Search

According to the 2018 report of the United Nations Office for Disaster Risk Reduction (UNISDR), the total number of recorded flood events between 1998 and 2017 was 3,148. This means there is an average of 13 monthly floods on our Earth. Satellite monitoring using Sentinel-1 synthetic aperture radar (SAR) can be effectively used to map floods. It provides a short revisit time, a high spatial resolution of 5 m, and a swath of up to 400 km. In this study, we compare two promising flood mapping algorithms to map floods which are common phenomena in the An Giang area, and to monitor the mapping difficulty of flood peaks. The two mapping algorithms are Global Flood Mapper (GFM), an open-source Google Earth Engine (GEE) application; and the other algorithm implemented the Otsu threshold method and change detection approach. They were visually compared using Sentinel-2 optical data and a land-use land cover (LULC) map of An Giang, Vietnam. The results show that the GFM overcomes the misclassification of flooded and non-flooded areas. A good statistical agreement showed that the performance of flood maps derived from the modified GFM algorithm on Sentinel-1 VH-VV-VH RGB composite and Sentinel-2 images have better accuracy than the Otsu Sentinel-1 technique by $\sim20\% $ ∼ 20 %. [ABSTRACT FROM AUTHOR]

The KAZE algorithm has been popularly used to match the remote sensing images. However, it has limited performance when its standard form is directly applied to match the synthetic aperture radar (SAR) images containing significant multiplicative speckle noise. In this paper, an effective SAR image matching algorithm is proposed, which is a combination of KAZE, phase congruency, and speckle noise removing anisotropic diffusion. In this algorithm, the ratio-based Phase Congruency (PC) information is used to eliminate the erroneous features, which improves the repeatability rate of the KAZE features. In addition, the scale layers of the input SAR images are constructed by speckle noise removing anisotropic diffusion method which significantly reduces the effect of noise. The proposed method can improve the repeatability of the extracted KAZE features. Moreover, it can give better recall and precision values than the state-of-the-art methods. It gives the root mean square error (RMSE) values in the range of 0.881 to 1.104 pixels in SAR image matching. Experiments on multi-temporal SAR images demonstrate the applicability of the proposed method. [ABSTRACT FROM AUTHOR]

The main purpose of this letter is to investigate the performance of the calibration of the Chinese civilian satellite carrying synthetic aperture radar (SAR), denoted as 1mC-SAR, which is a successor of the previous Gaofen-3 (GF-3) satellite. In total, more than 500 1mC-SAR images in co-polarization modes (vertical–vertical [VV] and horizontal–horizontal [HH]) are collocated with the wind products from Haiyang-2 (HY-2B/2C/2D) scatterometers in April – May 2023. The well-developed geophysical model function (GMF) CMOD7 and polarization ratio PR2011 in the C-band are used to simulate the normalized radar cross section (NRCS) employing the wind vector from HY-2. There are more than 2000 matchups for VV-polarization and more than 1000 matchups for HH-polarization. Comparison of the simulations and 1mC-SAR measurements yields a root mean squared error (RMSE) of 3.03 dB, a correlation coefficient (r) of 0.93, a scatter index (SI) of −0.14, and a bias of −2.22 dB for the VV-polarized NRCS, which is slightly better than the 3.68 dB RMSE, 0.90 r, −0.15 SI, and −3.08 dB bias of the HH-polarized NRCS. The variations in the difference in the VV-polarized NRCS (1mC-SAR measurements minus simulations) with respect to the wind speed and incidence angle are analysed. Despite the difference in the VV-polarization (i.e. <2 dB) at wind speeds of greater than 3 m s−1, the difference in the VV-polarization approaches 3 dB at several incidence angles, which needs to be further improved. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) measures backscattering characteristics of man-made targets and has the advantage of not being affected by weather and time. For target recognition in SAR images, the target reconstruction results based on attribute scattering centres are introduced for decision fusion. In comparison with original images, the reconstructed targets better relieve corruptions caused by noises, clutters, etc. A deep learning model, i.e. ResNet, is adopted as a basic classifier to classify both original and reconstructed images. According to the energy relationship between the reconstruction target and residual, the noise level of the original SAR image is defined. Then, the adaptive weights of original and reconstructed images are determined and a weighted decision fusion process is conducted to combine decisions from both images to confirm the target label. The proposed method is tested based on MSTAR dataset, and experimental results show its effectiveness. [ABSTRACT FROM AUTHOR]

The longer the baseline length of an airborne array synthetic aperture radar (SAR) is, the richer the ground object information can be resolved. The airborne array flexible SAR can overcome the limitation that the baseline length is limited by the belly length by attaching the antenna to the wing. However, fusing the airborne array flexible SAR distributed position and orientation system (POS) data under the wing and the belly is a major difficulty. To address this issue, the paper analyzes the inconsistency of airborne array flexible SAR data with a distributed POS and proposes a three-dimensional imaging method of airborne array flexible SAR based on distributed POS by fusing the Doppler centre, yaw angle and time of the observation data. By using distributed POS data, flexible data can obtain higher resolution than single-POS processing. The experiment of airborne array flexible SAR data proves the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Synthetic Aperture Radar (SAR) images have been widely used for surface water identification due to their all-weather capabilities. However, the presence of inherent speckle noise in SAR data poses a challenge for accurate water identification. Additionally, annotating high-quality water body samples requires significant human labour, which can be costly and time-consuming. Aiming at the above problems, a noise-robust automatic water identification architecture without artificial labels is proposed. First, a two-stage automatic sample collection method that utilizes k-means++ clustering and morphological concepts is designed. Then, a weakly supervised noise-resistant SAR water body segmentation method NRM-ACUNet has been developed based on U-Net combined with LNR-Dice loss function and Conditionally Parameterized Convolutions (CondConv) to minimize the impact of sample noises. Experimental results show that the morphological processing can improve water body sample quality compared to k-means++, and compared with U-Net, NRM-ACUNet performs superior with noise-containing pseudo-samples, achieving 96.8% F1 accuracy and 52.06% accuracy improvement. [ABSTRACT FROM AUTHOR]

Drone-based synthetic aperture radar (SAR) analysis offers a simple and efficient technique for acquiring observational data from various research areas. However, fundamental research on observation techniques, including the investigation of drone flight parameters, remains insufficient. This study focuses on improving the resolution of intensity images of a single-look complex (SLC) via drone-based SAR observations. By conducting experiments at different flight speeds and analysing the obtained SLC images, we demonstrated that higher drone speeds resulted in enhanced image resolution under the conditions investigated at flight speeds of ≤ 5 m s−1. A comparison with satellite SAR images highlighted the superior performance of drone-based SAR in capturing topographic features. The analysis also suggested that factors including trajectory deviation and Doppler rate affected the resolution and azimuth position. The simulation results supported the experimental findings on the relationship between drone speed and resolution. Our results emphasize the potential of drone-based SAR observations for monitoring slope failures and suggest the importance of the drone flight parameter selection. In the future, drone-based SAR is expected to produce intensity images of slopes and locate slope deformations associated with blasting for mining and earthquakes. [ABSTRACT FROM AUTHOR]

The mining-induced subsidence has caused severe casualties and poses a threat to the safe operation of coal extraction, especially under the roads, buildings and rivers. This alarming situation demands the large-area subsidence detection and the large-gradient deformation monitoring to ensure safety. Although the traditional Interferometric Synthetic Aperture Radar (InSAR) technology can monitor surface deformation in high accuracy, unfortunately, it often fails to recover the sufficient deformation when facing the effects of decorrelation and large gradient displacement. Therefore, this study firstly proposes a Stacking-InSAR and coherence gradient fusion method to detect the mining-induced subsidence over large area. Secondly, the robust offset tracking method is applied to retrieve the large-gradient deformation in Buertai mining site. Both Gaofen-3 and Sentinel-1A Synthetic Aperture Radar (SAR) datasets are acquired over the border of Shaanxi and Inner Mongolia, China, for experimental test. The results show that 179 mining-induced subsidence sites are detected with the proposed method, and the complete spatial temporal deformation fields are obtained by both Gaofen-3 and Sentinel-1A offset tracking method with maximum deformation of 3 m in range direction. This study provides a valuable tool for detecting the fine boundary of mining subsidence and recovering the accurate deformation time series. [ABSTRACT FROM AUTHOR]

Radar forward-looking imaging has gained increasingly significance in diverse applications, including battlefield reconnaissance, target surveillance, and precision guidance. While Synthetic Aperture Radar techniques are widely employed to achieve high azimuth resolution, the domain of forward-looking imaging has its challenges. These challenges arise due to the minimal Doppler frequency changes in forward-looking areas and the symmetrical Doppler history about flight path. These constraints negatively impact the radar's ability to accurately detect and image targets in forward-looking areas. Consequently, there has been a growing interest in the development of novel approaches to these problems. Due to the application of deep learning, image super-resolution methods based on convolutional neural networks (CNNs) have received extensive attention and demonstrated good performance. This avenue of research provides fresh insights into improving the azimuth resolution of radar imaging. In this letter, we propose a novel convolutional neural network, called MSUS-Net (Multi-scale U-shaped Network), for radar forward-looking imaging. It combines asymmetric feature fusion with a unique connection structure to replace traditional layers. Moreover, different from conventional image restoration networks that use content loss, we use the L1 distance in the frequency domain, restoring missing high-frequency details. Simulated and real radar data validate that this approach offers robust recovery and better imaging performance. [ABSTRACT FROM AUTHOR]

Deep learning methods have greatly promoted the ship detection performance in synthetic aperture radar (SAR) images. However, due to the different imaging mechanisms and target characteristics, detecting ships in SAR imagery precisely and efficiently is still challenging. To address this issue, a streamlined rotational ship detection network (SRSD-Net) is proposed in this letter. First, we construct a lightweight network using one-level feature and a receptive field block (RFB) to extract ship features with rich context. Second, a simple and effective representation for rotated ship bounding boxes using the minimum external horizontal box and two phase factors is proposed. Finally, to reduce the missed detections of small ships, k-nearest label assignment (KNLA) is designed to make all targets equally contribute to the network training. Extensive experiments are conducted on high-resolution SAR images dataset (HRSID) and the results show that the proposed method can achieve high detection accuracy with low computational cost. [ABSTRACT FROM AUTHOR]

This letter presents a novel method for synthetic aperture radar (SAR) image change detection using the global patch similarity-based graph (GPSG). First, the SAR image is divided into a number of square patches, which are then vectorized and stacked to form a global patch matrix (GPM). The GPSG is constructed by connecting the global similar patches that are selected through traversing the GPM. With the support of GPSG, the second-order random walk matrix is designed to widely aggregate the attributes of patches within the two-hop global neighbourhood. The difference image with good separability can be generated by comparing bitemporal patches and their aggregated attributes. Finally, the Otsu thresholding is adopted to obtain the change map showing changed and unchanged portions. Experiments conducted on three real SAR datasets demonstrate the superiority of the proposed GPSG method in terms of both robustness and accuracy for change detection. [ABSTRACT FROM AUTHOR]

A 3-D scattering centre model-based synthetic aperture radar (SAR) automatic target recognition (ATR) method is proposed. Multi-level dominant scattering areas (DSAs) are generated from the test sample and model predicted target images, which can describe the target region and scattering centres distribution from coarse to fine. The DSA matching is conducted at each level using the morphological erosion operation and Euclidean distance transform. Finally, the energy of the region residuals is taken as the distance measure between the test sample and different target models so as to confirm the target label. The 3-D scattering centre model can provide very complete descriptions of the target. In addition, the designed distance measure comprehensively considers the possible nonlinear transformations between the model and test image using the morphological erosion operation and Euclidean distance transform. Experiments are performed on the electromagnetic (EM) simulated data of three military targets. [ABSTRACT FROM AUTHOR]

Rice plays a significant role in regional food supply and international food trading, especially for Asian countries. However, accurate mapping of tropical rice cultivationis a challenging task due to the flexible planting calendar and the complicated topography. With the all-day and all-weather imaging ability, Synthetic Aperture Radar (SAR) provides an encouraging resolution for this task. In this research, we presented a case study in Hainan, China that identify the cultivation patterns of rice. As the only tropical province in China, Hainan has abundant crop resources and diverse rice cultivation practices. First, our previously proposedrice mapping method based on time-series Sentinel-1 data and U-Net modelwas applied to Hainan to generate the candidate rice fields in 2019. Then, a start of season (SOS) detection strategy was proposed to discriminate rice cultivation patterns. The accuracy of the annual rice map and the multi-season discrimination results were validated in pixel level and parcel level, respectively. The annual rice map achieved an overall accuracy of 95.03%. Besides, for 313 out of 360 rice parcels, the cultivation patterns were correctly identified. These results proved the effectiveness of phenology information of time-series SAR, and the proposed scheme will be easily extended to complex tropical agronomy zones. [ABSTRACT FROM AUTHOR]

Quickly and accurately obtaining information on rice planting areas is crucial for maintaining food security and social stability. Based on the time series of Sentinel-1A images, this study proposes an automatic rice mapping method called the Dynamic Threshold Time-Weighted Dynamic Time Warping method (DT-TWDTW). This method consists of two parts. First, rice samples are extracted automatically by a dynamic threshold method based on the unique flooding signal before and after rice transplanting. Then, the Time-Weighted Dynamic Time Warping (TWDTW) method is used for rice mapping based on the cumulative cost matrix between unknown pixels and rice samples. The results show that the DT-TWDTW method can achieve satisfactory mapping accuracy with an overall accuracy of 94.6%, higher than random forest (89.9%), and support vector machine (91.7%). This study provides a new approach and perspective for remote sensing monitoring of rice, especially for some regions without prior phenological knowledge and collection of the rice samples. [ABSTRACT FROM AUTHOR]

The increasing imaging capability of spaceborne Synthetic Aperture Radar (SAR) systems makes a challenging task in matching slant-range (SR) SAR images acquired from different orbit and side-viewing directions. However, currently existing algorithms for the SR SAR images matching are unsatisfactory and the correct homologous points cannot be matched. In this work, a robust and adaptable high-precision method for matching flipped SAR images based on oriented descriptors is proposed. Firstly, the ratio gradient calculation based on Gabor odd filter is introduced. Secondly, the oriented descriptors are built to eliminate the diversity of feature-based descriptors of homologous points of flipped SR SAR images, enabling directly match the flipped SAR images. Finally, the feasibility and reliability of the proposed method are verified by the experimental results of simulated flipped SAR images and actual flipped SAR images, and two directional descriptors of the slave image are constructed to match non-flipped SAR images. The results show that the proposed method performs better in precision with the root mean square error (RMSE) less than 1 pixel, and the proposed oriented descriptors can match SR SAR images. [ABSTRACT FROM AUTHOR]

Tropical cyclones (TC) are the most intense interactions between the ocean and atmosphere. Synthetic aperture radar (SAR) can capture high resolution marine atmospheric and oceanic features. Therefore, high-resolution sea surface wind information can be retrieved from SAR imagery through various methods. In this study, a new modified local gradient (ILG-Mod) method is proposed for TC wind direction retrieval. It combines SAR image smoothing and local gradient computation in the spectral domain, and changes the way to obtain the main local gradient direction in the spatial domain. Nineteen TC images acquired from Sentinel-1 SAR have been collected for validation. The proposed method is compared with the local gradients (LG) method, modified local gradients (LG-Mod) method, and the improved local gradients (ILG) method. In addition, ERA5 (the fifth generation European Centre for Medium-Range Weather Forecasts) reanalysis data and CCMP (the Cross-Calibrated Multi-Platform) data are used for validation. Statistical results show that wind directions retrieved by the ILG-Mod method are highly consistent with the validated data. Comparisons with validated data reveal bias, root mean square error (RMSE), and correlation coefficient (r) values of 17.09°, 19.63° and 0.96, respectively. The proposed ILG-mod method is therefore suitable to retrieve wind direction of TCs. [ABSTRACT FROM AUTHOR]

To improve the accuracy of canola identification in the southwest cloudy and foggy mountains of China, this letter puts forward a group of canola recognition features with the multi-temporal Sentinel-1 images. The Jeffries-Matusita (J-M) distance of sample types and the accuracy of canola extraction of support vector machine were compared. The results showed that the J-M values of canola crop with forest land and other green vegetation were improved obviously, among which the J-M value of forest land category increased to 1.93, and that of other green vegetation 1.82. The producer's accuracy (PA) of canola was improved to 70.93%, and the user's accuracy (UA) was increased to 75.86% with F1-measure 73.31%. The canola crop area recognition accuracy was 82.58%. The feature combination can enhance the distinguishability among sample categories, improve the extraction accuracy of canola crops, but the results still have a gap with the reliable results proposed by previous scholars due to the complex crops planting structure, the irregularity of the plot and the resolution of Synthetic Aperture Radar (SAR) images. This research can provide a reference for the distribution extraction of canola in mountain areas. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) provides high-resolution observations day and night, whose resulting images can be interpreted for different applications. For the SAR automatic target recognition (ATR) problem, this letter proposes a multi-view method based on adaptive decision fusion. The joint sparse representation (JSR) model is first employed to classify the multiple views. For the output decisions from different views, adaptive weights are determined based on Shannon entropy theory. The resulting weights are used for decision fusion to linearly combine the individual decisions from different SAR images to determine the target label. The MSTAR dataset is used for the experiments, on which both the standard operating condition (SOC) and two representative extended operating conditions (EOCs) are setup. By comparison with several state-of-the-art multi-view SAR ATR methods, the validity and robustness of the proposed method can be effectively confirmed. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) plays a vital role in change detection (CD) analysis due to the ability to produce remote sensing images throughout the day, regardless of weather conditions. Nowadays, deep learning methods have gained popularity in multitemporal SAR image CD applications. However, false labels generated during the preclassification stage limit the performance of the CD process. This work employs a fast and robust fuzzy c-means clustering to generate the pseudo-label samples and lightweight spatial pyramid pooling attention network (SPPANet) to detect changes in multitemporal SAR images. The spatial pyramid structure in SPPANet applies adaptive pooling layers to provide better contextual information without incurring computational overhead. The log-ratio operator is used to generate the difference image (DI), and the pseudo-label samples are created from DI. The pseudo-label samples are used to create the training and testing patches. Finally, the trained SPPANet is used to classify testing samples into unchanged and changed classes. The SPPANet achieves an accuracy of 98.70%, 99.06%, 96.40%, and 99.10% for Ottawa, San Francisco, Yellow River, and Farmland datasets, respectively. [ABSTRACT FROM AUTHOR]

The carrier platform in synthetic aperture radar (SAR) imaging may deviate from the correct trajectory due to end currents in the atmosphere, causing motion errors and ultimately degrading the quality of the radar image. Techniques for motion compensation can reduce the impact of motion errors on the results of the imaging process. Motion error parameters are necessary for motion compensation algorithms. In this paper, the line of sight (LOS) error of the carrier platform is estimated based on the vortex-selected SAR imaging system by analysing the vortex SAR echo data using the relationship between the Bessel magnitude term, the phase term, and the motion error. The validity of the method is verified through numerical simulations. [ABSTRACT FROM AUTHOR]

The aim of our work is to improve the modulation transfer function (MTF) of velocity bunching by the quad-polarized GF-3 synthetic aperture radar (SAR). In total, more than 10,000 Gaofen-3 (GF-3) images acquired in wave mode (WV) in quad-polarization are collocated with ocean waves simulated from WAVEWATCH-III (WW3) model and ocean currents from the HYbrid Coordinate Ocean Model (HYCOM). The validation of WW3-simulated significant wave height (SWH) against the measurements from National Data Buoy Center (NDBC) of National Oceanic and Atmospheric Administration (NOAA) yields a 0.48 m root-mean-square error (RMSE) of SWH with a 0.88 correlation coefficient (r). The winds are inverted using the geophysical model function (GMF), denoted as CSARMOD-GF, which is specially for the re-calibrated GF-3 SAR in vertical-vertical (VV) polarization. The azimuthal cut-off wavelength λc caused by velocity bunching is theoretically calculated by wave spectrum from WW3 model and practically estimated from SAR intensity image, respectively. It is found that the velocity bunching is independent with polarization; however, the bias (theoretic minus SAR-derived λc) increases with sea state growing and has no obvious relationship with current. Following this rationale, the MTF of velocity bunching is empirically constructed by including the term of wind speed and the analysis shows a 3.76 m RMSE with a 0.92 r through comparing the SAR-derived λc in VV polarization and simulated values. Under this circumstance, the accuracy of wave retrieval is anticipated to be improved when using the proposed MTF of velocity bunching. [ABSTRACT FROM AUTHOR]

The traditional multi-view synthetic aperture radar (SAR) automatic target recognition (ATR) is based on the assumption of viewpoint integrity of the dataset, requiring that the viewpoint information of all samples participating in the recognition is complete. To address the problem that SAR ATR is difficult under the training condition of an incomplete view sample set, a new cross-view SAR target recognition method based on view correlation extension and multi-task learning is proposed. To mine the complementary information between multi-view SAR images, the training datasets are clustered into subsets of different views based on the correlation coefficient. To improve the accuracy of cross-view recognition, we adopt the strategy of multi-view subsets corresponding to multi-task kernel sparse representation. For the decision values of each view subset, the fusion error is obtained by linear weighting. Finally, the total reconstruction errors accumulated from all tasks are identified. Experimental results on the MSTAR database in cross-view conditions show that our method outperforms the widely-used SAR ATR methods in several performance indicators. [ABSTRACT FROM AUTHOR]

Although numerous classification algorithms have been successfully applied to synthetic aperture radar (SAR) automatic target recognition (ATR), the performance of most methods is limited by the lack of SAR images under different views. Considering traditional generative adversarial networks (GAN) generate fake SAR images same as the input, a novel cross-view GAN (CVGAN) is proposed to generate SAR images under disparate views for enhancing view diversity. In this network, a generator is first used to learn the mapping relations of image pairs to generate cross-view SAR images. Then, a pre-trained classifier is utilized to improve the view angle and category authenticity of the generated images. Finally, image pairs are fed into the discriminator to distinguish the generated SAR images from the real SAR images. To make the proposed network better learn the mapping relations of cross-view images, a modified loss function is extended to our network. The experimental results on the moving and stationary target acquisition and recognition (MSTAR) dataset demonstrate the effectiveness of the CVGAN on SAR image generation. [ABSTRACT FROM AUTHOR]

Digital beamforming with null steering in elevation can be used to separate echoes corresponding to different sub-pulses, but the pulse extension loss will decrease performance on echo separation. To resolve this problem, an improved null steering beamformer based on multi-group time delays is proposed. In the proposed beamformer, the phase weighting matrix is multiplied and the multi-group time delays are used to improve the coherence between echoes from different channels. Then, the modified phase weighing matrix for null steering is introduced to further improve the capacity of echo separation. Compared with the conventional null steering beamformer, the multi-group time delays can provide the more accurate time delay compensation for better echo separation. Simulation results on both point and distributed targets validate the proposed beamformer. [ABSTRACT FROM AUTHOR]

Sustainable oil palm production is a key issue in global biodiversity conservation and sustainable development. As one of the world's major vegetable oil crops, oil palm has expanded exponentially to meet increased demand over the past decades. However, previous monitoring and assessments of oil palm plantations were hampered because of the lack of high-resolution annual maps at the global scale. We produced annual oil palm plantation maps in 4 major producer countries (Indonesia, Malaysia, Thailand and Papua New Guinea) in Asia-Pacific from 2007 to 2018 at 100-m resolution using advanced remote sensing techniques with Phased Array L-band Synthetic Aperture Radar (PALSAR) and Moderate Resolution Imaging Spectroradiometer (MODIS) data. We uncover the global patterns of oil palm expansion and find that global oil palm expansion has a very high degree of potential conflict with local biodiversity. Globally, 99.9% of oil palm plantations overlapped with Conservation Priority Zones (CPZs) and oil palm plantations encroached on 231 protected areas. We suggest to incorporate the related issues into the Post-2020 Global Biodiversity Framework. [ABSTRACT FROM AUTHOR]

In this paper, an essential modulation transfer function (MTF) on synthetic aperture radar (SAR), denoted as tilt modulation, is improved through five vertical–vertical (VV) polarized Gaofen-3 (GF-3) SAR images acquired in Global Observation (GLO) and Wide ScanSAR (WSC) mode that were captured during the 2017 tropical cyclone (TC) season in the western Pacific Ocean. These images are reprocessed to be Level-1B products and collocated with the wind fields from Global and Regional Assimilation and Prediction System – Typhoons Model. More than 9,000 matchups are exploited for studying the dependence of tilt on the wind speed and incidence angle. It is found that quantitative tilt MTF (herein denoted as the derivative scattering coefficient), which is proportional to first-order derivative normalized radar cross-section with respect to incidence angle, is inversely related with the incidence angle between 14° and 46°. In addition, the wind speed has an obvious influence on the derivative scattering coefficient due to each scattering facet essentially being modulated by strong wind-induced long-waves, whereas traditional tilt MTF does not conform to the actual situation. However, the sensitivity of wind speed on the derivative scattering coefficient weakens at larger incidence angles (>40°). Following this rationale, the empirical tilt MTF in VV polarization has been developed. The reanalysis results indicate that the values from empirical tilt MTF could depict the role of wind. The improvement herein could overcome the limitation in tilt MTF, namely, the lack of a strong wind's influence, and it is expected to be implemented for SAR wave retrieval in TCs. [ABSTRACT FROM AUTHOR]

Synthetic Aperture Radar (SAR) plays an important role in marine ship management due to its all-weather, multi-band, and multi-polar characteristics, but the complex environment on the ground and the limitation of the imaging principle lead to the existence of large noise interference in SAR imaging, which poses a greater challenge to the existing ship detection algorithm. To address the problem of low detection rate and high false alarm rate of ship detection in noise-laden SAR images, we propose a detection algorithm based on convolutional neural network and transformer. First, we design the processing structure of multi-scale feature maps based on the advantages of transformer global processing capability, then we use weighted fusion of different scale feature maps after global processing to enhance the detection capability of the model for noise-containing targets, and finally, the model is evaluated using the high noise SAR datasets (HNSAR) we constructed. The experimental results show that our model has better detection of noise-containing SAR images compared with other deep learning models. [ABSTRACT FROM AUTHOR]

The sparse array imaging radar could largely reduce the system complexity, but it also suffers from poor azimuthal resolution due to the limited aperture length and incomplete observation caused by non-uniform array configuration. In this letter, a radar forward-looking imaging method with improved imaging quality and reduced system complexity is proposed. Firstly, the matrix completion method is utilized to recover the missing signals of sparse array. Then, based on the plane-wave assumption, a fast 2-dimensional atomic norm minimization (ANM) is adopted to solve the positions and amplitudes of scatterers. As a continues compressed sensing method, the result of ANM-based method earns sharper main-lobes and much lower-level side-lobes compared to the fast Fourier transform -based method. Although the resolution of proposed method is still limited by Rayleigh resolution, the imaging quality can be greatly improved by emphasizing main-lobes and avoiding the perturbation of side-lobes. Simulations and experiments are carried out to demonstrate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

Pixel offset tracking (OT) algorithm is a useful tool for measuring large surface displacements by matching amplitudes in master and slave synthetic aperture radar (SAR) images. However, strong backscatters can cause homogeneous errors within a matching window (referred to as patch-like errors) in traditional OT processing, thereby misleading the interpretation of displacement events, especially over a small area. In this letter, we proposed an improved SAR OT algorithm to reduce patch-like errors. In which, a logarithmic transformation was firstly utilized to narrow the SAR amplitude range between strong and weak back scatterers. Strong backscatters causing patch-like errors were then statistically detected with an indicator of median absolute deviation. Finally, those strong backscatters were excluded from SAR OT processing using a weighted normalized cross-correlation scheme, in order to reduce the caused patch-like errors. Two real data tests over the Shuozhou and Yulin coal mining areas, China, suggest that the mean accuracy of the displacements estimated by the presented method improved about 30%, with respect to that estimated by the traditional OT algorithm. The proposed SAR OT algorithm offers a robust option to measure large displacements, especially over a small area, associated with anthropologic or geophysical activities. [ABSTRACT FROM AUTHOR]

In the global navigation satellite system-based synthetic aperture radar interferometry (GNSS-based InSAR) system, 3D deformation retrieval needs to associate multiangle images, but due to the low 2D resolution, traditional texture-based algorithm cannot be used in GNSS-based system. At the same time the association error introduced by the elevation error also cannot be compensated for. In this paper, a multiangle images aggregation analysis and association algorithm is proposed under elevation error conditions. Instead of using texture information, focus positions from different angles are adopted here. First, the focus position with elevation error is modelled in the bistatic configuration. Then, the images association is achieved based on the aggregation analysis, which is further quantitatively expressed through minimum circle principle. Raw data from four satellites are used to demonstrate the validity of the proposed algorithm. [ABSTRACT FROM AUTHOR]

This paper presents a new methodology for detecting bare rough soil in non-urban areas using synthetic aperture radar data. Our approach involves matching the covariance/coherence matrix to the extended Bragg (xBragg) model. Our method uses automatic OTSU's thresholding based on roughness angles histogram to determine output classes and avoid complex calculations, making it suitable for onboard computing. The procedure is suitable for a range of soil roughness but not for highly specular surfaces. The performance of the suggested methodology was evaluated using simulated data and real L- and P-band SAR datasets from two different sites with various features. Our results demonstrate the effectiveness of the method for rough agricultural and natural surfaces, with good performance and a success rate of up to about 95 % across a wide range of roughness values from 0 ∘ to 65 ∘ . [ABSTRACT FROM AUTHOR]

RetinaNet has been widely used in the field of ship detection in synthetic aperture radar (SAR) images due to its simple structure and excellent performance. Although various modified RetinaNet models for SAR ship detection have been proposed, there are still some problems, such as how to capture more complete ship features and the unbalanced distribution of positive samples in different tasks, which have not been mentioned in previous studies. Aiming at the above problems, this paper proposes a modified RetinaNet detector for SAR ship detection, namely D 2 ADet. Specifically, a modified feature pyramid network is proposed to capture adaptive features that are more suitable for the ship. Then, a dual-activation domain training method is designed to decouple the sample assignment processes of different tasks to meet their requirements of positive samples for each task. Experimental results show that compared with RetinaNet, D 2 ADet obtains 7.67 % , 11.57 % and 7.28 % gains in AP50, AP75 and mAP on the public SAR ship detection dataset (SSDD), respectively. And D 2 ADet achieves the state-of-the-art performance compared with other methods. [ABSTRACT FROM AUTHOR]

Helicopter-borne wide-angle synthetic aperture radar (SAR) can be used in a wide range of military and civil applications because of their high resolution and mobility. The absolute geometric positioning accuracy is an important index for SAR image quality assessment. However, the effectiveness of using traditional geometric calibration methods with helicopter-borne SAR images obtained by time-domain imaging algorithms remains to be verified. Based on the range-Doppler model, a high-precision geometric calibration method for helicopter-borne SAR is proposed in this study and verified by experiments using real data with different resolutions. The results show that after calibration, the positioning accuracy of SAR images with 20 m and 1 m resolutions is improved from 51.33 m to 4.71 m and from 62.49 m to 3.03 m, respectively. These results verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In view of inverse synthetic aperture radar (ISAR) imaging, sparse aperture not only affects the imaging quality of traditional range Doppler (RD) algorithm but also increases the difficulty of translational motion compensation (TMC). For the purpose of reducing the influence of motion errors on the target imaging, reweighted alternating direction method of multipliers (RADMM) is proposed to realize the joint processing of phase adjustment and ISAR imaging. Combined with pattern-coupled sparse structure (PCSS) information of the target, the reweighted l 1 norm minimization problem is established first. Then, the minimum entropy method is adopted to estimate the phase error, and the iterative process of RADMM is deduced. Finally, the experimental results based on two sets of measured data demonstrate the validity of the proposed algorithm under the conditions of noise and sparse aperture. [ABSTRACT FROM AUTHOR]

We study the shape of convective rolls in the Marine Atmospheric Boundary Layer from Synthetic Aperture Radar images of the ocean. We propose a multiscale analysis with structure functions which allow an easy generalization to analyse high-order statistics and so to finely describe the shape of the rolls. The two main results are 1) second-order structure function characterizes the wavelength and direction of rolls just like correlation or power spectrum do, 2) high-order statistics can be studied with skewness and flatness which characterize the asymmetry and intermittency of rolls, respectively. This is the first time that the asymmetry and intermittency of rolls are shown from radar images of the ocean surface. [ABSTRACT FROM AUTHOR]

Terrain-induced variations of radar backscatter represent an important limiting factor of many Synthetic Aperture Radar (SAR)-based applications. Radiometric terrain flattening (RTF) is a well-established method that minimizes these variations in SAR imagery. To fully understand the implications of SAR RTF, validation of its impact on the derived products is needed. In this study, we quantified the influence of the RTF on a forest mapping and classification algorithm over Austria, and compared the classification results for the conventional sigma naught and radiometrically terrain-corrected gamma backscatter. The overall accuracy for forest/non-forest mapping and forest type classification improved by 2% and 4%, respectively, over the whole of Austria, with improvements of up to 16% and 20%, respectively, in regions with strong topography. [ABSTRACT FROM AUTHOR]

Spaceborne P-band full-polarimetric wideband synthetic aperture radar (SAR) will be affected by ionospheric phase dispersion and polarimetric dispersion. An improved contrast optimization autofocus (COA) method using four polarization channel data is proposed to compensate for both phase dispersion and polarimetric dispersion. The basis and steps of the method are given. Its performance is analysed by typical target simulation under clutter background and compared with traditional COA method for single channel. The residual quadratic phase error (QPE) of single channel compensated by traditional method may exceed 45° focusing threshold, while the proposed method can reduce the error to less than 10°. Due to the lack of spaceborne P-band SAR data, we simulate the effect of ionosphere on P-band SAR based on ALOS2 PALSAR2 data and verify the proposed method. The simulation result shows the proposed method is effective and superior to traditional method. [ABSTRACT FROM AUTHOR]

Remote Sensing-based multi-time point imagery helps incorporate the potentiality of time series analysis in crop classification studies. Single timepoint imagery has limitations in this aspect due to a comparatively similar spectral signature of crops. Even in multi-time point imagery, the viability of optical datasets is thwarted by clouds in crop growing seasons. Hence, researchers have used the all-weather imaging capabilities of Synthetic Aperture Radar (SAR) for crop classification. This all-weather capability of SAR and the multispectral capability of Optical datasets combined with their time series data hold great potential in remote sensing-based crop classification. Nevertheless, the underlying patterns of the combined SAR and optical time series haven't been meticulously explored, especially for socio-economically important crops like sugarcane. Therefore, this study proposes using the Sentinel-1 and Sentinel-2 time-series datasets and crop phenology information to develop a Neural Network capable of identifying sugarcane crops in the sugar belts of Western-Uttar Pradesh. Studies are done for the year 2020 using a Long short-term memory (LSTM) neural network obtaining high accuracies in classifying sugarcane and non-sugarcane classes using optical and SAR time-series analysis datasets. The synthetic dataset has also been developed to enhance the convergence of the deep learning model, producing higher accuracy than utilizing the original dataset. Consequently, the study contributes to remote sensing-based crop classification and sets a foundation for further advancements in this domain. [ABSTRACT FROM AUTHOR]

Change detection is an essential step for synthetic aperture radar (SAR) image interpretation. We propose a change detection method that included a denoising method and a difference operator to improve the change precision. The SAR images are first transformed into logarithmic domain before filtering. And then, the log-intensity image is filtered by the iterative guided filter that is the combination of a traditional guided filter and iterative regularization technique. The operator based on the log-ratio (LR) and neighbourhood pixels is developed to generate the difference image, which can reflect the change trends between images effectively. The final image that consisted of changed pixels and unchanged pixels is obtained by using the fuzzy c-means (FCM) algorithm. The experimental results accomplished over two real SAR datasets demonstrate that the proposed method can improve the accuracy of change detection. [ABSTRACT FROM AUTHOR]

This paper proposes a new ship detection algorithm via saliency analysis on multiscale superpixels. First, Synthetic Aperture Radar (SAR) imagery is over-segmented into many homogenous regions, called superpixels. In contrast to prior work, these superpixels are generated with various scales to handle size variation of ships in the SAR imagery. At each scale, local contrast and ship geometrical characteristics are considered to enhance the saliency of ship pixels. Then, the final saliency map is computed by integrating saliency values of all scales. Experimental results on two real SAR images demonstrate the efficiency of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Due to all-weather and all-time work characteristics, synthetic aperture radar (SAR) images have been widely used in remote sensing. There is great difficulty in understanding SAR images because they are quite different from optical images in imaging mechanism, geometric characteristics, and radiation characteristics. It can greatly improve the readability of SAR images if we can translate them into optical image styles. In this paper, we propose a variational generative adversarial network for SAR images to optical images translation (S2O-VGAN). To demonstrate the validity of the proposed model, a new large-scale dataset called SARGB is proposed. Experimental results based on the proposed dataset with several evaluations show the superiority of the proposed model over the existing methods on SAR-optical image translation. [ABSTRACT FROM AUTHOR]

The empirical mode decomposition (EMD) based method for mitigating radio frequency interference (RFI) has the capability of removing narrow-band RFI and the advantage of easy-to-implement. The signals are decomposed by EMD into a finite number of intrinsic mode functions (IMFs) in a pulse-by-pulse manner, and then the IMFs corresponding to the RFI are subtracted to achieve interference mitigation. However, useful signals will be lost by using this method, which reduces the synthetic aperture radar (SAR) imaging quality. In this letter, a modified EMD method is proposed to tackle this problem. Firstly, the interferences of the echoes are detected through the relative energy ratio. Secondly, echoes containing RFI are decomposed into many IMFs via EMD. These IMFs are transformed into time-frequency (TF) domain via short-time Fourier transform (STFT), and then IMFs corresponding to RFI are screened out via the robust bisecting k-means algorithm. Finally, the RFIs in TF domain are located with the robust OTSU algorithm and mitigated by a notch operation. Experimental results of real SAR data with simulated RFIs demonstrate the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Distributed scatterer interferometry (DS-InSAR) is an important time-series technique of surface deformation monitoring. The accuracy of the results highly depends on the density of DS, which is mainly influenced by synthetic aperture radar (SAR) image resolution and land cover. In this work, firstly, we use split spectrum technique to enable construction of SAR images with different bandwidth (i.e. different resolution). With the land-cover map as classification basis and mean coherence as DS selection index, we quantitatively analyse the behaviour of DS pixel density on different resolution SAR images and three land-cover types. We find that the DS density increases linearly with the increase of SAR system bandwidth, but the linear coefficient decreases with the increase of mean coherence. Secondly, we extend the coherence framework proposed by Zebker in 1992 to establish a model of relationship between mean coherence and bandwidth. The DS density by three land cover and different bandwidth SAR images was predicted by this model. When the mean coherence threshold is set to 0.4, the prediction model can agree with empirical results within 50% difference, and rather closer on impervious surface. [ABSTRACT FROM AUTHOR]

To handle synthetic aperture radar (SAR) image target recognition problem, a multi-view method is proposed. For the multi-view SAR images be recognized, they are first clustered based on the correlation coefficients and divided into several view sets. Afterwards, for the view set containing two or more images, the multiset canonical correlation analysis (MCCA) is employed to fuse them as a single feature vector. For the view set with a single image, its corresponding feature vector is directly used. Finally, the joint sparse representation is used to characterize and classify the feature vectors from different view sets and determine the target label of the multi-view SAR images. Experiments and analysis on the moving and stationary target acquisition and recognition (MSTAR) dataset show that the proposed method can achieve an average recognition rate of 99.42% for 10 types of targets under the standard operating condition (SOC). Its performance is also better than several reference methods under the extended operating conditions (EOC) including noise interference and target occlusion. [ABSTRACT FROM AUTHOR]

Automatic target recognition for Synthetic Aperture Radar (SAR) images is a very challengeable problem due to the lack of enough training examples. Currently, there are two options for a network to learn on small SAR data sets: to increase the depth and width or to use the Capsule Network (CapsNet). To take advantage of both options, this paper proposes a Dense CapsNet for SAR target recognition. The contributions are mainly twofold. First, we replace the original convolutional layers with dense blocks in order to increase the depth of the network. Second, we substitute deconvolutional layers for fully connected layers to build the reconstruction network, which is beneficial for improving training efficiency. The experimental results on the MSTAR data set demonstrate that our Dense CapsNet outperforms other state-of-the-art methods upon limited data. [ABSTRACT FROM AUTHOR]

The application of interferometric synthetic aperture radar (InSAR) has been challenged over hilly and vegetated regions due to significant phase decorrelation. The spatiotemporal filtering techniques can partially overcome this, whereas the so-called 'homogeneous pixels' are selected based on the synthetic aperture radar (SAR) datasets and the computation is not satisfactorily efficient. In this letter, we propose the idea of interferogram homogeneous filtering relying on Light Detection and Ranging (LiDAR) derived topography features, instead of using amplitude or phase from the SAR images. The method exploits terrain slope and aspect to cluster topographically homogeneous pixels (THPs) and uses a coherence weighted phase-link (WPL) algorithm to reconstruct pixel phase histories. We compare the proposed method with two representative strategies, i.e., the amplitude statistically homogeneous pixels (SHPs) and the similar time-series interferometric phase pixels (STIPs). Results are given using L-band ALOS-1 data covering the forested Island of Lantau, Hong Kong. It is demonstrated that the LiDAR-derived THP has substantially decreased the computation load whereas the coherence of phase and the InSAR parameter estimation have been improved. [ABSTRACT FROM AUTHOR]

A feature space projection based on the nearest polarimetric and spatial neighbours is proposed for polarimetric synthetic aperture radar (PolSAR) images. The objective function is designed such that increases the class discrimination while preserves the contextual properties. Each polarimetric or spatial neighbour is contributed proportional to with its Wishart distance to the given pixel. A guidance image, which involves all polarimetric channels, is also introduced for spatial filtering of the initial classification map, which is efficient for improving the final classification map. The proposed method shows high performance for PolSAR image classification using limited training samples. [ABSTRACT FROM AUTHOR]

Polarimetric synthetic aperture radar (PolSAR) image classification is a significant task of PolSAR applications. In recent years, deep learning-based methods using spatial features have achieved satisfactory results in PolSAR classification. However, there are still some important features left to be exploited. One is the target scattering orientation diversity feature that contains rich hidden information. In this letter, a hybrid network is proposed to model the target scattering orientation diversity feature and spatial feature. First, the polarimetric scoherency matrix is expanded to polarization coherency matrix sequence by different polarization orientation angles (POAs). Then, the long short-term memory (LSTM) network and the convolutional neural networks (CNN) are proposed to process the polarization coherency matrix sequence and the square neighbourhood of pixels, respectively. Finally, the features of the two sub-networks are combined to improve the classification accuracy. The experiments demonstrate that the proposed method can obtain superior and robust performance. [ABSTRACT FROM AUTHOR]

This study analyses the roles of all the classes in the segmentation procedure, and aims to improve the segmentation accuracy of synthetic aperture radar (SAR) images. Accordingly, weight variables of all classes for SAR images are defined and incorporated into the segmentation model by Bayes' rule and geometric partition tessellation. In addition, the Generalized Multiple-Try Reversible Jump (GMTRJ) and expectation maximization (EM) algorithms are designed to obtain the values of weight variables and the optimal segmentation by simulating the segmentation model. The greater the value of the weight variable, the more important the role of its class. The proposed approach can automatically determine the classes' roles in the segmentation procedure and segment SAR images accurately and quickly. Finally, the proposed and comparison approaches are tested on SAR images, and the results fully demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Recently, Convolutional Neural Network (CNN) has achieved some success in synthetic aperture radar (SAR) image classification. This outstanding performance mainly depends on a large number of training samples, but achieved satisfactory classification results with limited training samples remains a challenge. To address this problem, we propose a dual-scale Siamese densely connected network with Markov Random Fields (DS-SDCNet-MRF) for single-polarization SAR image classification. First, the Siamese densely connected network (SDCNet) is proposed to fully extract discriminative features under limited samples. Then, the proposed DS-SDCNet is constructed with two SDCNets of different scales to produce complementary classification results. Among them, the large-scale SDCNet has a better classification result in the homogeneous region, while the small-scale SDCNet tends to provide good detail preservation. Finally, an improved MRF model which combined the category probability information is proposed to further improve the classification performance. Experimental results on simulated and real single-polarization SAR data demonstrate that the proposed method achieves more encouraging classification performance than the current state-of-the-art classification methods with limited samples. [ABSTRACT FROM AUTHOR]