Ship detection with RadarSat-2 Quad-Pol sar data using a notch filter based on perturbation analysis

Target detection of marine feature is a major topic for the security and monitoring of coastlines. Synthetic Aperture Radar (SAR) has been shown to be particularly useful for this application because of its all-weather and night capability. In this paper a new ship and iceberg detection methodology is described. The algorithm proposed is based on a perturbation analysis in the target space recently developed and published by the authors, which was focused on land based target detection. The algorithm can be considered to be a negative filter focused on sea. Consequently, all the features which have a polarimetric behaviour different from the sea are detected. To demonstrate and validate the technique two RadarSat Fine Quad-Pol mode scenes were acquired off the south coast of the UK at Portsmouth harbour. An extensive ground truth campaign was also conducted that was coincident with these acquisitions. Portsmouth is one of the busiest harbours in the UK and this afforded the opportunity to capture a wide range of vessel sizes and types for analysis.