SYNTHESIS OF IMAGING SPECTROMETER AND MULTISPECTRAL SCANNER DATA OF DIFFERENT SPATIAL RESOLUTIONS

Abstract

The results of modeling the procedure of synthesizing data from an imaging spectrometer (IS) and a multispectral scanner (MSS) with low and high spatial resolutions, respectively, are presented. The synthesis method makes it possible to develop a classification and determine the spatial distribution of the sensed features at the high spatial resolution of MSS imagery and to retrieve their spectra with the spectroradiometric detail of IS measurements. It is shown that with an IS signal-to-noise ratio of no less than 50, a geometric matching of IS and MSS data with an accuracy of no less than 0.1 of an IS pixel along both axes, and with allowance for the IS real point-spread function, the spectral nonuniformity of the classes and the errors in retrieving the spectra fall within the range of several percentage points, with a ratio of the IS and MSS spatial resolutions of 4-8 and ∼10% with a ratio of resolutions of 16-32.