Estimation of ship velocity using SAR imagery - simulation methods

Abstract

The estimation of ship heading and speed from a high resolution Synthetic Aperture Radar (SAR) image of ship and its wake is important for monitoring and tracking ships. In order to get a correct and focused image of the moving ships, it is necessary to know the real positions and velocities of the ships. Though the ship can often be imaged clearly, its orientation may be difficult to estimate from its image because of the effects of ship motion. The wake can provide direct information about the ship heading. The cross range separation between the location of the ship in the image and the wake can also provide an estimate of the ship’s velocity. We propose a strategy based on the generalized Radon Transform and the Stochastic Matched Filtering where the locus of the wake signature in the 2D spectrum of the image is to be detected. In this work we propose three simulation methods for estimation of ship velocities, and directions using SAR imagery. The knowledge of these parameters is used to compensate the SAR image, i.e. to eliminate the imaging errors due to object motion. The proposed methods are based on evaluation of a sequence of single–look SAR images, generated from conventional SAR raw data.