Skip to Main Content
An image processing procedure is presented to construct a three-dimensional representation of a vehicle from a set of two-dimensional circular synthetic aperture radar (SAR) images. Given a previously generated digital elevation map, a third dimension is inferred from layover and polarization. With high data rates and circular flight paths, SAR scenes may be large relative to the standoff distance. Consequently, variations in elevation angles (with respect to terrain) across the scene become crucial for identifying signatures. The proposed 3D processing inverts elevated point layover to provide partial invariance to elevation angle and produces an unordered set of scattering centers with attributes including an azimuth attribute to preserve anisotropy. The proposed processing is demonstrated on both measured and simulated data.