Automatic Ship Detection in Space-borne Sar Imagery

Synthetic aperture radar (SAR) imagery has proven to be a promising data source for the surveillance of maritime activity, and its application for automatic ship detection has been the focus of many research studies. Apart from the well-known CFAR detector, there has emerged a novel method for automatic ship detection, based on the wavelet transform. Since the underlying principles for both methods are fundamentally different, their advantages and disadvantages concerning various image features also differ. Within this paper we will present a prototype ship detection system that attempts to combine the benefits yielded by the two aforementioned techniques, thus gaining both sensitivity for weak targets and robustness against false alarms in inhomogeneous areas. For this, a wavelet-based prescreening stage is applied, which is followed by an object analysis, and a final adaptive-threshold test. The prototype has been tested and assessed on ALOS PALSAR and RADARSAT-1 data, especially with respect to the behavior toward sea-ice areas and irregularities such as beam seams in ScanSAR imagery. The results indicate a compensation of the intrinsic drawbacks held by the individual detection methods, producing a reliable and versatile detection system.