In this paper we describe a multi-temporal classification procedure for crops in LANDSAT scenes. The method involves the creation of crop signatures which characterize multi-spectral observations as functions of phenological growth states. The phenological signature models spectral reflectance explicitly as a function of crop maturity rather than a function of observation date. This means that instead of stacking spectral vectors of one observation on another, as is usually done for multi-temporal data, we establish for each possible crop category a correspondence of time to growth state which minimizes the smallest difference between the given multi-spectral multi-temporal vector and the category mean vector indexed by growth state. The results of applying this procedure to winter wheat show that the method is capable of discrimination with about the same degree of accuracy as more traditional multi-temporal classifiers. It shows some potential to label degree of maturity of the crop with out crop condition information in the training set.

Date of this Version