FUNCTION APPROXIMATION FRAMEWORK FOR REGION OF INTEREST DETERMINATIONIN SYNTHETIC-APERTURE RADAR IMAGES

Region of interest (ROI) determination is a first and crucial step per formed in an automatic target recognition (ATR) system. The goal of RO I determination is to identify candidate regions that may have potenti al targets. To be most effective, this initial detection (or focus of attention) stage must reject clutter (noise or countermeasures that pr ovide target like characteristics), while ensuring that regions with t rue targets are not missed. We present a novel approach to ROI determi nation in synthetic aperture radar (SAR) images for ATR based on the p remise that regions with targets would require a model with more free parameters to smoothly approximate the magnitude of the return. Toward that end, we use a sigmoidal multilayered feed-forward neural network with selected lateral connections between hidden layer neurons to app roximate the return in disjoint square patches of the SAR image. This network probably uses as few neurons as possible to produce a desired approximation and thus enables the determination of the number of para meters used in approximating the return in an image patch. Those squar es of :he image that require a large number of neurons (more free para meters) are then labeled as ROIs. Results obtained with synthetic and real-world SAR images are used to demonstrate the effectiveness of the proposed method. A significant advantage of the proposed method is th at it does not require the presence of a training data set, which, giv en the variability in SAR images and target signatures, is difficult t o obtain. (C) 1998 Society of Photo-Optical instrumentation Engineers. [S0091-3286(98)02310-1].