A neural-statistical approach to multitemporal and multisource remote-sensing image classification

A data fusion approach to the classification of multisource and multitempor al remote-sensing images is proposed. The method is based on the applicatio n of the Bayes rule for minimum error to the "compound" classification of p airs of multisource images acquired at two different dates. In particular, the fusion of multisource data is obtained by using multilayer perceptron n eural networks for a nonparametric estimation of posterior class probabilit ies. The temporal correlation between images is taken into account bg the p rior joint probabilities of classes at the two dates, As a novel contributi on of this paper, such joint probabilities are automatically estimated by a pplying a specific formulation of the expectation-maximization (ERI) algori thm to the data to be classified, Experiments carried out on a multisource and multitemporal data set confirmed the effectiveness of the proposed appr oach.