A hybrid algorithm of the BP-ANN/GA for classification of urban terrain surfaces with fused data of Landsat ETM+ and ERS-2 SAR

摘要

A hybrid algorithm of the back propagation artificial neural network and genetic algorithm (BP-ANN/GA) was developed to optimize initial weights and to enable fast convergence of the BP-ANN. This algorithm was applied to the classification of urban terrain surfaces with fused data of Landsat ETM + and ERS-2 SAR. An example of the Shanghai Pudong area using both Landsat ETM + and ERS-2 SAR image data demonstrates well the classification when applying our fusion algorithm.