A dynamic learning neural network for remote sensing applications

The neural network learning process is to adjust the network weights to the selected training data. Based on the multilayer feedforward perceptron neural network, a dynamic learning algorithm (DL) is proposed. The presented learning algorithm makes use of Kalman filtering technique to update the network weights, in the sense that the stochastic characteristics of incoming data sets are implicitly incorporated into the network. The Kalman gains which represent the learning rates of the network weights are updated and calculated through the U-D factorization. By concatenating all of the network weights at each layer to form a long vector such that it can be updated without propagating back, the proposed algorithm improves in convergence substantially over the backpropagation (BP) learning algorithm. Numerical illustrations are carried out using two types of problems: multispectral image classification and surface parameter retrieval. Results indicate that the use of the Kalman filtering algorithm not only substantially improves the convergence rate in the learning stage, but also enhances the separability for problems with highly nonlinear boundaries, as compared to BP algorithm, suggesting that the proposed DL neural network provides a practical and efficient tool for remote sensing applications