Polynomial expression for analysis of hyperspectral remote sensing data
暂无分享,去创建一个
Presents a new method to analyze the relation between canopy spectral reflectance and component spectral properties. The polynomial decomposition method differs from linear spectral unmixing because it takes into consideration the multiple scattering inside canopy. It is consistent with physical BRDF models and more flexible because it does not depend on certain assumption on canopy structure. This method is superior for some kinds of canopy whose structure is ambiguous between homogeneous and discrete. The output of the analysis is "angular-structural coefficients" which is possible to be related directly to canopy biophysical parameters.
[1] John F. Mustard,et al. Spectral unmixing , 2002, IEEE Signal Process. Mag..
[2] Robert F. Cromp,et al. Analyzing hyperspectral data with independent component analysis , 1998, Other Conferences.
[3] W. Verhoef. Light scattering by leaf layers with application to canopy reflectance modeling: The Scattering by Arbitrarily Inclined Leaves (SAIL) model , 1984 .
[4] Sylvain G. Leblanc,et al. A four-scale bidirectional reflectance model based on canopy architecture , 1997, IEEE Trans. Geosci. Remote. Sens..