Optimal channels selection for underlying surface classification using orthogonal subspace projection algorithm

Based on orthogonal subspace projection algorithm, the algorithm was used to satellite optimum channels selection for underlying surface classification. The weight functions were obtained by analyzing target spectra distribution and background radiance, transmittance computed by MODTRAN. Satellite optimum bands for underlying surface classification were selected by using weight function spectra distribution, optimum index function and the usages channels of MODIS. Channels composing result shows that the algorithm is effective for underlying surface classification.