Feature Extraction and Shape Reconstruction for SAR to LADAR Target Transfer

"Laser radar (LADAR) has emerged as one of the key technology candidates for application as a precision weapon terminal seeker. Targets of interest include unplanned targets of opportunity, such as fixed buildings and bunkers, and relocatable and movingtargets. To discriminate such targets, a seeker requires a template, based on the target surface shape, that must be constructed from the available information. LADAR can provide an image with three channels of target information, namely a map of range,active reflectance, and passive reflectance and emission. The RF frequency for SAR doesn't provide information about active or passive target emission in the visible, but it can be applied to construct the target surface shape. Accordingly, we investigatehow best to construct the 3D target shape from standoff SAR to support LADAR terminal seeker autonomous target acquisition and track. We leverage our experience from the model-based SAR ATR MSTAR program in deriving features from the SAR data andconverting these features to suitable 3D surface representations, but the proposed system here will not be an ATR. In particular, we propose to investigate a direct shape reconstruction approach and a shape matching approach. In order to analyze theperformance tradeoffs of the different approaches and understand their benefits and limitations, we will perform extensive analysis on available in-house SAR imagery covering a broad range of operating