Discriminant models for crop identification of complex agricultural landscapes

This study presents a discriminant model for crop identification, which follows the pattern and process rules in Geosciences and is found by the variables closely related to crop growth. On this basis, a method of classification based on probability calibration algorithm is developed and compared with other two supervised classifiers. Spectral analysis was carried out to extract the dominant impact factors for crop mapping. Accuracy assessment and kappa analysis were used to evaluate the identification results. Statistical test procedure was applied to measure the differences between classification maps and the reference data based on individual categories. Through experiments using data sets from a region in south China, it was found that the proposal method can be used for identifying crop types for complex agricultural landscapes.