A neural network correlator for satellite imagery and ground truth data in a geographical information system

The authors describe results of research on using neural networks as a computational tool to capture the correlation between ground truth data and corresponding SPOT satellite imagery, and then incorporating the correlator within GRASS (Geographic Resources Analysis Support System), a geographical information system (GIS), to transform the whole imagery into land condition maps covering an entire installation. Correlators for any transect value can be built by directly training a multilayer feedforward network on the transect value data and the corresponding SPOT imagery. This approach is exemplified by the construction of a neural-network-based correlator for the transect value of percent land cover. The main benefits of this approach are that it uses the minimal information available, requires little input from the user, and is relatively easy to implement