NEURAL PROCESSING OF TARGETS IN VISIBLE MULTISPECTRAL IR AND SAR IMAGERY

We have designed and implemented computational neural systems for targ et enhancement, detection, learning and recognition in visible, multis pectral infrared (IR), and synthetic aperture radar (SAR) imagery. The system architectures are motivated by designs of biological vision sy stems, drawing insights from retinal processing of contrast and color, occipital lobe processing of shading, color and contour, and temporal lobe processing of pattern and shape. Distinguishing among similar ta rgets, and accumulation of evidence across image sequences is also des cribed. Similar neurocomputational principles and modules are used acr oss these various sensing domains. We show how 3D target learning and recognition from visible silhouettes and SAR return patterns are relat ed. We show how models of contrast enhancement, contour, shading and c olor vision can be used to enhance targets in multispectral IR and SAR imagery, aiding in target detection.