Urban land use mapping with multi-spectral and SAR satellite datausing neural networks

摘要

Statistical, textural and Gabor features were extracted fromintegrated multitemporal multispectral TM data and ERS-1 SAR imagery forurban land use mapping. The computed features are first normalised usingthe SOM algorithm and then a decision tree algorithm is applied forfeature selection. The classification procedure was carried out with amultilayer perceptron, trained with the resilient backpropagationalgorithm. The authors' results demonstrate the potential of theproposed methodology