The Turbulent Wake Detection and Beam Estimation of a Ship in SAR Images

Synthetic aperture radar (SAR) images of sea surface often reveal ship wakes, which appear in the form of bright and dark streaks. The bright streaks are always attributed to the Kelvin wake and bright narrow V-wake, and the dark ones are always to the turbulent wake. In some SAR images the Kelvin wakes and the bright V-wakes are distinct and there are plenty of papers have discussed on the detection of them. But in some other SAR images the turbulent wakes are more obvious. The present work deals with the turbulent wakes of the ships, proposed a method called ellipse normalize-Radon transform to detect the dark lines of the turbulent wake. Once the turbulent wake is detected, the ship beam is estimated by a semi-empirical relation between the ship and the width of the turbulent wake in SAR images and the ship beam developed by Greory Zilman. The experiment results demonstrate that this algorithmpsilas robustness in the presence of noise, as well as its ability to detect the turbulent wake and estimate the beam of a ship.