Classifier performance using enhanced resolution SAR data

In support of the DARPA/DARO-sponsored SAIP program [1], MlT Lincoln Laboratory has developed a new, state-of-the-art ATR (automatic target recognition) system that provides significantly improved target recognition performance compared with ATR systems that use conventional synthetic aperture radar (SAR) image processing techniques. This significant improvement is achieved by using a new superresolution image processing technique that enhances SAR image resolution (and image quality) prior to performing target recognition; a template-based classifier is then used to perform target recognition. This paper quantifies the improvement in target recognition performance achieved using superresolution image processing in the ATR system.