Land-use classification using multitemporal ERS-1, Radarsat and JERS SAR-images

Land-use classification was performed by using a set of ERS-1, JERS- and Radarsat images. Classes were water, forests (with subclasses according to stem volume), agricultural field, mire and urban area. Median filtering was used for speckle reduction and principal component analysis for feature extraction. Spectral classification was performed by using self-organizing feature map and learning vector quantization. Contextual classification was performed as a post-processing step. The overall accuracy of the spectral classification was 86.4% and the best contextual classification 89.8%.