Synthetic Aperture Radar Detection Research Articles

Fusion Detection via Distance-Decay Intersection over Union and Weighted Dempster–Shafer Evidence Theory

In recent years, increasing attention has been paid to object detection in remote sensing imagery. However, traditional optical detection is highly susceptible to illumination and weather anomalies. It is a challenge to effectively utilize the cross-modality information from multisource remote sensing images, especially from optical and synthetic aperture radar images, to achieve all-day and all-weather detection with high accuracy and speed. Toward this end, a fast multisource fusion detection framework FusionDet is proposed in the current paper. A novel distance-decay intersection over union is employed to encode the shape properties of the targets with scale invariance. Therefore, the same target in multisource images can be paired accurately. Furthermore, the weighted Dempster–Shafer evidence theory is utilized to combine the optical and synthetic aperture radar detection, which overcomes the drawback in a feature-level fusion that requires a large amount of paired data. In addition, the paired optical and synthetic aperture radar images are taken to demonstrate our fusion algorithm. Considerable experiments and ablation studies on multiple datasets show that our method is generally effective and achieves state-of-the-art detection performance.

Feedback-Assisted Automatic Target and Clutter Discrimination Using a Bayesian Convolutional Neural Network for Improved Explainability in SAR Applications

In this paper, a feedback training approach for efficiently dealing with distribution shift in synthetic aperture radar target detection using a Bayesian convolutional neural network is proposed. After training the network on in-distribution data, it is tested on out-of-distribution data. Samples that are classified incorrectly with high certainty are fed back for a second round of training. This results in the reduction of false positives in the out-of-distribution dataset. False positive target detections challenge human attention, sensor resource management, and mission engagement. In these types of applications, a reduction in false positives thus often takes precedence over target detection and classification performance. The classifier is used to discriminate the targets from the clutter and to classify the target type in a single step as opposed to the traditional approach of having a sequential chain of functions for target detection and localisation before the machine learning algorithm. Another aspect of automated synthetic aperture radar detection and recognition problems addressed here is the fact that human users of the output of traditional classification systems are presented with decisions made by “black box” algorithms. Consequently, the decisions are not explainable, even to an expert in the sensor domain. This paper makes use of the concept of explainable artificial intelligence via uncertainty heat maps that are overlaid onto synthetic aperture radar imagery to furnish the user with additional information about classification decisions. These uncertainty heat maps facilitate trust in the machine learning algorithm and are derived from the uncertainty estimates of the classifications from the Bayesian convolutional neural network. These uncertainty overlays further enhance the users’ ability to interpret the reasons why certain decisions were made by the algorithm. Further, it is demonstrated that feeding back the high-certainty, incorrectly classified out-of-distribution data results in an average improvement in detection performance and a reduction in uncertainty for all synthetic aperture radar images processed. Compared to the baseline method, an improvement in recall of 11.8%, and a reduction in the false positive rate of 7.08% were demonstrated using the Feedback-assisted Bayesian Convolutional Neural Network or FaBCNN.

Synthetic aperture radar detection and characteristic analysis of cyanobacterial scum in Lake Taihu

To compensate for the limitations of optical remote sensing when restricted by cloud cover, it is worth exploring how to detect cyanobacterial blooms using synthetic aperture radar (SAR), which can penetrate clouds. A satellite–ground synchronous experiment was conducted in Lake Taihu, the third largest freshwater lake in China. A lipopeptide biosurfactant was detected in the algal scum layer, with an average content of 1.8 g/L. The viscosity (1.41 to 332 mPa.s) of the scum was significantly higher than that of scum-free water. The surface tension of the algal scum decreased by 12.5%, and the SAR microwave backscatter was reduced by 7.3 dB. This indicated that the cyanobacterial scum could effectively attenuate capillary waves and appear as dark patches in SAR images. SAR has the potential to be developed as a tool for the remote sensing of algal scum in lake waters.

Sandglass transformation for synthetic aperture radar detection and imaging of ship at low signal-to-clutter-plus-noise ratio

Space/air-borne synthetic aperture radar (SAR) detection and imaging of moving ships at sea are important for ocean reconnaissance and fishery monitoring. A novel algorithm based on sandglass transformation to detect weak ships and form high-quality images under the conditions of low signal-to-clutter-plus-noise ratio is presented here. It requires no prior information about motion parameters of targets. The sandglass transformation can decouple the time and lag time in the instantaneous autocorrelation function of a linear frequency modulated (LFM) signal. Thus the cross-range signals of ship in the approximate form of LFM signals can be integrated coherently via the two-dimensional fast Fourier transformation. The proposed algorithm can not only achieve high-energy accumulation gain, but also suppress the interference of cross terms effectively without loss of resolution. Hence, it is effective to detect weak ships and generate high-resolution images. Numerical and experimental results confirm the effectiveness of the proposed algorithm.

Satellite synthetic aperture radar detection of Delaware Bay plumes: Jet‐like feature analysis

From the physics of radar imaging of ocean surface processes a theoretical model for the radar imaging of an ocean jet was derived. The theoretical model predicts that the jet current structure, sech2η, constitutes a determining factor for the radar image. The modeled image intensity depends on the axial velocity, decreasing as x−1 along the jet axis, and is sensitive to the wind direction with respect to the jet axis. The model was used to interpret synthetic aperture radar (SAR) images of Delaware Bay plumes taken by the RADARSAT‐1 (Canadian Radar Satellite) and ERS‐2 (European Remote Sensing satellite) satellites during a period from summer 1996 to spring 1997. In all seasons the low‐salinity plumes appear as relatively bright, jet‐like patterns on SAR images. From a morphological interpretation of the image features, both summer and winter plumes can be divided into three sections along the axis: source, jet, and dispersion region. Along the transverse direction, SAR image interpretations indicate that the plumes have a twin jet structure, which conforms to decomposition of field measurements. In summer the typical axial velocity is estimated at 6 × 10−1 m s−1 with a Reynolds number of 18. At about 5 km downstream from the source the plume behaves like a turbulent jet, and beyond that range down to about 10 km downstream, it behaves like a laminar jet. In winter the plumes become weaker than in summer. The typical axial velocity is estimated at 4 × 10−1 m s−1 with a Reynolds number of 3. The jet behaves like the laminar jet off the source down to about 4 km downstream. In both summer and winter cases the SAR images were taken at maximum flood tide, and the plumes appeared as jets. In the spring case the SAR image was taken at early flood tide; the plume appeared as an integrated body with relatively uniform bright tunes. In all cases the plume disperses within about 25 km downstream.

Synthetic aperture radar detection of the snowline on Commonwealth and Howard Glaciers, Taylor Valley, Antarctica

AbstractSynthetic aperture radar (SAR) images of Taylor Valley, Antarctica, were acquired in January 1999 in coordination with ground-based measurements to assess SAR detection of the snowline on dry polar glaciers. We expected significant penetration of the radar wave resulting in an offset of the SAR-detected snowline relative to the true snowline. Results indicated no detectable displacement of the SAR snowline. Snow depths of 15 cm over ice can be detected on the imagery. We hypothesize that the optical depth of thin snowpacks is enhanced by reflection and refraction of the radar beam by internal snow layers. The enhanced optical depth increases the volume scattering, and thereby enhances backscatter sufficiently to be detected by the SAR. Consequently, SAR imagery may be used directly to image the position of transient snowlines in dry polar regions.

Ultrawide-band synthetic aperture radar for detection of unexploded ordnance: modeling and measurements

Electromagnetic (EM) scattering from subsurface unexploded ordnance (UXO) is investigated both theoretically and experimentally. Three EM models are considered: the multilevel fast multipole algorithm (MLFMA), the method of moments (MoM), and physical optics (PO). The relative accuracy of these models is compared for several scattering scenarios. Moreover, the model results are compared to data measured by an experimental synthetic aperture radar (SAR) system, SAR images have been generated for subsurface UXO targets, in particular 155-mm shells. We compare SAR images from the measured data with theoretical images produced by the MoM and PO simulations, using a standard back-projection imaging technique. In addition to such comparisons with measurement, we consider additional buried-UXO scattering scenarios to better understand the underlying wave phenomenology.

Synthetic aperture radar detection, recognition, and clutter rejection with new minimum noise and correlation energy filters

Synthetic aperture radar detection, recognition, and clutter rejection with new minimum noise and correlation energy filters

Synthetic aperture radar detection and clutter rejection minace filters

New shift-invariant distortion-invariant Synthetic Aperture Radar (SAR) filters for detection are described. They employ new modified Minimum Noise and Correlation Energy (MINACE) filters. Test results are provided for real SAR data with 360° aspect views of two different objects and for 100 real SAR clutter chips (natural and man-made) that passed the second stage of the Lincoln Laboratory's SAR processor. We obtained high detection rates P D=100% and few false alarms P FA ⩽ 0.82% using only three filters per object class. The shift-invariance of the filters allow their use as adjuncts for all stages (detection, discrimination, classification) of a SAR image processor. Classification results are also briefly reported.

A point target model for the synthetic aperture radar detection of ships and ice conditions during a swell

A running swell affects the synthetic aperture radar (SAR) imagery of ships, smaller icebergs, and other floating objects because the targets signal is no longer matched with the azimuth processor. This effect is analyzed analytically and numerically for the case of conventional image processing.