Cross-Platform SAR Image Quality Metric for ATR

The intelligence community uses the National Imagery Interpretability Rating Scale (NIIRS) to quantify the information that an image analyst can extract from a visible image. NIIRS ratings relate the visual quality of an image to the interpretation tasks for which it may be used. A General Image Quality Equation (GIQE) has been used to predict NIIRS ratings for EO/IR images from image quality parameters (for example, image resolution, sharpness, signal-to-noise ratio, etc.) measured directly from the image by the image analyst. There is considerable interest in developing NIIRS image ratings for synthetic aperture radar (SAR) imagery. A NIIRS prediction equation for SAR needs to be developed (a SAR GIQE) that would take into account both the amplitude and phase of SAR imagery and would be applicable to advanced SAR modes utilized by image analysts in exploitation of SAR imagery. This new SAR GIQE would represent multiple SAR products generated from complex, full polarization, and/or multi-pass imagery; advanced modes utilizing these imagery types include detection and recognition, coherent/non-coherent change detection, interferometric and bistatic imaging, super-resolution and ATR processing. SSCI is developing a General Image Quality Equation for predicting NIIRS ratings of SAR imagery. BENEFIT: The development of a NIIRS prediction equation for synthetic aperture radar, a SAR-specific General Image Quality Equation, will provide SAR system designers with a tool for predicting the performance of various SAR exploitation modes (detection and tracking, coherent change detection, super-resolution and ATR processing, interferometric imaging, etc.) prior to actually building the SAR system. This SAR-specific GIQE will provide the system designer with the ability to predict functional performance of a SAR design across both employment and scenario, thereby allowing design and procurement decisions guided by the functions the SAR supports; it will also reveal the capabilities, limitations, and sensitivities critical to determining the best use of sensor resources.