基于特征空间分析和光谱相关制图法相结合的纯光谱提取方法(FSASCM),具有复杂度低、对大多数高光谱图像数据普遍适用的特点.
Abstract The analysis endmembers consisting complex effective analysis calculation and of hyp in hyper can not erspectral data using linear unmixing requires the determination of spectral data. All endmember determination methods at present need be proved universal for all the hyperspectral data. According to the analysis of the information presented in hyperspectral data,a new method based on feature space and spectral correlation mapping （FSASCM） was proposed. The method mentioned is simple and universal for almost all the hyperspectral data.
出处
《光子学报》
EI
CAS
CSCD
北大核心
2005年第9期1336-1339,共4页
Acta Photonica Sinica
关键词
纯光谱
高光谱图像
解混合
特征空间
光谱相关制图法
Endmember
Hyperspectral data
Unmixing
Feature space
Spectral correlation mapping
