Velocity Estimation of Maritime Targets in Spaceborne Single-Channel SAR images: Methods and Performance Assessment

High-resolution images provided by satellite Synthetic Aperture Radar (SAR) systems are of great interest for maritime surveillance applications. As ship targets can be characterized by complex motions, they usually appear defocused in level-1 SAR images. Previous works have shown the effectiveness of the applications of Inverse SAR (ISAR) techniques to refocus the images of such targets. Apart from achieving well-focused images, also the knowledge of ship velocity is strategic information for surveillance. Therefore, in this work, we analyze the capability of ISAR techniques to estimate the ship translational motion: to this purpose, two different autofocus approaches are considered and compared, one based on the image contrast and the other on the phase gradient algorithm. Achievable performances are analyzed by means of Cosmo-SkyMed data and exploiting the Automatic Identification System (AIS) information as the reference ground truth.