3D reconstruction of target geometry based on 2D ISAR image movies

Inverse synthetic aperture radar (ISAR) images are frequently used in target classification and recognition applications. If we keep observing the target and consecutively generate multiple ISAR images, the target image varies considerably due to the motion of the target. In this paper, we propose a new ISAR system with three receivers. The phases measured from two orthogonal baselines are used to jointly estimate the effective rotation vector and the scattering center positions. The scattering center is assigned based on the probabilistic least squares (PLS) approach. However, the algorithm requires a priori knowledge of the relative motion of the target. The effectiveness and robustness of the proposed algorithm is proven theoretically and tested against by the simulation.