Multispectral image-processing with a three-layer backpropagation network

The authors report on the use of a three-layer backpropagation neural network to do a pixel-by-pixel classification of a multispectral image of the Earth's surface. Standard techniques are used to specify and train the network; however, the output slab is tailored to present the results of the pixel classification as standard three-color video display. Classification results are presented.<
>