David Blacknell

Spatially variant incoherence trimming for improved SAR CCD

Conventional synthetic aperture radar (SAR) Coherent Change Detection (CCD) has been found to be of great utility in detecting changes that occur on the ground. The CCD procedure involves performing repeat pass radar collections to form a coherence product, where ground disturbances can induce detectable incoherence. However there is always a difference in the radar collection geometry which can lead to incoherent energy noise entering the CCD. When sensing flat terrain in a far-field regime, the incoherence due to collection geometry difference can be removed through a conventional global Fourier image support trimming process. However, it has been found that when the terrain is either in a near-field regime or contains non-flat topography, the optimal trimming process is substantially more involved, so much so that a new per-pixel SAR back-projection imaging algorithm has been developed. The new algorithm removes incoherent energy from the SAR CCD collection pair on a per-pixel basis according to the local radar geometry and topography, leaving a higher coherence CCD product. In order to validate the approach, change detection measurements were conducted with GB-SAR, a ground-based indoor radar measurement facility.

Very high resolution Coherent Change Detection

Synthetic Aperture Radar (SAR) Coherent Change Detection (CCD) has been found to be of great utility in detecting changes that occur on the ground. Detectable changes of interest include vehicle tracks, water flow, and small scale subsidence. The CCD procedure involves performing repeat pass radar collections to form a coherence product, where ground disturbances can induce detectable incoherence. Currently, SAR imagery of between 10cm and 30cm resolution is considered to be a high resolution, allowing the detection of subtle changes on the ground, however it is of interest to examine CCD images resulting from very high resolution SAR down to 1cm resolution, which in principle could be collected through airborne or spaceborne radar platforms. To perform this study, laboratory data was generated with a ground-based SAR system.

Comparison of vibration and multipath signatures from simulated and real SAR images

The formation of Synthetic Aperture Radar (SAR) imagery requires many assumptions during data acquisition (e.g. assuming a flat focal plane, imaging in the far field, isotropic scattering). If a SAR image formation processor (IFP) does not account for these, then subtle, real-world effects such as those from a vibrating target or multipath signatures can become difficult to isolate from processor-induced artefacts. A SAR simulator has been developed that simulates the raw phase history that would be collected from a scene and input into a SAR IFP. This can then be passed through an operational SAR processor and the resulting image compared with the field-collected SAR image. This allows a precise observation of subtle features (such as those from a vibrating target) to be made. A super-fine resolution Ground-Based SAR system developed by Cranfield University was used to collect true multipath and vibrating signatures and a comparison made.

Hyperspectral 10–50GHz SAR imaging of building materials

A laboratory-based study was carried out to assess the performance and unique intelligence capabilities of an extreme wideband 10-50GHz SAR. Very high resolution range-profile measurements were obtained of samples used in building construction. Periodic features in reflectivity with frequency were interpreted as a resonance between the front and back face reflections of a sample. The features characteristics were in good agreement with the idea of a sample as a resonant microwave cavity. The resonance behaviour was preserved in SAR imaging of the samples, examined by sub-band analysis. Selecting a composite Perspex-MDF target, there was good agreement between simulation and measurement.

Target detection in SAR imagery by diffraction patterning

We report on an investigation into the detection of power and telephone cables in SAR imagery by the presence of diffraction patterning. Laboratory SAR imaging measurements on metal pipes suspended above a gravel surface produced downrange fringe patterns in both backscatter and interferometric phase. Modelling was carried out using an incremental diffraction algorithm, and the positioning and modulation characteristics of the fringes could be understood by the relative positioning of the targets above the gravel and the imaging geometries. Whereas previous studies have relied upon the direct return from the cables, this study has shown their presence might still be inferred from the persistent presence of ground fringes even when a direct return is absent. The effect could find an application in collision avoidance of power and telephone cables by low-flying aircraft, as well as in surveillance and monitoring.

Detection of aerial features by ground diffraction patterning in SAR imagery

We report on an investigation into the secondary detection of objects in SAR imagery by the presence of diffraction patterning. Laboratory SAR imaging measurements were carried on metal pipes suspended above a gravel surface. Clear fringe patterns were seen across the imaged gravel in both backscatter and interferometric phase. Modelling was carried out using an incremental diffraction algorithm. The positioning and modulation characteristics of the fringes could be understood by the relative positioning of the targets above the gravel and the imaging geometries. Even if the pipes or wires are not visible in the imagery from the direct return, their presence might be inferred from the persistent presence of ground fringes.

The physics of vibrating scatterers in SAR imagery

Measurement times for synthetic aperture radar (SAR) image collection can take from the order of seconds to minutes and consequently the technique is subject to imaging artefacts due to target motion. For example, imaged moving targets can be displaced and unfocussed and similarly for vibrating targets. Current understanding of this phenomenon is somewhat esoteric however this paper puts forward and demonstrates a visual explanation via the physics of modulated scatterer SAR images in the Fourier domain. This novel approach has led to an imagery analyst aid which associates a distinctive signature to modulated scatterer artefacts in SAR imagery and to an associated filter.

Utilizing feedback in adaptive SAR ATR systems

Existing SAR ATR systems are usually trained off-line with samples of target imagery or CAD models, prior to conducting a mission. If the training data is not representative of mission conditions, then poor performance may result. In addition, it is difficult to acquire suitable training data for the many target types of interest. The Adaptive SAR ATR Problem Set (AdaptSAPS) program provides a MATLAB framework and image database for developing systems that adapt to mission conditions, meaning less reliance on accurate training data. A key function of an adaptive system is the ability to utilise truth feedback to improve performance, and it is this feature which AdaptSAPS is intended to exploit. This paper presents a new method for SAR ATR that does not use training data, based on supervised learning. This is achieved by using feature-based classification, and several new shadow features have been developed for this purpose. These features allow discrimination of vehicles from clutter, and classification of vehicles into two classes: targets, comprising military combat types, and non-targets, comprising bulldozers and trucks. The performance of the system is assessed using three baseline missions provided with AdaptSAPS, as well as three additional missions. All performance metrics indicate a distinct learning trend over the course of a mission, with most third and fourth quartile performance levels exceeding 85% correct classification. It has been demonstrated that these performance levels can be maintained even when truth feedback rates are reduced by up to 55% over the course of a mission.