摘要
植被光谱数据采集以3个尺度光谱为主:材料光谱、端元光谱、像元光谱。材料光谱通常是指在严格的实验室条件下测量的样本光谱。端元光谱指在野外测量得到的相对均一目标的光谱,对于遥感图像而言,端元指像元或亚像元级纯像元。像元光谱通常是指航空航天平台获取的遥感图像,以遥感像元尺度为观测目标的光谱。用植被几何光学模型和冠层辐射传输模型来描述材料光谱和端元光谱之间的关系。采用解析模型和计算机模拟模型,将端元光谱向上尺度转化到像元光谱。从遥感影像中提取端元及像元解混可以认为是从像元光谱向下尺度转化到端元光谱。遥感影像空间分辨率重采样方法可以实现像元光谱内多尺度转换。
Vegetation spectrum collection includes three-scale spectrum,which are material spectrum,endmember spectrum and pixel spectrum.Material spectrum usually refers to the spectral samples measurement under strict laboratory conditions. Endmember spectrum is the relative homogeneous object that is measured in the field,which refers to the pure pixel at pixel or sub-pixel level in remote sensing images.Pixel spectrum usually comes from remote sensing image at the aerospace platform,the observed object spectrum at pixel scale.The relationship between material spectrum and endmember spectrum is described by vegetation geometrical optics model and canopy radiative transfer model.Scaling-up of the endmember spectrum is transformed into the pixel spectrum by analytical model and computer simulation models.Endmember extracted from remote sensing images and pixel unmixing are scaling-down method that are converted from pixel spectrum to endmember spectrum.Resampling method can complete multi-scale spectrum conversion in remote sensing image pixel level.
出处
《遥感信息》
CSCD
2014年第1期108-113,120,共7页
Remote Sensing Information
基金
福建省科技厅重点项目(2011N0004)
福建省教育厅项目(JA12267
JA10298)
关键词
材料光谱
端元光谱
像元光谱
光谱尺度转换
material spectrum
endmember spectrum
pixel spectrum
spectral scaling-up
