Abstract
ABSTRACTThis article describes how a detectability model can be trained in the form of a binary classifier from a data set of synthetic aperture radar (SAR) images of ship wakes, augmented by automatic identification system data. While detectability models for ship signatures exist, ship wake detectability models are only available for simulated data. In order to improve existing ship wake detection algorithms on SAR imagery, there is a need for building a data-driven detectability model which may provide useful a-priori information. A binary L2-regularized logistic regression classifier is trained for each investigated data subset. The dependency on the SAR working frequency is evaluated by analysing a large number of X- and C-band images. In the X-band, the probability of detecting a wake shows dependencies on vessel size and velocity as well as prevailing wind speed. In the C-band, these dependencies are maintained, but with a general reduction in the correlation. This fact led us to the conclusion that, for our data set, ship wakes are more easily imaged in the X-band rather than in the C-band. This is an important outcome, which is supported by a qualitative and quantitative analysis of a large data set of TerraSAR-X and two independent C-band sensors, specifically RADARSAT-2 and Sentinel-1.
Highlights
Synthetic aperture radar (SAR) sensors are utilized by public and private users to support maritime safety and security in worldwide oceans and coastal waters (ESA 2017)
In order to eliminate the possibility that differences in the detectability of ship wakes on C-band and X-band data arise from the effects of this new Terrain Observation by Progressive Scans (TOPS) mode or from the reduced resolution of Sentinel-1 Interferometric Wide Swath (IW) images, which is lower in comparison to TerraSAR-X Stripmap and Spotlight images, a second C-band collection consisting of 30 RADARSAT-2 Fine images was analysed, all with HH-polarization
A detectability model based on L2-regularized logistic regression trained with TerraSARX (X1-MIX), TerraSAR-X HH-polarization (X1-HH), TerraSAR-X vertical receive (VV)-polarization (X1-VV), Sentinel-1 VV-polarization (C1-VV), and RADARSAT-2 HH-polarization (C2-HH) data is presented
Summary
VELOTTO ship wake signatures is often approached by researchers (Copeland, Ravichandran, and Trivedi 1995; Eldhuset 1996; Graziano, D’Errico, and Rufino 2016b). The appearance of ship wakes varies strongly in extent and visibility of the diverse wake components, depending on the SAR sensor, SAR frequencies, and SAR processing architectures, as well as image acquisition parameters and environmental conditions (Soloviev et al 2010). Data from the TerraSAR-X X-band SAR satellite, the RADARSAT-2 C-band SAR satellite, and the Sentinel-1 C-band SAR satellite are used. For these sensors, detectability analyses based on simulated SAR data are available (Zilmann, Zapolski, and Marom 2014), but statistics based on real data are still lacking. The resulting detectability models could be applied to control an automatic process for wake detection
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
