Alternate neural network models as supervised classifiers for satellite data

Investigations on the effectivity of different neural network architectures, viz. number of hidden neurons, constrained neuronal connections (hierarchical network), and fuzzy aggregation based synaptic neuronal functions (fuzzy neural network) for satellite data classification are presented. Performance of networks trained with varied number of training sizes for classification in large spatial extensions are used as illustration through two case studies, viz. land use/land cover classification of Delhi Ridge and species classification of floral resources in Shimla and Chopal regions in India. The results have been compared with statistical methods.