Flexible, High-Speed Video Display Processor using Field Programmable Gate Array

Not Available For robust target detection, we propose the TNE paradigm, which has recently been extensively exercised against the MSTAR derived HRR radar database. We will develop a demonstration test bed to characterize the use of TNE for target vehicle detection in a classify before detect mode. Our goal is to demonstrate proof of concept for a robust target detection systems capable of mitigating naturally induced false alarms in medium resolution magnitude detected microwave synthetic radar imagery from airborne collection systems. We will define and characterize the performance of prototype direct TNE methods for automated use of scene context in Foliage Penetration (FOPEN) Synthetic Aperture Radar (SAR) imagery to improve detection of camouflaged or concealed targets. Our testing will use the radar signatures of both individual and groups of vehicles in areas with significant background clutter density resulting from target-like natural and man made objects. The TNE paradigm has been shown to be a computationally efficient method for robust target detection and recognition using HRR signature data. We will extend testing the to the SAR/FOPEN arena. The TNE algorithm is very suitable for parallelization and porting to Field Programmable Gate Arrays and other Digital Signal Processing devices for real-time implementation.