Summary form only given. A research effort was carried out to explore the use of neural networks in processing hyperspectral imagery for target detection and classification. Pixel registered imagery containing 32 spectral bands in the 2.0 to 2.5 mu m range was used to train and test a backpropagation neural net for detection of camouflaged targets. Because of the high degree of correlation between features, the dimensionality of the feature set was reduced using a Karhunen-Loeve expansion.<<ETX>>