Abstract
The classification maps are required for management and for the estimation of agricultural disaster compensation; however, those techniques have yet to be established. Some supervised learning models may allow accurate classification. In this study, the Random Forest (RF) classifier and the classification and regression tree (CART) were applied to evaluate the potential of multi-temporal ALOS/PALSAR HH polarization data for classification of crop type. Furthermore, comparisons of the two algorithms and three orbits including one type of descending and two type of ascending data were carried out. In the study area, beans, beet, grasslands, maize, potato and winter wheat were cultivated, and these crop types were classified using the data set acquired in 2010. The classification results of RF were superior to those of CART and the overall accuracies were 83.2% with the kappa statistics of 0.785. This work indicates the usefulness of regular monitoring of agricultural fields using the ALOS-2/PALSAR-2 system, which was launched in 2014.
