In this paper we compare coherent change detection performance obtained using the maximum likelihood estimate
(MLE) of the SAR image-pair coherence versus using the complex correlation coefficient coherence estimate (CCD).
We also compare the non-coherent change detection performance (P<sub>D</sub> vs. P<sub>FA</sub>) versus the performance of the coherent
change detection algorithms.
In this paper we present a new method for restoring multi-pass synthetic aperture radar (SAR) images containing
arbitrary gaps in SAR phase history data. Frequency and aspect gaps in SAR image spectrum manifest themselves
as artifacts in the associated SAR imagery. Our approach, which we term LDREG for the (cursive ell);<sub>1</sub> difference
regularization, jointly processes multi-pass interrupted data using sparse magnitude and sparse magnitude difference
constraints, and results in improved quality imagery. We find that the joint processing of LDREG results
in coherent change detection gains over independent processing of each data pass. To illustrate the capabilities
of LDREG, we evaluate coherent change detection performance using images from the Gotcha SAR.
In this paper we apply a sparse signal recovery technique for synthetic aperture radar (SAR) image formation
from interrupted phase history data. Timeline constraints imposed on multi-function modern radars result in
interrupted SAR data collection, which in turn leads to corrupted imagery that degrades reliable change detection.
In this paper we extrapolate the missing data by applying the basis pursuit denoising algorithm (BPDN) in the
image formation step, effectively, modeling the SAR scene as sparse. We investigate the effects of regular and
random interruptions on the SAR point spread function (PSF), as well as on the quality of both coherent (CCD)
and non-coherent (NCCD) change detection. We contrast the sparse reconstruction to the matched filter (MF)
method, implemented via Fourier processing with missing data set to zero. To illustrate the capabilities of the
gap-filling sparse reconstruction algorithm, we evaluate change detection performance using a pair of images
from the GOTCHA data set.
Optical coherence tomography (OCT) is a valuable technique for non-invasive imaging in medicine and biology.
In some applications, conventional time-domain OCT (TD-OCT) has been supplanted by spectral-domain OCT
(SD-OCT); the latter uses an apparatus that contains no moving parts and can achieve orders of magnitude faster
imaging. This enhancement comes at a cost, however: the CCD array detectors required for SD-OCT are more
expensive than the simple photodiodes used in TD-OCT. We explore the possibility of extending the notion of
compressed sensing (CS) to SD-OCT, potentially allowing the use of smaller detector arrays. CS techniques can
yield accurate signal reconstructions from highly undersampled measurements, i.e., data sampled significantly
below the Nyquist rate. The Fourier relationship between the measurements and the desired signal in SD-OCT
makes it a good candidate for compressed sensing. Fourier measurements represent good linear projections for
the compressed sensing of sparse point-like signals by random under-sampling of frequency-domain data, and
axial scans in OCT are generally sparse in nature. This sparsity property has recently been used for the reduction
of speckle in OCT images. We have carried out simulations to demonstrate the usefulness of compressed sensing
for simplifying detection schemes in SD-OCT. In particular, we demonstrate the reconstruction of a sparse axial
scan by using fewer than 10 percent of the measurements required by standard SD-OCT.
In this paper we study the impact of sparse aperture data collection of a SAR sensor on reconstruction quality of
a scene of interest. Different mono and multi-static SAR measurement configurations produce different Fourier
sampling patterns. These patterns reflect different spectral and spatial diversity trade-offs that must be made
during task planning. Compressed sensing theory argues that the mutual coherence of the measurement probes
is related to the reconstruction performance of sparse domains. With this motivation we compare the mutual
coherence and corresponding reconstruction behavior of various mono-static and ultra-narrow band multi-static
configurations, which trade-off frequency for geometric diversity. We investigate if such simple metrics are related
to SAR reconstruction quality in an obvious way.
In this paper we present an algorithm for wide-angle synthetic aperture radar (SAR) image formation. Reconstruction
of wide-angle SAR holds a promise of higher resolution and better information about a scene, but it
also poses a number of challenges when compared to the traditional narrow-angle SAR. Most prominently, the
isotropic point scattering model is no longer valid. We present an algorithm capable of producing high resolution
reflectivity maps in both space and aspect, thus accounting for the anisotropic scattering behavior of targets. We
pose the problem as a non-parametric three-dimensional inversion problem, with two constraints: magnitudes
of the backscattered power are highly correlated across closely spaced look angles and the backscattered power
originates from a small set of point scatterers. This approach considers jointly all scatterers in the scene across all
azimuths, and exploits the sparsity of the underlying scattering field. We implement the algorithm and present reconstruction results on realistic data obtained from the XPatch Backhoe dataset.