Detection of ship wakes in SAR images based on Radon transform

The Radon transform is widely used to detect ship wakes in synthetic aperture radar (SAR) images. Ship wakes have linear features in the image, and correspond to peaks and troughs in the Radon transform. Therefore, the ship wakes can be detected by searching for peaks and troughs in the Radon transform. In this paper, a novel algorithm based on the Radon transform is presented to detect ship wakes in SAR images. Considering the variation of the sea clutter, this algorithm uses a locally-adaptive method to search for peaks and troughs in the Radon transform. In addition, a maximum directional-derivative method is used to locate the starting points of the detected ship wakes. The algorithm is tested on real SAR images, and the results demonstrate its effectiveness.