Performance estimation of SAR imagery using NIIRS techniques

We propose a windowed General Image Quality Equation that incorporates automatic target recognition specific features. Because the metric adapts to local clutter statistics, we expect it to better represent the non-uniform characteristics of the imagery. The feature weights are determined via linear programming techniques constrained to be monotonic. BENEFIT: Defining exploitation specific image quality is a key challenge to multiple disciplines including non-destructive evaluation of airframes and biomedical imaging. We have been developing and marketing tools for these purposes that will be able to leverage the results of this effort.