Coherent Change Detection Research Articles

Convolutional networks for vehicle track segmentation

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple and fast models to label track pixels. These models, however, are unable to capture natural track features, such as continuity and parallelism. More powerful but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3×3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate in low power and have limited training data. As a result, we aim for small and efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our six-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.

Enabling Reliable Change Detection for Independently Compressed SAR Images

This paper develops distortion metrics for compressed synthetic aperture radar (SAR) imagery from change detection test statistics. These metrics are used to predict lossy image compression’s impact on change detection performance. The metrics do not require the intended change detection comparison image to provide these benefits. An SAR compression system leveraging the distortion metrics is proposed. The system generates a bad-pixel mask highlighting potential false alarms that are generated due to compression and are subsequently discarded in the change detection process. The proposed system’s performance is demonstrated through noncoherent change detection analysis after JPEG2000 and JPEG image compression. Similarly, a coherent change detection system is evaluated after JPEG2000 image compression. For noncoherent change detection at large compression ratios (CRs) using JPEG2000, the proposed system provides a 33% reduction in false alarms at a 0.1 probability of detection as well as the ability to maintain near-distortionless false alarm rates across a wide range of CRs. At a 0.1 probability of detection for coherent change detection, the system provides a 37% reduction in false alarms at modest CRs. The coherent change detection system is also demonstrated to maintain low false alarm rates across a range of CRs.

Options for continuous radar Earth observations

Near Real Time (minutes or hours) radar imaging of ground targets located anywhere on an hemi- sphere, with or without interferometric coherence with previous passes, can be obtained with different solutions that are considered here. Geosynchronous systems, from the one proposed in 1978 by Tomiyasu to telecom satellite compatible solutions, and Low, Medium or Geosynchronous Earth Orbit constellations are discussed. Their benefits, problems, and sizes are briefly summarized, and a comparative table is presented. If interfer-ometric coherence is requested, continuous imaging is obtained only if a very wide geostationary aperture is progressively scanned, eventually using a MIMO (Multiple Input Multiple Output) combination of several slow librating small satellites. Instead, fast librating, strip mapping, large geosynchronous satellites do provide high resolution imaging, but interferometry (and thus coherent change detection) is achievable only after a minimum delay of 12 h, i.e., when the target comes in sight without need to squint the antenna. Hence, both complex and simple systems reach full resolution interferometric imaging and thus coherent change detection capability only after 12 h.

Computing navigation corrections for co-registration of repeat-pass synthetic aperture sonar images

Interferometric processing of repeat-pass Synthetic Aperture Sonar (SAS) data requires a precise co-registration of the images in order to exploit the phase information for applications such as differential interferometry or coherent change detection. Co-registration of SAS images is made difficult by changing environmental conditions and positioning ambiguities that give rise to warping functions which are not well modeled by affine or polynomial transformations, resulting in non-negligible residual misregistration errors. The objective of this presentation will be to examine these warping functions in greater detail to gain a better physical understanding of the estimated along- and across-track displacements. A non-linear least-squares optimization is used to determine the corrections to be applied to the repeat-pass data that will yield a repeat-pass image that is co-registered with the original. These corrections include sway, heave and yaw as well as sound speed errors and element-level surge adjustments. Repeat-pass SAS data collected a Kongsberg HISAS 1020 system is used to demonstrate this optimization, showing good agreement with modeled and real data for most parameters, while determining surge estimates are shown to be particularly challenging. An examination of the resulting repeat-pass coherence will also be performed.

Mitigating Effects of Missing Data for SAR Coherent Images

Missing samples within synthetic aperture radar data result in image distortions. For coherent data products, such as coherent change detection and interferometric processing, the image distortion can be devastating to these second-order products, resulting in missed detections, and inaccurate height maps. Previous approaches to repair the coherent data products focus upon reconstructing the missing data samples. This paper demonstrates that reconstruction is not necessary to restore the quality of the coherent data products.

Trace Coherence: A New Operator for Polarimetric and Interferometric SAR images

Quadratic forms play an important role in the development of several polarimetric and interferometric synthetic aperture radar (Pol-InSAR) methodologies, which are very powerful tools for earth observation. This paper investigates integrals of Pol-InSAR operators based on quadratic forms, with special interest on the Pol-InSAR coherence. A new operator, namely Trace Coherence, is introduced, which provides an approximation for the center of mass of the coherence region (CoRe). The latter is the locus of points on the polar plot containing all the possible coherence values. Such center of mass can be calculated as the integral of Pol-InSAR coherences over the scattering mechanisms (SMs). The trace coherence provides synthetic information regarding the partial target as one single entity. Therefore, it provides a representation, which is not dependent on the selection of one specific polarization channel. It may find application in change detection (e.g., coherent change detection and differential DEM), classification (e.g., building structure parameters), and modeling (e.g., for the retrieval of forest height). In calculating the integral of the Pol-InSAR coherences, an approximate trace coherence expression is derived and shown to improve the calculation speed by several orders of magnitude. The trace coherence approximation is investigated using Monte Carlo simulations and validated ESA (DLR) L-band quad-polarimetric data acquired during the AGRISAR 2006 campaign. The result of the analysis using simulated and real data is that the average error in approximating the integral of the coherence region is 0.025 in magnitude and 3° in phase (in scenarios with sufficiently high coherence).

Vehicle Track Segmentation Using Higher Order Random Fields

We present an approach to segment vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times. The approach uses multiscale higher order random field models to capture track statistics, such as curvatures and their parallel nature, that are not currently utilized in existing methods. These statistics are encoded as 3-by-3 patterns at different scales. The model can complete disconnected tracks often caused by sensor noise and various environmental effects. Coupling the model with a simple classifier, our approach is effective at segmenting salient tracks. We improve the F-measure on a standard vehicle track data set to 0.963, up from 0.897 obtained by the current state-of-the-art method.

Seasonal Change in Wetland Coherence as an Aid to Wetland Monitoring

Water is an essential natural resource, and information about surface water conditions can support a wide variety of applications, including urban planning, agronomy, hydrology, electrical power generation, disaster relief, ecology and preservation of natural areas. Synthetic Aperture Radar (SAR) is recognized as an important source of data for monitoring surface water, especially under inclement weather conditions, and is used operationally for flood mapping applications. The canopy penetration capability of the microwaves also allows for mapping of flooded vegetation as a result of enhanced backscatter from what is generally believed to be a double-bounce scattering mechanism between the water and emergent vegetation. Recent investigations have shown that, under certain conditions, the SAR response signal from flooded vegetation may remain coherent during repeat satellite over-passes, which can be exploited for interferometric SAR (InSAR) measurements to estimate changes in water levels and water topography. InSAR results also suggest that coherence change detection (CCD) might be applied to wetland monitoring applications. This study examines wetland vegetation characteristics that lead to coherence in RADARSAT-2 InSAR data of an area in eastern Canada with many small wetlands, and determines the annual variation in the coherence of these wetlands using multi-temporal radar data. The results for a three-year period demonstrate that most swamps and marshes maintain coherence throughout the ice-/snow-free time period for the 24-day repeat cycle of RADARSAT-2. However, open water areas without emergent aquatic vegetation generally do not have suitable coherence for CCD or InSAR water level estimation. We have found that wetlands with tree cover exhibit the highest coherence and the least variance; wetlands with herbaceous cover exhibit high coherence, but also high variability of coherence; and wetlands with shrub cover exhibit high coherence, but variability intermediate between treed and herbaceous wetlands. From this knowledge, we have developed a novel image product that combines information about the magnitude of coherence and its variability with radar brightness (backscatter intensity). This product clearly displays the multitude of small wetlands over a wide area. With an interpretation key we have also developed, it is possible to distinguish different wetland types and assess year-to-year changes. In the next few years, satellite SAR systems, such as the European Sentinel and the Canadian RADARSAT Constellation Mission (RCM), will provide rapid revisit capabilities and standard data collection modes, enhancing the operational application of SAR data for assessing wetland conditions and monitoring water levels using InSAR techniques.

UTILIZING SAR AND MULTISPECTRAL INTEGRATED DATA FOR EMERGENCY RESPONSE

Abstract. Satellite images are used widely in the risk cycle to understand the exposure, refine hazard maps and quickly provide an assessment after a natural or man-made disaster. Though there are different types of satellite images (e.g. optical, radar) these have not been combined for risk assessments. The characteristics of different remote sensing data type may be extremely valuable for monitoring and evaluating the impacts of disaster events, to extract additional information thus making it available for emergency situations. To base this approach, two different change detection methods, for two different sensor's data were used: Coherence Change Detection (CCD) for SAR data and Covariance Equalization (CE) for multispectral imagery. The CCD provides an identification of the stability of an area, and shows where changes have occurred. CCD shows subtle changes with an accuracy of several millimetres to centimetres. The CE method overcomes the atmospheric effects differences between two multispectral images, taken at different times. Therefore, areas that had undergone a major change can be detected. To achieve our goals, we focused on the urban areas affected by the tsunami event in Sendai, Japan that occurred on March 11, 2011 which affected the surrounding area, coastline and inland. High resolution TerraSAR-X (TSX) and Landsat 7 images, covering the research area, were acquired for the period before and after the event. All pre-processed and processed according to each sensor. Both results, of the optical and SAR algorithms, were combined by resampling the spatial resolution of the Multispectral data to the SAR resolution. This was applied by spatial linear interpolation. A score representing the damage level in both products was assigned. The results of both algorithms, high level of damage is shown in the areas closer to the sea and shoreline. Our approach, combining SAR and multispectral images, leads to more reliable information and provides a complete scene for the emergency response following an event.

UTILIZING SAR AND MULTISPECTRAL INTEGRATED DATA FOR EMERGENCY RESPONSE

Satellite images are used widely in the risk cycle to understand the exposure, refine hazard maps and quickly provide an assessment after a natural or man-made disaster. Though there are different types of satellite images (e.g. optical, radar) these have not been combined for risk assessments. The characteristics of different remote sensing data type may be extremely valuable for monitoring and evaluating the impacts of disaster events, to extract additional information thus making it available for emergency situations. To base this approach, two different change detection methods, for two different sensor's data were used: Coherence Change Detection (CCD) for SAR data and Covariance Equalization (CE) for multispectral imagery. The CCD provides an identification of the stability of an area, and shows where changes have occurred. CCD shows subtle changes with an accuracy of several millimetres to centimetres. The CE method overcomes the atmospheric effects differences between two multispectral images, taken at different times. Therefore, areas that had undergone a major change can be detected. To achieve our goals, we focused on the urban areas affected by the tsunami event in Sendai, Japan that occurred on March 11, 2011 which affected the surrounding area, coastline and inland. High resolution TerraSAR-X (TSX) and Landsat 7 images, covering the research area, were acquired for the period before and after the event. All pre-processed and processed according to each sensor. Both results, of the optical and SAR algorithms, were combined by resampling the spatial resolution of the Multispectral data to the SAR resolution. This was applied by spatial linear interpolation. A score representing the damage level in both products was assigned. The results of both algorithms, high level of damage is shown in the areas closer to the sea and shoreline. Our approach, combining SAR and multispectral images, leads to more reliable information and provides a complete scene for the emergency response following an event.

A New Maximum-Likelihood Change Estimator for Two-Pass SAR Coherent Change Detection

In past research, two-pass repeat-geometry synthetic aperture radar (SAR) coherent change detection (CCD) predominantly utilized the sample degree of coherence as a measure of the temporal change occurring between two complex-valued image collects. Previous coherence-based CCD approaches tend to show temporal change when there is none in areas of the image that have a low clutter-to-noise power ratio. Instead of employing the sample coherence magnitude as a change metric, in this paper, we derive a new maximum-likelihood (ML) temporal change estimate—the complex reflectance change detection (CRCD) metric to be used for SAR coherent temporal change detection. The new CRCD estimator is a surprisingly simple expression, easy to implement, and optimal in the ML sense. This new estimate produces improved results in the coherent pair collects that we have tested.

Forcing Scale Invariance in Multipolarization SAR Change Detection

This paper considers the problem of coherent (in the sense that both amplitudes and relative phases of the polarimetric returns are used to construct the decision statistic) multipolarization synthetic aperture radar change detection starting from the availability of image pairs exhibiting possible power mismatches/miscalibrations. The principle of invariance is used to characterize the class of scale-invariant decision rules which are insensitive to power mismatches and ensure the constant false alarm rate property. A maximal invariant statistic is derived together with the induced maximal invariant in the parameter space which significantly compresses the data/parameter domain. A generalized likelihood ratio test is synthesized both for the cases of two- and three-polarimetric channels. Interestingly, for the two-channel case, it is based on the comparison of the condition number of a data-dependent matrix with a suitable threshold. Some additional invariant decision rules are also proposed. The performance of the considered scale-invariant structures is compared to those from two noninvariant counterparts using both simulated and real radar data. The results highlight the robustness of the proposed method and the performance tradeoff involved.

Two-Stage Change Detection for Synthetic Aperture Radar

Coherent change detection using paired synthetic aperture radar (SAR) images is often performed using a classical coherence estimator that is invariant to the true variances of the populations underlying each paired sample. While attractive, this estimator is biased and requires a significant number of samples to yield good performance. Increasing sample size often results in decreased image resolution. Thus, we propose the use of Berger's coherence estimate because, with the same number of pixels, the estimator effectively doubles the sample support without sacrificing resolution when the underlying population variances are equal or near equal. A potential drawback of this approach is that it is not invariant since its distribution depends on the pixel pair population variances. While Berger's estimator is inherently sensitive to the inequality of population variances, we propose a method of insulating the detector from this acuity. A two-stage change statistic is introduced to combine a noncoherent intensity change statistic given by the sample variance ratio, followed by the alternative Berger estimator, which assumes equal population variances. The first-stage detector identifies pixel pairs that have nonequal variances as changes caused by the displacement of sizeable object. The pixel pairs that are identified to have equal or near-equal variances in the first stage are used as an input to the second stage. The second-stage test uses the alternative Berger coherence estimator to detect subtle changes such as tire tracks and footprints. We show experimentally that the proposed method yields higher contrast SAR change detection images than the classical coherent change detector (state of the art), the alternative coherent change detector, and the intensity change detector. Experimental results are presented to show the effectiveness and robustness of the proposed algorithm for SAR change detection.

Phase measurement accuracy in noise waveform synthetic aperture radar

Detection of small changes in natural formations and human-made constructions is a promising application of synthetic aperture radar (SAR). This paper deals with a specific radar, which is a ground-based SAR using noise waveform. Being fully coherent, SAR provides both phase and amplitude information for every pixel in the mapped area. This gives rise to the possibility of comparing phases between the images taken from the same position at different times in order to extract information about the changes that have occurred between subsequent radar measurements. Such a technique, called coherent change detection (CCD), allows changes of a subwavelength size to be detected. The use of a noise waveform for sounding ensures high electromagnetic compatibility and interference immunity. Moreover, it enables long integration time in the detection stage. However, the random nature of probing signals used in noise radars leads to randomness of the output signals and the preservation of residual fluctuation in the images generated. This random residual, along with external noise, influences the overall accuracy of the phase measurement in noise radar. In this paper, the theoretical and experimental investigation of this specific influence of noise on the precision of phase measurements is presented. The theoretical part of the paper is focused on the estimation of the Cramer-Rao lower bound for the phase measurements using both statistical theory and geometrical interpretation of complex signals. In the experimental part, these findings are verified with dedicated measurements of a metallic sphere displacement, carried out with the help of a ground-based noise waveform SAR.

Overview of the RADARSAT Constellation Mission

. The RADARSAT Constellation Mission (RCM) is a Canadian synthetic aperture radar (SAR) mission consisting of the design, construction, launch, and operation of three C-band SAR satellites as well as the design and build of an associated ground segment. RCM is currently being implemented by a Canadian industrial team led by MacDonald Dettwiler and Associates (MDA) under contract to the Canadian Space Agency (CSA). This article provides an overview of the RCM. A description of how mission requirements drove the selection of key mission parameters is given, including number of satellites, orbit selection, and major radar parameters. Emphasis is placed on providing a description of those capabilities of RCM that are unique or innovative. These capabilities include a constellation optimized for coherent change detection (CCD), a compact polarimetry mode, special imaging modes optimized for ship detection, and a combined SAR and automatic identification system (AIS) capability. To support CCD, the system design includes maintaining the satellites in a narrow orbital tube, a tight requirement on azimuth antenna pointing control, and ScanSAR burst timing calculated from on-orbit spacecraft location data in order to accurately synchronize ScanSAR bursts with along-track position. For compact polarimetry, a capability to transmit circular polarization and receive horizontal (H) and vertical (V) polarizations simultaneously is part of the design. For ship detection, innovative ScanSAR modes are used with a large number of beams and variable resolution and number of looks designed to optimize ship detection across a wide swath. The combined SAR and AIS capability is achieved with an AIS instrument using both on-board AIS processing and on-ground AIS processing.Résumé. La mission de la Constellation RADARSAT (MCR) est une mission canadienne de radars à synthèse d’ouverture (SAR) qui consiste en la conception, la construction, le lancement et l’exploitation de trois satellites SAR en bande C ainsi que de la conception et la construction d’un segment au sol associé. Le MCR est actuellement mis en œuvre par une équipe industrielle canadienne dirigée par MacDonald Dettwiler and Associates (MDA) sous contrat avec l’Agence spatiale canadienne (ASC). Ce document donne un aperçu de la MCR. Nous décrivons la façon dont les exigences de la mission ont conduit à la sélection des paramètres clés de la mission, y compris le nombre de satellites, la sélection de l’orbite et les principaux paramètres radar. L’accent est mis sur la description des capacités uniques ou innovantes de la MCR. Ces capacités comprennent une constellation optimisée pour la détection de changement cohérent (CCD), un mode de polarimétrie compacte, des modes d’imagerie spéciaux optimisés pour la détection des navires et une capacité combinée du SAR et du système d’identification automatique (AIS). Pour permettre la CCD, la conception du système comprend la maintenance des satellites dans un tube orbital étroit, une exigence stricte sur le contrôle de pointage de l’antenne d’azimut, et la synchronisation des rafales du ScanSAR calculée à partir des données de localisation en orbite du satellite afin de synchroniser avec précision les rafales du ScanSAR avec la position le long de la piste. Pour la polarimétrie compacte, une capacité de transmettre la polarisation circulaire et de recevoir les polarisations horizontales (H) et verticales (V) simultanément est une partie de la conception. Pour la détection des navires, les modes innovants du ScanSAR sont utilisés avec un grand nombre de faisceaux, des résolutions et des nombres de visées variables qui sont conçus pour optimiser la détection des navires à travers une large fauchée. La capacité combinée du SAR et du AIS est réalisée avec un instrument AIS utilisant à la fois le traitement AIS à bord et le traitement AIS au sol.

A Generalized Likelihood Ratio Test for Coherent Change Detection in Polarimetric SAR

Several detection statistics have been proposed for detecting fine ground disturbances between two synthetic aperture radar (SAR) images, such as vehicle tracks. The standard method involves estimating a local correlation coefficient between images. Other methods have been proposed using various statistical hypothesis tests. One of these alternative methods is a generalized likelihood ratio test (GLRT), which compares a full-correlation image model to a no-correlation image model. In this letter, we expand the GLRT to polarimetric SAR data and derive the appropriate GLRT detection statistics. Additionally, we explore relaxing the equal variance/equal polarimetric covariance assumptions used in previous results and find improved performance on macroscopic scene changes.

Man-Made Change Detection Using High-Resolution Cosmo-SkyMed SAR Interferometry

The technique of coherent change detection (CCD) in repeat-pass synthetic aperture radar (SAR) imagery has the potential to detect very subtle scene changes. With the advent of high-resolution space-borne SAR images, CCD has received a lot of interest. In CCD, the interferometric coherence is evaluated and analyzed to detect changes. The sample coherence estimator is biased, especially for low-coherence values, which complicates CCD map analysis and interpretation. The bias decreases when the number of samples used to evaluate the coherence increases. However, a large number of samples causes the loss of the small size changes. In this work, an adaptive method for coherence estimation is described and evaluated. The local fringe frequency is used as an additional change indicator to adapt the number of samples used to estimate the coherence. The results obtained using Cosmo-SkyMed (CSK) SAR images indicate that the proposed adaptive method improves detection performance while preserving the small size changes.

Accurate Coherent Change Detection Method Based on Pauli Decomposition for Fully Polarimetric SAR Imagery

Accurate Coherent Change Detection Method Based on Pauli Decomposition for Fully Polarimetric SAR Imagery

Interrupted SAR persistent surveillance via group sparse reconstruction of multipass data

In this paper we present a new method for synthetic aperture radar (SAR) image formation from interrupted, multipass SAR phase history data, with application to persistent surveillance SAR imaging. We propose a new compressed sensing-motivated approach to reconstruction that jointly processes multipass interrupted data using a sparse recovery technique with a group support constraint and results in improved imagery. We compare our approach, a group sparsity (GS) algorithm, to methods that independently process each data pass, namely the basis pursuit denoising and iterative adaptive approach methods. We find that the joint processing of GS results in coherent change detection gains over the other approaches regardless of interrupt pattern. To illustrate the capabilities of GS, we evaluate coherent change detection performance using images from the Gotcha SAR.

Noncoherent Approach for Through-the-Wall Moving Target Indication

A moving target indication, noncoherent change detection algorithm is introduced to detect moving targets inside buildings. The proposed algorithm is designed to attenuate image artifacts observed in coherent change detection images by utilizing noncoherent change detection, a positive threshold operation, and sidelobe minimization. The proposed algorithm is compared with coherent change detection for three moving target scenarios. It is shown that the proposed algorithm significantly reduces imaging artifacts while preserving the moving target signature.