Sea-Spike Detection in High Grazing Angle X-Band Sea-Clutter

Luke Rosenberg
2013 IEEE Transactions on Geoscience and Remote Sensing  
Knowledge of radar sea-clutter phenomenology allows accurate models to be developed for assessing target detection performance. The majority of work in this area has been at low-grazing angles from clifftops or wave tanks and does not consider scattering in the high grazing angle region beyond 10 • . To improve our understanding at high grazing angles 15 • to 45 • , the DSTO's Ingara airborne X-band fully polarimetric radar has been used to collect 12 days worth of sea-clutter data. This report
more » ... r data. This report focuses on understanding the characteristics of sea-spikes as they are often the cause of false detections in a radar processor. Using the Ingara data, a threshold is used to isolate these scatterers in the range/time domain with results verified against the KK probability distribution function. Detections due to discrete and persistent scattering are then isolated to provide more information regarding the underlying cause of seaspikes and answer the question of whether Walker's three component mean Doppler spectrum model is suitable at high grazing angles. APPROVED FOR PUBLIC RELEASE UNCLASSIFIED Executive Summary This report builds on work undertaken at the Defence Science and Technology Organisation (DSTO) in characterising the maritime environment from high altitude airborne platforms. Traditionally, maritime surveillance of small targets is conducted from low altitude platforms and hence with low grazing angles. This surveillance scenario has been well studied and relevant models have been developed. However, little data has been collected and analysed from the high grazing angles typically expected with the operation of high altitude airborne platforms. In August 2004 and July 2006, the DSTO's Ingara X-band airborne radar collected fine resolution fully polarimetric data in the high grazing angle region, 15 • − 45 • . The data was collected in the ocean off the coasts of Port Lincoln and Darwin respectively. This report investigates the detection and characterisation of sea-spikes which result from non-Bragg scattering with the goal of being able to better distinguish them from targets of interest. The approach taken is to threshold the magnitude of the raw backscatter data in the range/time domain. The percentage of sea-spikes present in the data is then measured. Results show that the majority occur in the lower grazing angle region for the horizontal transmit, horizontal receive (HH) channel and are slightly higher in the cross wind directions for the horizontal transmit, vertical receive (HV) and vertical transmit, vertical receive (VV) channels. These results are verified by comparing the trends with a separate analysis which used the KK probability distribution function (PDF) to model the sea-clutter. An image processing algorithm is then used to isolate the discrete and persistent scatterers and test whether Walker's mean Doppler spectrum model is valid with the higher grazing angles. The results show that the persistent 'whitecaps' are spread quite evenly in grazing and azimuth for the HH channel with a clear trend in the cross-wind directions for the HV and VV channels. While there were many common peaks in both the HH and VV channels, there are however, a lot of detections present in HH but not in VV. Also, there are many discrete scatterers detected in the VV channel. This indicates that Walker's three channel model is not totally valid at high grazing angles.
doi:10.1109/tgrs.2013.2239112 fatcat:xguk5lcparhlvhebwlthot4v2u