3D Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) in Contested Environments

Synthetic aperture radar (SAR) works where other sensing modalities fail, in the dead of night and through cloud cover. However, the projection from a 3D scene to a 2D image is different than the perspective projection analysts and algorithms are used to interpreting, and high-resolution images require long integration times that can lead to distorted images if the scene is moving. The required integration time is especially long for 3D SAR, which is often considered infeasible in contested environments. This proposal considers a collaborative imaging scenario, with at least two different platforms, to address these two issues. The proposed approach will estimate a 3D target using non-coherent images from different sensors and it will interpret this 3D scene using an automatic target recognition (ATR) algorithm that performs inference on this 3D representation. While non-coherent 3D scene estimation is still not a solved problem, Toyon has made recent progress using both airborne and satellite sensors. Our solution has been evaluated by different independent governmental organizations with positive reviews.