Resolution Synthetic Aperture Radar Research Articles

Fine-grained ship classification based on deep residual learning for high-resolution SAR images

ABSTRACTAs the resolution of Synthetic Aperture Radar (SAR) images increases, the fine-grained classification of ships has become a focus of the SAR field. In this paper, a ship classification framework based on deep residual network for high-resolution SAR images is proposed. In general, networks with more layers have higher classification accuracy. However, the training accuracy degradation and the limited dataset are major problems in the training process. To build deeper networks, residual modules are constructed and batch normalization is applied to keep the activation function output. Different fine tuning strategies are used to select the best training scheme. To take advantage of the proposed framework, a dataset including 835 ship slices is augmented by different multiples and then used to validate our method and other Convolutional Neural Network (CNN) models. The experimental results show that, the proposed framework can achieve a 99% overall accuracy on the augmented dataset under the optimal fine-tuning strategy, 3% higher than that in other models, which demonstrates the effectiveness of our proposed approach.

Improved range model based on parameter estimation for high resolution and high squint LEO SAR

Due to the curve of the LEO synthetic aperture radar (SAR) orbit and the rotation of the earth, the high resolution imaging for LEO SAR is always facing the problem of obtaining accurate range history. Meanwhile, with the increase of squint angle, it will take longer azimuth time to meet the same high resolution. In this paper, an improved hyperbolic range equation (IHRE) is presented to model the accurate range history for high squint LEO SAR. First, the state vectors of satellite are used to estimate the Doppler parameters. Then, the improved hyperbolic range equation is recovered based on the estimation of the Doppler parameters. Compared with other existing methods, the superiority of IHRE is more precise in the condition of high squint LEO SAR which can provide sufficient azimuth time for high resolution imaging. The simulation and analysis validate the precision of the proposed range history model.

DSM Quality of Korean Satellite KOMPSAT-3 in Comparison to AW3D30 and Sentinel-1A in Respect of Airborne Laser Scanning

With significant technological improvements on earth observation instruments, three dimensional space-borne remote sensing data became indispensable for modelling large areas and continents. To acquire high resolution 3D earth data, actual optical and synthetic aperture radar (SAR) satellite missions are launched frequently. Korean Multi-Purpose Satellite 3 (KOMPSAT-3), which was launched in May, 2012 by Korea Aerospace Research Institute (KARI), is a high-resolution optical observation mission. The 0.7 m ground sampling distance (GSD) and stereo imaging capability of the satellite enables to derive qualified digital surface models (DSM). We generated a 5 m spacing KOMPSAT-3 DSM in Istanbul historic peninsula and estimated its quality by comprehensively analysing the absolute and relative accuracies and the morphological detail description capability. In the analysis, 1 m spacing airborne laser scanning (ALS) DSM was used as reference. Additionally, the ALOS (Advanced Land Observation Satellite) World 3D 30 m (AW3D30) and Sentinel-1A (S-1A) DSMs have been used in comparison with the KOMPSAT-3 DSM. As accuracy metrics, standard deviation (SZ) and normalized median absolute deviation (NMAD) of height differences between KOMPSAT-3, AW3D30, S-1A and ALS reference DSM were used. The results demonstrated that KOMPSAT-3 DEM quality is better than AW3D30 and S-1A both in accuracy and morphologic detail analysis.

Impact of Image Artifact and Solution to the Image Quality Issues in Real Time SAR Images

Basically, the Synthetic Aperture Radar (SAR) images are often degraded due to three factors namely noise, blur and artifact. The noise is the undesirable fluctuation in a random portion of the image and is often detracts from the image. The blur will reduce the object visibility. According to the recent literatures the most dangerous effect which appear in real time images are artifacts. The shadowing effect is the best example to depict the image artifact. The presence of shadows mostly affects the vital information of an image. In the shadowing effect, the portion of the object is totally obscured or hidden from the image. In this paper, we focus the impact of image artifact such as shadow in real time images and we focus how to detect the shadowing effect. Further, this paper is devoted to removal of shadows from very high resolution (VHR) SAR images and aerial view Images.

Time series of flood mapping in the Mekong Delta using high resolution satellite images

The Mekong Delta, located in Southern Vietnam, is one of the most affected areas in the world by climate change and sea level rises, especially flooding. Therefore, flood mapping is essential for understanding the flood regime and mitigating its impacts. Remote sensing and GIS can support the accurate and area wide evaluation of floods. In this study, high spatial resolution synthetic aperture radar (SAR) and optical data were used to generate a dense satellite data time series for analysing the spatio-temporal patterns of flooding in the Mekong Delta. To derive water masks, a total of 777 Sentinel-1, 515 Sentinel-2 and 57 Landsat-8 scenes were used to generate cloud free water masks at a 10m spatial resolution in regular 10-day intervals throughout the observation period of hydrological years 2016-2017. The results show a spatial explicit information on the core zones of the seasonal flooding processes for the entire Mekong Delta and their effect of using floodwater for rice cultivation. The outcome maps provide an overall understanding of Mekong Delta flood patterns and many valuable information for policymakers and water resources managers.

New Insights into Multiclass Damage Classification of Tsunami-Induced Building Damage from SAR Images

The fine resolution of synthetic aperture radar (SAR) images enables the rapid detection of severely damaged areas in the case of natural disasters. Developing an optimal model for detecting damage in multitemporal SAR intensity images has been a focus of research. Recent studies have shown that computing changes over a moving window that clusters neighboring pixels is effective in identifying damaged buildings. Unfortunately, classifying tsunami-induced building damage into detailed damage classes remains a challenge. The purpose of this paper is to present a novel multiclass classification model that considers a high-dimensional feature space derived from several sizes of pixel windows and to provide guidance on how to define a multiclass classification scheme for detecting tsunami-induced damage. The proposed model uses a support vector machine (SVM) to determine the parameters of the discriminant function. The generalization ability of the model was tested on the field survey of the 2011 Great East Japan Earthquake and Tsunami and on a pair of TerraSAR-X images. The results show that the combination of different sizes of pixel windows has better performance for multiclass classification using SAR images. In addition, we discuss the limitations and potential use of multiclass building damage classification based on performance and various classification schemes. Notably, our findings suggest that the detectable classes for tsunami damage appear to differ from the detectable classes for earthquake damage. For earthquake damage, it is well known that a lower damage grade can rarely be distinguished in SAR images. However, such a damage grade is apparently easy to identify from tsunami-induced damage grades in SAR images. Taking this characteristic into consideration, we have successfully defined a detectable three-class classification scheme.

Building Detection in Very High Resolution SAR Images Via Sparse Representation Over Learned Dictionaries

The spatial resolution of synthetic aperture radar (SAR) images has significantly increased in the past few years. As a result, very high resolution images are available for interpretation of the urban area. Buildings are one of the most important parts of the urban area. In this paper, we propose a novel method for building detection from SAR images. First, a set of primary detections, using double bounce, layover, and shadow features, are made. These features are extracted using the order statistics constant false alarm rate method and power ratio detector. The detectors parameters are adjusted in such a way that all bright and dark areas are well extracted. Second, the K-singular value decomposition method is used to learn two dictionaries for discrimination of buildings from clutters. The main idea is to utilize recognition techniques in detection algorithms in order to reduce the number of the false alarms. Gray level statistical, textural, and shape features are used to form the dictionaries. In this step, a secondary set of detections are available, and the final detections are made by the fusion of these two sets. The effectiveness of the algorithm is evaluated by two different TerraSAR-X images from a densely built-up area. The results show that the proposed method is an effective way to detect building through single SAR images.

Staring Spotlight TerraSAR-X SAR Interferometry for Identification and Monitoring of Small-Scale Landslide Deformation

We discuss enhanced processing methods for high resolution Synthetic Aperture Radar(SAR) interferometry (InSAR) to monitor small landslides with difficult spatial characteristics,such as very steep and rugged terrain, strong spatially heterogeneous surface motion,and coherence-compromising factors, including vegetation and seasonal snow cover. The enhancedmethods mitigate phase bias induced by atmospheric effects, as well as topographic phase errorsin coherent regions of layover, and due to inaccurate blending of high resolution discontinuouswith lower resolution background Digital Surface Models (DSM). We demonstrate the proposedmethods using TerraSAR-X (TSX) Staring Spotlight InSAR data for three test sites reflecting diversechallenging landslide-prone mountain terrains in British Columbia, Canada. Comparisons withcorresponding standard processing methods show significant improvements with resultingdisplacement residuals that reveal additional movement hotspots and unprecedented spatial detailfor active landslides/rockfalls at the investigated sites.

Assessment of the High Resolution SAR Mode of the RADARSAT Constellation Mission for First Year Ice and Multiyear Ice Characterization

Simulated compact polarimetry from the RADARSAT Constellation Mission (RCM) is evaluated for sea ice classification. Compared to previous studies that evaluated the potential of RCM for sea ice classification, this study focuses on the High Resolution (HR) Synthetic Aperture Radar (SAR) mode of the RCM associated with a higher noise floor (Noise Equivalent Sigma Zero of −19 dB), which can prove challenging for sea ice monitoring. Twenty three Compact Polarimetric (CP) parameters were derived and analyzed for the discrimination between first year ice (FYI) and multiyear ice (MYI). The results of the RCM HR mode are compared with those previously obtained for other RCM SAR modes for possible CP consistency parameters in sea ice classification under different noise floors, spatial resolutions, and radar incidence angles. Finally, effective CP parameters were identified and used for the classification of FYI and MYI using the Random Forest (RF) classification algorithm. This study indicates that, despite the expected high noise floor of the RCM HR mode, CP SAR data from this mode are promising for the classification of FYI and MYI in dry ice winter conditions. The overall classification accuracies of CP SAR data over two test sites (96.13% and 96.84%) were found to be comparable to the accuracies obtained using Full Polarimetric (FP) SAR data (98.99% and 99.20%).

Tracking Human-Induced Landscape Disturbance at the Nasca Lines UNESCO World Heritage Site in Peru with COSMO-SkyMed InSAR

The “Lines and Geoglyphs of Nasca and Palpa” in Peru are among the most well-known UNESCO World Heritage Sites globally, and an exemplar of site where heritage assets cannot be separated from their natural and anthropogenic environment. The site is exposed to interactions with natural processes, as well as human presence. In this work, 3-m resolution synthetic aperture radar (SAR) StripMap HIMAGE HH-polarised scenes acquired by the X-band COSMO-SkyMed constellation are exploited to track two events of human-induced landscape disturbance that occurred in December 2014 and January 2018. Pre-, cross-, and post-event interferometric SAR (InSAR) pairs characterised by small temporal and normal baselines allow the detection of temporal decorrelation associated with the two events, the extent and time reference of which match with online photographic and video evidence, published literature, web news, and press releases by the Ministry of Culture in Peru. Further elements enhancing the understanding of the 2018 event come from 10-m resolution Sentinel-2B satellite data that reveal the occurrence of apparent changes of surface reflectance due to uncovering of the light grey-yellow clay underneath the darker pebble constituting the fragile surface of the Pampa de Jumana. This scientific study confirms that SAR imagery archives, such as those being built by COSMO-SkyMed for Nasca, prove valuable for the retrospective analysis and digital recording of human-induced landscape disturbance events from space. These archives therefore act as essential sources of geospatial information on the conservation history of heritage sites and assets.

Constraints on the geomorphological evolution of the nested summit craters of Láscar volcano from high spatio-temporal resolution TerraSAR-X interferometry

Small-scale geomorphological changes that are associated with the formation, development, and activity of volcanic craters and eruptive vents are often challenging to characterize, as they may occur slowly over time, can be spatially localized, and difficult, or dangerous, to access. Using high-spatial and high-temporal resolution synthetic aperture radar (SAR) imagery collected by the German TerraSAR-X (TSX) satellite in SpotLight mode in combination with precise topographic data as derived from Pleiades-1A satellite data, we investigate the surface deformation within the nested summit crater system of Lascar volcano, Chile, the most active volcano of the central Andes. Our aim is to better understand the structural evolution of the three craters that comprise this system, to assess their physical state and dynamic behavior, and to link this to eruptive activity and associated hazards. Using multi-temporal SAR interferometry (MT-InSAR) from ascending and descending orbital geometries, we retrieve the vertical and east-west components of the displacement field. This time series indicates constant rates of subsidence and asymmetric horizontal displacements of all summit craters between June 2012 and July 2014, as well as between January 2015 and March 2017. The vertical and horizontal movements that we observe in the central crater are particularly complex and cannot be explained by any single crater formation mechanism; rather, we suggest that short-term activities superimposed on a combination of ongoing crater evolution processes, including gravitational slumping, cooling and compaction of eruption products, as well as possible piston-like subsidence, are responsible for the small-scale geomorphological changes apparent in our data. Our results demonstrate how high-temporal resolution synthetic aperture radar interferometry (InSAR) time series can add constraints on the geomorphological evolution and structural dynamics of active crater and vent systems at volcanoes worldwide.

Monitoring Hydro Temporal Variability in Alberta, Canada with Multi-Temporal Sentinel-1 SAR Data

ABSTRACTFreely available high temporal and spatial resolution synthetic aperture radar (SAR) satellite data such as Sentinel-1 have made it possible for almost near-real time monitoring of surface water extent. We present a method to track temporal variability in surface water extent, hereafter hydro temporal variability (HTV) in Alberta, Canada. Multi-temporal Sentinel-1-C band SAR data was used to classify each pixel in a pixel stack across time into water or non-water. This dataset can tell the percent of time a given 10 m × 10 m pixel was seen as open water. HTV was then summarized by calculating the percentage of the total pixel stack, which was detected as water. Comparison to the waterbodies in the Government of Alberta Base Features Hydrography Polygon dataset shows that the HTV dataset is able to differentiate between permanent and recurring lakes as well as capture rivers with a width of over 30 m. It is anticipated that the methodology presented here will be further enhanced and refined with imagery available for 2018 and beyond, due to now operational Sentinel-1B satellite and future RADARSAT Constellation Mission (2018), which will both provide improved data opportunities.

Robust algorithm for detecting floodwater in urban areas using synthetic aperture radar images

Flooding is a major hazard in both rural and urban areas worldwide, but it is in urban areas that the impacts are most severe. High resolution Synthetic Aperture Radar (SAR) sensors are able to detect flood extents in urban areas during both day- and night-time. If obtained in near real-time, these flood extents can be used for emergency flood relief management or as observations for assimilation into flood forecasting models. In this paper a method for detecting flooding in urban areas using near real-time SAR data is developed and extensively tested under a variety of scenarios involving different flood events and different images. The method uses a SAR simulator in conjunction with LiDAR data of the urban area to predict areas of radar shadow and layover in the image caused by buildings and taller vegetation. Of the urban water pixels visible to the SAR, the flood detection accuracy averaged over the test examples was 83%, with a false alarm rate of 9%. The results indicate that flooding can be detected in the urban area to reasonable accuracy, but that this accuracy is limited partly by the SAR’s poor visibility of the urban ground surface due to shadow and layover.

Derivation of Bistatic SAR Resolution Equations Based on Backprojection

This letter introduces ground-range and cross-range resolution equations for the side-looking bistatic synthetic aperture radar (SAR). The derivation is based on the backprojection integral and the method of stationary phase. The ground-range and cross-range resolution equations are provided in closed form, making them easy for calculation. They are, therefore, helpful for bistatic SAR system development. The derived ground-range and cross-range resolution equations are validated with the bistatic data simulated mainly using the parameters of the LORA system.

Mixture Statistical Distribution Based Multiple Component Model for Target Detection in High Resolution SAR Imagery

This paper proposes an innovative Mixture Statistical Distribution Based Multiple Component (MSDMC) model for target detection in high spatial resolution Synthetic Aperture Radar (SAR) images. Traditional detection algorithms usually ignore the spatial relationship among the target’s components. In the presented method, however, both the structural information and the statistical distribution are considered to better recognize the target. Firstly, the method based on compressed sensing reconstruction is used to recover the SAR image. Then, the multiple component model composed of a root filter and some corresponding part filters is applied to describe the structural information of the target. In the following step, mixture statistical distributions are utilised to discriminate the target from the background, and the Method of Logarithmic Cumulants (MoLC) based Expectation Maximization (EM) approach is adopted to estimate the parameters of the mixture statistical distribution model, which will be finally merged into the proposed MSDMC framework together with the multiple component model. In the experiment, the aeroplanes and the electrical power towers in TerraSAR-X SAR images are detected at three spatial resolutions. The results indicate that the presented MSDMC Model has potential for improving the detection performance compared with the state-of-the-art SAR target detection methods.

Enhanced Resolution Stripmap Mode Using Dynamic Metasurface Antennas

To maintain sufficient signal-to-noise ratio (SNR) for image reconstruction and image interpretation, conventional synthetic aperture radar (SAR) systems must trade off resolution and scene size. This paper proposes a new SAR mode of operation, which improves resolution while maintaining good SNR and a large scene size. It leverages the unique properties of dynamic metasurface antennas (MSAs) to subsample a large virtual beamwidth utilizing multiple small distinct antenna beams. Due to this parallelization in scene sampling, the constraints on the azimuth sampling rate can be relaxed while maintaining an aliasing-free cross range. Due to the versatile properties of MSAs and their cost effective manufacturing process, this paper proposes SAR systems, which can obtain high resolution images over a wide scene size with lower cost and complexity than competing approaches. Point-spread functions and proof-of-concept SAR simulations are shown to verify this approach. In addition, laboratory experiments using a commercial prototype MSA are presented, which show an improvement of 62% in cross-range resolution of the proposed approach, compared with the cross-range resolution of stripmap mode SAR with the same aperture.

Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images

Improving the geo-localization of optical satellite images is an important pre-processing step for many remote sensing tasks like monitoring by image time series or scene analysis after sudden events. These tasks require geo-referenced and precisely co-registered multi-sensor data. Images captured by the high resolution synthetic aperture radar (SAR) satellite TerraSAR-X exhibit an absolute geo-location accuracy within a few decimeters. These images represent therefore a reliable source to improve the geo-location accuracy of optical images, which is in the order of tens of meters. In this paper, a deep learning-based approach for the geo-localization accuracy improvement of optical satellite images through SAR reference data is investigated. Image registration between SAR and optical images requires few, but accurate and reliable matching points. These are derived from a Siamese neural network. The network is trained using TerraSAR-X and PRISM image pairs covering greater urban areas spread over Europe, in order to learn the two-dimensional spatial shifts between optical and SAR image patches. Results confirm that accurate and reliable matching points can be generated with higher matching accuracy and precision with respect to state-of-the-art approaches.

Synthesis of conformal array antenna for hypersonic platform SAR using modified particle swarm optimisation

The synthetic aperture radar (SAR) system equipped on the hypersonic platform in the near space presents many advantages compared with the conventional airborne SAR. However, due to the complex aerodynamic configuration and flight characteristics of the hypersonic platform, the radar design is facing some new challenges. Taking into account the new features of the platform, in this study the authors analyse the conformal design requirements of the array antenna, as well as the prominent contradiction between the high resolution and wide swath of SAR. After establishing the model of the conformal array antenna and the space coordinate system of the far-field radiation, a specific process of synthesising the antenna pattern is given. Furthermore, the antenna mask templates of optimising the antenna pattern are generated based on the main performance measures of the SAR system, and the modified meta particle swarm optimisation algorithm is exploited to obtain the optimal excitation phase of each array element in synthesising the ultimate antenna pattern. Finally, simulation results verify the effectiveness of the method, where the performance of the optimised antenna pattern is evaluated numerically.

Performance of PolSAR backscatter and PolInSAR coherence for scattering characterization of forest vegetation using single pass X-band spaceborne synthetic aperture radar data

Airborne synthetic aperture radar (SAR) data have been successfully used for forest height inversion; however, there is limited applicability in spaceborne scenarios due to high temporal decorrelation. This study investigates the potential of a high-resolution fully polarimetric interferometric pair of TerraSAR-X/TanDEM-X SAR data with no temporal decorrelation to analyze the backscatter and coherence response and to implement polarimetric SAR interferometry-based height inversion algorithms. The data were acquired over Barkot forest region of Uttarakhand state in India. Yamaguchi decomposition was implemented onto the dataset to express total backscatter as a sum of different scattering components from a single SAR resolution cell. Coherency matrix was used to compute complex coherence for different polarization channels. Forest areas suffered from low coherence due to volume decorrelation, whereas a dry river bed had shown high coherence. The coherence amplitude inversion approach overestimated the forest height and also resulted in false heights for this dry river bed. These limitations were overcome by implementing three-stage inversion modeling, which assumes polarization-independent volume coherence. The results were validated using ground truth data available for 49 plots, and the latter was found to be more accurate with an overall accuracy of 90.15% and root-mean-square error of 2.42 m.

Determining Rice Growth Stage with X-Band SAR: A Metamodel Based Inversion

Rice crops are important in the global food economy, and new techniques are being implemented for their effective management. These techniques rely mainly on the changes in the phenological cycle, which can be investigated by remote sensing systems. High frequency and high spatial resolution Synthetic Aperture Radar (SAR) sensors have great potential in all-weather conditions for detecting temporal phenological changes. This study focuses on a novel approach for growth stage determination of rice fields from SAR data using a parameter space search algorithm. The method employs an inversion scheme for a morphology-based electromagnetic backscattering model. Since such a morphology-based model is complicated and computationally expensive, a surrogate metamodel-based inversion algorithm is proposed for the growth stage estimation. The approach is designed to provide estimates of crop morphology and corresponding growth stage from a continuous growth scale. The accuracy of the proposed method is tested with ground measurements from Turkey and Spain using the images acquired by the TerraSAR-X (TSX) sensor during a full growth cycle of rice crops. The analysis shows good agreement for both datasets. The results of the proposed method emphasize the effectiveness of X-band PolSAR data for morphology-based growth stage determination of rice crops.