Synthetic aperture radar (SAR) images contain a grainy pattern, called speckle, that is a consequence of a coherent imaging system. For fine resolution SAR images speckle can obscure subtle features and reduce visual appeal. Many speckle reduction methods result in a loss of image resolution and reduce visual appeal which can obscure subtle features. Another approach to maintain resolution while reducing speckle is to register and combine multiple images. For persistent surveillance applications it is more efficient for an airborne platform to fly circles around the particular area of interest. In these cases, it would be beneficial to combine multiple circle mode SAR images, however the image registration process is not so straightforward because the layover angle changes in each image. This paper develops a SAR image registration process for combining multiple circle mode SAR images to reduce speckle while preserving resolution. The registration first uses a feature matching algorithm for a coarse rotation and alignment, and then uses a fine registration and warp. Ku band SAR data from a circle mode SAR collection is used to show the effectiveness of the registration and enhanced visual appeal from multi-looking.
Interference and interference mitigation techniques degrade synthetic aperture radar (SAR) coherent data products. Radars utilizing stretch processing present a unique challenge for many mitigation techniques because the interference signal itself is modified through stretch processing from its original signal characteristics. Many sources of interference, including constant tones, are only present within the fast-time sample data for a limited number of samples, depending on the radar and interference bandwidth. Adaptive filtering algorithms to estimate and remove the interference signal that rely upon assuming stationary interference signal characteristics can be ineffective. An effective mitigation method, called notching, forces the value of the data samples containing interference to zero. However, as the number of data samples set to zero increases, image distortion and loss of resolution degrade both the image product and any second order image products. Techniques to repair image distortions,1 are effective for point-like targets. However, these techniques are not designed to model and repair distortions in SAR image terrain. Good terrain coherence is important for SAR second order image products because terrain occupies the majority of many scenes. For the case of coherent change detection it is the terrain coherence itself that determines the quality of the change detection image. This paper proposes an unique equalization technique that improves coherence over existing notching techniques. First, the proposed algorithm limits mitigation to only the samples containing interference, unlike adaptive filtering algorithms, so the remaining samples are not modified. Additionally, the mitigation adapts to changing interference power such that the resulting correction equalizes the power across the data samples. The result is reduced distortion and improved coherence for the terrain. SAR data demonstrates improved coherence from the proposed equalization correction over existing notching methods for chirped interference sources.
KEYWORDS: Radar, Synthetic aperture radar, Image acquisition, Scene simulation, Data modeling, Radar signal processing, Signal processing, Time-frequency analysis, Detection and tracking algorithms, Motion models
We propose a new laboratory method for characterizing synthetic aperture radar (SAR) systems through the use of a synthetic scene generator. Flight tests are the only definitive way to characterize the system level performance of airborne synthetic aperture radar systems. However, due to the expense of flights tests it is beneficial to complete as much testing as possible in a laboratory environment before flight testing is performed. There are many existing tests that are employed to measure the performance of various subsystems in a SAR system, find defective hardware, and indicate design problems that need to be mitigated. However, certain issues can only be found on an integrated system, and laboratory testing at a system level is typically confined to characterizing the impulse response (IPR) of a single point target through the use of an optical delay line. While useful, delay line testing requires running a modified version of real-time image formation code as the delay line does not completely mimic a real target. Ideally, system level tests are performed on unmodified code. On modern SAR systems many algorithms are data driven (e.g., autofocus) and require a substantially more sophisticated data model for testing. We desire to create a complete system test by combining an arbitrary number of point targets and clutter patterns to mimic radar responses from a real scene. This capability enables complete testing of radar systems in a laboratory environment according to prescribed terrain/scene characteristics. This paper presents an overview of the system requirements for a synthetic scene generator. The analysis is limited to SAR systems utilizing chirp waveforms and stretch processing. Furthermore, we derive relationships between IF bandwidth, target position, and the phase history model. A technique to properly compensate for motion pulse to pulse is presented. Finally, our concept is demonstrated with simulation data.
Synthetic aperture radar systems that use the polar format algorithm are subject to a focused scene size limit inherent to
the polar format algorithm. The classic focused scene size limit is determined from the dominant residual range phase
error term. Given the many sources of phase error in a synthetic aperture radar, a system designer is interested in how
much phase error results from the assumptions made with the polar format algorithm. Autofocus algorithms have limits
to the amount and type of phase error that can be corrected. Current methods correct only one or a few terms of the
residual phase error. A system designer needs to be able to evaluate the contribution of the residual or uncorrected phase
error terms to determine the new focused scene size limit. This paper describes a method to estimate the complete
residual phase error, not just one or a few of the dominant residual terms. This method is demonstrated with polar format
image formation, but is equally applicable to other image formation algorithms. A benefit for the system designer is that
additional correction terms can be added or deleted from the analysis as necessary to evaluate the resulting effect upon
image quality.
The trans-rectal implementation of NIR optical tomography makes it possible to assess functional status like hemoglobin
concentration and oxygen saturation in prostate non-invasively. Trans-rectal NIR tomography may provide tissue-specific
functional contrast that is potentially valuable for differentiation of cancerous lesions from normal tissues. Such
information will help to determine if a prostate biopsy is needed or can be excluded for an otherwise ambiguous lesion.
The relatively low spatial resolution due to the diffuse light detection in trans-rectal NIR tomography, however, limits
the accuracy of localizing a suspicious tissue volume. Trans-rectal ultrasound (TRUS) is the clinical standard for guiding
the positioning of biopsy needle owing to its resolution and convenience; nevertheless, TRUS lacks the pathognomic
specificity to guide biopsy to only the suspicious lesions. The combination of trans-rectal NIR tomography with TRUS
could potentially give better differentiation of cancerous tissue from normal background and to accurately localize the
cancer-suspicious contrast obtained from NIR tomography. This paper will demonstrate the design and initial evaluation
of a trans-rectal NIR tomography probe that can conveniently integrate with a commercial TRUS transducer. The transrectal
NIR tomography obtained from this probe is concurrent with TRUS at matching sagittal imaging plane. This
design provides the flexibility of simple correlation of trans-rectal NIR with TRUS, and using TRUS anatomic
information as spatial prior for NIR image reconstruction.
Near-infrared optical tomography is an interesting technique of imaging with high blood-based contrast. Unfortunately non-invasive NIR tomographic imaging has been restricted to specific organs like breast that can be transilluminated externally. In this paper, we demonstrate that near-infrared (NIR) optical tomography can be employed at the endoscope-scale, and implemented at a rapid sampling speed that allows translation to in vivo use. A spread-spectral-encoding technique based on a broadband light source is combined with light delivery by linear-to-circular fiber bundle, to provide endoscopic probing of multiple source/detector fibers for tomographic imaging as well as parallel sampling of all source-detector pairs for rapid data acquisition. Endoscopic NIR tomography is demonstrated by use of a 12mm diameter probe housing 8 sources and 8 detectors at 8 Hz frame rate. Transrectal NIR optical tomography by use of tissue specimen is also presented. This novel approach provides the key feasibility studies to allow this blood-based contrast imaging technology to be tried in cancer detection of internal organs via endoscopic interrogation.
Endoscopic near-infrared (NIR) optical tomography is a novel approach that allows the blood-based high intrinsic
optical contrast to be imaged for the detection of cancer in internal organs. In endoscopic NIR tomography, the imaging
array is arranged within the interior of the medium as opposed to the exterior as seen in conventional NIR tomography
approaches. The source illuminates outward from the circular NIR probe, and the detector collects the diffused light
from the medium surrounding the NIR probe. This new imaging geometry may involve forward and inverse approaches
that are significantly different from those used in conventional NIR tomography. The implementation of a hollow-centered
forward mesh within the context of conventional NIR tomography reconstruction has already led to the first
demonstration of endoscopic NIR optical tomography. This paper presents some fundamental computational aspects
regarding the performance and sensitivity of this endoscopic NIR tomography configuration. The NIRFAST modeling
and image reconstruction package developed for conventional circular NIR geometry is used for endoscopic NIR
tomography, and initial quantitative analysis has been conducted to investigate the "effective" imaging depth, required
mesh resolution, and limit in contrast resolution, among other parameters. This study will define the performance
expected and may provide insights into hardware requirements needed for revision of NIRFAST for the endoscopic NIR
tomography geometry.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.