Eigenspace transformation for automatic clutter rejection

The goal of our research is to develop an effective and efficient clutter r ejector with the use of an eigenspace transformation and a multilayer perce ptron (MLP) that can be incorporated into an automatic target recognition s ystem. An eigenspace transformation is used for feature extraction and dime nsionality reduction. The transformations considered in this research are p rincipal-component analysis (PCA) and the eigenspace separation transformat ion (EST). We fed the result of the eigenspace transformation to an MLP tha t predicts the identity of the input, which is either a target or clutter. Our proposed clutter rejector was tested on two huge and realistic datasets of second-generation forward-looking infrared imagery for the Comanche hel icopter. In general, both the PCA and EST methods performed satisfactorily with minor differences. The EST method performed slightly better when a sma ller amount of transformed data was fed to the MLP, or when the positive an d negative EST eigentargets were used together. (C) 2001 Society of Photo-O ptical Instrumentation Engineers.