Development of an ATR Workbench for SAR Imagery

In order to assist in the development of automation techniques for target recognition within SAR imagery, we have implemented an evaluation platform, the ATR Workbench. This will allow researchers and Geospatial Image Analysts (GIAs) to investigate the capabilities of various commercial and experimental applications, singly or in combination, as applied to the target recognition process. The platform will enable studies to determine which aspects improve GIA performance when automated, which methods best improve classifier performance, as well as which methods work better for particular environments and target class definitions. The ATR Workbench was developed, using several open source tools, to provide a platform independent bridge between Automatic Target Detection (ATD) applications and target classifiers. It is capable of importing several kinds of ATD reports, of applying different feature extraction and preprocessing algorithms and of implementing various aspects of Automatic Target Recognition (ATR) applications while importing, displaying and reporting their results. Each step may be automated or operated interactively, as required. Initially, this capability is demonstrated on imagery based upon the public MSTAR data set.