Linear mixing model based on optimization method for land cover mapping using LANDSAT ETM+ and SPOT-HRV data

The study was aimed at determining the relative proportions of ground cover components in a mixed pixel. A new estimator is introduced, which is based on optimization method and is used for the derivation of land cover classification maps. Fraction images were derived from the two reflective channels 3 and 4 of LANDSAT ETM+ and channels 2 and 3 of SPOT data. This research indicated that the use of optimization method might lead to improvements in mapping and analyses of land cover changes. It has been shown that the use of non-standard estimators can lead to significant increase in the accuracy of estimation.