Synthetic Aperture Radar Images Research Articles

LRMSNet: A New Lightweight Detection Algorithm for Multi-Scale SAR Objects

In recent years, deep learning has found widespread application in SAR image object detection. However, when detecting multi-scale targets against complex backgrounds, these models often struggle to strike a balance between accuracy and speed. Furthermore, there is a continuous need to enhance the performance of current models. Hence, this paper proposes LRMSNet, a new multi-scale target detection model designed specifically for SAR images in complex backgrounds. Firstly, the paper introduces an attention module designed to enhance contextual information aggregation and capture global features, which is integrated into a backbone network with an expanded receptive field for improving SAR image feature extraction. Secondly, this paper develops an information aggregation module to effectively fuse different feature layers of the backbone network. Lastly, to better integrate feature information at various levels, this paper designs a multi-scale aggregation network. We validate the effectiveness of our method on three different SAR object detection datasets (MSAR-1.0, SSDD, and HRSID). Experimental results demonstrate that LRMSNet achieves outstanding performance with a mean average accuracy (mAP) of 95.2%, 98.9%, and 93.3% on the MSAR-1.0, SSDD, and HRSID datasets, respectively, with only 3.46 M parameters and 12.6 G floating-point operation cost (FLOPs). When compared with existing SAR object detection models on the MSAR-1.0 dataset, LRMSNet achieves state-of-the-art (SOTA) performance, showcasing its superiority in addressing SAR detection challenges in large-scale complex environments and across various object scales.

Satellite Geodesy Unveils a Decade of Summit Subsidence at Ol Doinyo Lengai Volcano, Tanzania

AbstractThe processing of hundreds of Synthetic Aperture Radar (SAR) images acquired by two satellite systems: Sentinel‐1 and COSMO‐SkyMed reveals a decade of ground deformation for a ∼0.5 km diameter area around the summit crater of the only active carbonatitic volcano on Earth: Ol Doinyo Lengai in Tanzania. Further decomposing ascending and descending orbits when the appropriate SAR data sets overlap allow us to interpret the imaged deformation as ground subsidence with a significant rate of ∼3.6 cm/yr for the pixels located just north of the summit crater. Using geodetic modeling and inverting the highest spatial resolution COSMO‐SkyMed data set, we show that the mechanism explaining this subsidence is most likely a deflating very shallow (≤1 km depth below the summit crater at the 95% confidence level) magma reservoir, consistent with geochemical‐petrological and seismo‐acoustic studies.

Full-Scale Aggregated MobileUNet: An Improved U-Net Architecture for SAR Oil Spill Detection.

Oil spills are a major threat to marine and coastal environments. Their unique radar backscatter intensity can be captured by synthetic aperture radar (SAR), resulting in dark regions in the images. However, many marine phenomena can lead to erroneous detections of oil spills. In addition, SAR images of the ocean include multiple targets, such as sea surface, land, ships, and oil spills and their look-alikes. The training of a multi-category classifier will encounter significant challenges due to the inherent class imbalance. Addressing this issue requires extracting target features more effectively. In this study, a lightweight U-Net-based model, Full-Scale Aggregated MobileUNet (FA-MobileUNet), was proposed to improve the detection performance for oil spills using SAR images. First, a lightweight MobileNetv3 model was used as the backbone of the U-Net encoder for feature extraction. Next, atrous spatial pyramid pooling (ASPP) and a convolutional block attention module (CBAM) were used to improve the capacity of the network to extract multi-scale features and to increase the speed of module calculation. Finally, full-scale features from the encoder were aggregated to enhance the network's competence in extracting features. The proposed modified network enhanced the extraction and integration of features at different scales to improve the accuracy of detecting diverse marine targets. The experimental results showed that the mean intersection over union (mIoU) of the proposed model reached more than 80% for the detection of five types of marine targets including sea surface, land, ships, and oil spills and their look-alikes. In addition, the IoU of the proposed model reached 75.85 and 72.67% for oil spill and look-alike detection, which was 18.94% and 25.55% higher than that of the original U-Net model, respectively. Compared with other segmentation models, the proposed network can more accurately classify the black regions in SAR images into oil spills and their look-alikes. Furthermore, the detection performance and computational efficiency of the proposed model were also validated against other semantic segmentation models.

The N-S direction strike-slip activities in the Pamir hinterland under oblique convergence: the 2015 and 2023 earthquakes

SUMMARY The prominent Pamir plateau holds considerable significance in comprehending the processes of Asian continental collisional orogeny. However, due to harsh natural conditions and low seismic activity within the Pamir hinterland, our understanding of this region remains deficient. Recent major events and the accumulation of geodetic observations present a rare opportunity for us to get insights into the tectonic activities and orogenic processes occurring in this region. First, employing Sentinel-1 and Advanced Land Observation Satellite (ALOS)-2 Synthetic Aperture Radar (SAR) images, we acquire coseismic displacements associated with the most recent earthquakes in 2015 and 2023. Subsequently, we conduct the source models inversion with the constraints of surface displacements based on a finite-fault model. Our results reveal displacements ranging from −0.8 to 0.8 m for the 2015 Mw 7.2 Tajik earthquake and −0.25 to 0.25 m for the 2023 Mw 6.9 Murghob event, respectively. The optimal three-segment model for the 2015 event ruptured a fault length of 89 km with a surface rupture extending 59 km along the Sarez–Karakul fault (SKF), characterized predominantly by left-lateral strike-slip motion, with a maximum slip of 3.5 m. Meanwhile, our preferred uniform slip model suggests that the 2023 event ruptured an unmapped fault in the southern Pamir region with a strike angle of 31° and a dip angle of 76.8°. The distributed slip model indicates that the 2023 event ruptured a fault length of 32 km, resulting in an 8 km surface rupture. This event is characterized by left-lateral strike slip, with a peak slip of 2.2 m. Secondly, the Coulomb stress calculations demonstrate that the 2023 event was impeded by the 2015 event. Finally, interseismic Global Positioning System data revel a relative motion of 3.4–5.7 mm yr−1 in the N-S component and 3.2–3.8 mm yr−1 in the E-W component along the SKF in the Pamir hinterland, respectively. These N-S direction strike-slip activities and slip behaviours support an ongoing strong shear and extension in the Pamir regime, which is a response to the oblique convergence between the Indian and Eurasian plates.

A Deep-Based Approach for Multi-Descriptor Feature Extraction: Applications on SAR Image Registration

Synthetic aperture radar (SAR) images have found increasing attention in various applications, such as remote sensing, surveillance, and disaster management. Unlike optical sensors, SAR can capture images in any weather conditions at any time, making it particularly valuable. However, it poses unique challenges for image processing tasks, like complex speckle noise, texture, and geometric distortions. For many image processing tasks such as image registration; feature detection and extraction are critical initial steps. In this work, we propose a deep learning-based method for detecting and describing keypoints in SAR images, which can be specifically applied to SAR image registration tasks. To address the challenges of SAR images, we utilize the R2D2 network and propose a training scheme specifically tailored for these images. Furthermore, our approach involves multiple descriptors utilization for outlier removal between image keypoints and a two-stage pipeline consisting of coarse and fine stages for robust feature point matching of SAR images. By leveraging the coarse registration stage, which maintains robustness, and also a fine stage that employs multiple descriptors along with a local search technique, our approach enhances the keypoint matching accuracy. Evaluation results show that the proposed method outperforms most of the traditional and deep learning-based methods in terms of the percentage of correct matches, while achieving comparable or even better root mean squared error (RMSE) results.

GDB‐YOLOv5s: Improved YOLO‐based model for ship detection in SAR images

AbstractIn recent years, deep learning methods were good solutions for object detection in synthetic aperture radar (SAR) images. However, the problems of complex scenarios, large object scale differences and imperfect fine‐grained classification in ship detection were intractable. In response, an improved model GDB‐YOLOv5s (Improved YOLOv5s model incorporating global attention mechanism (GAM), DCN‐v2 and BiFusion) is designed. This model introduces deformable convolution networks (DCN‐v2) into the Backbone to enhance the adaptability of the receptive field. It replaces the original Neck's PANet structure with a BiFusion structure to better fuse the extracted multiscale features. Additionally, it integrates GAM into the network to reduce information loss and improve global feature interaction. Experiments were conducted on single‐class dataset SSDD and multi‐class dataset SRSSD‐V1.0. The results show that the GDB‐YOLOv5s model improves mean average precision scores (mAP) significantly, outperforming the original YOLOv5s model and other traditional methods. GDB‐YOLOv5s also improves Precision‐score (P) and Recall‐score (R) for fine‐grained classification to some extent, thereby reducing false alarms and missed detections. It has been proved that the improved model is relatively effective.

Semi-Supervised SAR Image Classification via Adaptive Threshold Selection

Semi-supervised learning is a good way to train a classification model using a small number of labeled and large number of unlabeled data. We applied semi-supervised learning to a synthetic aperture radar(SAR) image classification model with a limited number of datasets that are difficult to create. To address the previous difficulties, semi-supervised learning uses a model trained with a small amount of labeled data to generate and learn pseudo labels. Besides, a lot of number of papers use a single fixed threshold to create pseudo labels. In this paper, we present a semi-supervised synthetic aperture radar(SAR) image classification method that applies different thresholds for each class instead of all classes sharing a fixed threshold to improve SAR classification performance with a small number of labeled datasets.

Multiple Signal Classification Algorithm Combined with Volume Reflectivity Models to Improve Accuracy of the Estimated Vegetation Height in Synthetic Aperture Radar Tomography

Multiple Signal Classification Algorithm Combined with Volume Reflectivity Models to Improve Accuracy of the Estimated Vegetation Height in Synthetic Aperture Radar Tomography

NRENet: Neighborhood removal-and-emphasis network for ship detection in SAR Images

In recent years, object detection in Synthetic-Aperture Radar (SAR) images still faces many challenges, especially for ship detection. Small or dense ships are vulnerable to the interference of complex scenes such as ports and land. In feature extraction, a large amount of redundant information on the feature map will further reduce the network’s attention to small-sized ships. Therefore, in this paper, we propose a network called neighborhood removal-and-emphasis network (NRE-Net), including an object neighborhood removal (ONR) strategy and a neighborhood feature emphasis (NFE) module. Among them, the ONR strategy directly removes the complex background information around the ship before feature extraction, only retaining effective contextual neighborhoods conducive to ship detection and avoiding the interference of complex background information on the network. The NFE module is based on ONR to extract features and form a weight map of small-sized ships or complex images. This module can adaptively recognize the detection neighborhood of each ship and highlight the detection ship on the feature map. Our network has validated the effectiveness of the method on multiple SAR ship datasets and improved the object AP for each size. Our code is available at: https://github.com/Xidian-AIGroup190726/NRENet.

Phase error compensation for Synthetic Aperture Radar (SAR) images by on non-iteration Phase Gradient Autofocus (PGA) algorithms

Phase error compensation for Synthetic Aperture Radar (SAR) images by on non-iteration Phase Gradient Autofocus (PGA) algorithms

Design of an integrated imaging system of airborne SAR and visible light camera based on common aperture antenna

In recent years, the advantages of multi-sensor integration are gradually highlighted. In this paper, an integrated imaging system (IIS) of airborne synthetic aperture radar (SAR) and visible light camera (VLC) based on common aperture antenna is designed to meet the mission requirement of acquiring visible light and SAR images simultaneously. The IIS adopts the coaxial scheme, its structure is simple and compact, and is more suitable for airborne platform. In this paper, a parameter design method of IIS based on common aperture antenna is proposed. After the system parameters design, the imaging performance of IIS is analyzed by simulation. The results show that the system parameters are designed reasonably and the imaging performance meets the mission requirement. The designed IIS can realize the visible and Ka-band (30 GHz) dual-band imaging. Under the condition of 20 km altitude, the imaging resolutions of VLC and SAR are higher than 0.2 m and 0.3 m respectively, the field of view of VLC is 1.46°, and the swath length of SAR is greater than 870 m.

Development of X-Band Geophysical Model Function for Sea Surface Wind Speed Retrieval with ASNARO-2

In the present study, a new geophysical model function (GMF) is developed for the X-band synthetic aperture radar (SAR) on board the Advanced Satellite with New System Architecture for Observation-2 (ASNARO-2) to retrieve accurate offshore wind speeds. Equivalent neutral wind speeds based on the local forecast model (LFM) are employed as reference wind vectors, and 12,259 matching points from 502 SAR images obtained with horizontal transmitting, horizontal receiving polarization around Japan are collected. To ensure convergence of the calculation, 8129 points are selected from the matching points to determine the basic formula for the GMF and 23 coefficients based on the relationships among the normalized radar cross section, wind speed, incidence angle, and relative wind direction. Compared with the reference wind speeds, the GMF wind speeds showed reproducibility with a bias of −0.10 m/s and an RMSD of 1.37 m/s. Additionally, it can be confirmed that the retrieved wind speed has the bias of 0.03 and the RMSD of 1.68 m/s when compared to the in situ wind speed from the Kuroshio Extension Observatory (KEO) buoy. The accuracy of these retrieved wind speeds is comparable to previous studies, and it is indicated that the developed GMF can be used to retrieve offshore winds from ASNARO-2 images.

Source Models of the 2016 and 2022 Menyuan Earthquakes and Their Tectonic Implications Revealed by InSAR.

The Haiyuan fault system plays a crucial role in accommodating the eastward expansion of the Tibetan Plateau (TP) and is currently slipping at a rate of several centimeters per year. However, limited seismic activities have been observed using geodetic techniques in this area, impeding the comprehensive investigation into regional tectonics. In this study, the geometric structure and source models of the 2022 Mw 6.7 and the 2016 Mw 5.9 Menyuan earthquakes were investigated using Sentinel-1A SAR images. By implementing an atmospheric error correction method, the signal-to-noise ratio of the 2016 interferometric synthetic aperture radar (InSAR) coseismic deformation field was significantly improved, enabling InSAR observations with higher accuracy. The results showed that the reliability of the source models for those events was improved following the reduction in observation errors. The Coulomb stress resulting from the 2016 event may have promoted the strike-slip movement of the western segment of the Lenglongling fault zone, potentially expediting the occurrence of the 2022 earthquake. The coseismic slip distribution and the spatial distribution of aftershocks of the 2022 event suggested that the seismogenic fault may connect the western segment of the Lenglongling fault (LLLF) and the eastern segment of the Tuolaishan fault (TLSF). Additionally, the western segment of the surface rupture zone of the northern branch may terminate in the secondary branch close to the Sunan-Qilian fault (SN-QL) strike direction, and the earthquake may have triggered deep aftershocks and accelerated stress release within the deep seismogenic fault.

SAR target recognition through adaptive kernel sparse representation model based on local contrast perception

Sparse representation-based methods have shown promising prospects in synthetic aperture radar (SAR) target recognition due to their robustness under complex application conditions. However, these methods are still susceptible to the feature representation of SAR images. In this paper, we propose a new SAR target recognition method using an adaptive kernel sparse representation model based on local contrast perception (LCP). The LCP method combines a specially designed random projection matrix with two-dimensional filtering to capture extensive intensity contrasts between local regions in SAR images. A response map reconstruction technique shows that LCP can obtain discriminative properties between target and background regions and improve feature representation. The LCP features are then transformed into the kernel representation to improve linear separability by defining a reproducing kernel Hilbert space (RKHS) and using a nonlinear kernel function. The kernel representation is utilized to construct a new kernel sparse representation model for classification. Furthermore, neighborhood analysis (NA) is proposed to establish the relationship between the parameter and feature distribution, with an adaptive regularization parameter selection strategy to mitigate the parameter sensitivity. Experiments conducted on the moving and stationary target acquisition and recognition dataset demonstrate the superiority and robustness of the proposed method under different application conditions.

On the physical mechanism causing strongly enhanced radar backscatter in C-Band SAR images of convective rain over the ocean

ABSTRACT Radar signatures of rain over the ocean have a complex structure since they receive contributions from surface scattering and volume scattering and attenuation by hydrometeors in the atmosphere. These contributions overlap and are often difficult to detangle. While most of the mechanisms contributing to radar signatures of rain over the ocean are well understood, there is one remaining issue that has been discussed controversially in the literature for a long time. It is the question what scattering mechanism causes the areas of strongly enhanced radar backscatter, also called ‘bright blobs’ or ‘bright patches’, which are frequently observed on spaceborne C-band SAR images acquired over tropical and subtropical oceans in the presence of convective rain. Recently, papers have been published in which it is hypothesized that they are caused by radar backscattering at hydrometeors in the melting layer (ML). Although many observational facts seem to support this hypothesis, there exists one strong argument against this hypothesis: It is the observation that the position of the ML radar signatures (bright blob) in the SAR image is not shifted in anti-range from the position, where the rain column hits the sea surface. This absence of a shift is observed when 1) comparing Sentinel-1 SAR images on which rain cells are visible with quasi-concurrently acquired weather radar images and 2) when inter-comparing of SAR images of rain cells acquired concurrently at different frequencies and polarizations. Based on these observations, we discard the hypothesis that the bright blobs are due to volume scattering at hydrometeors in the ML and hypothesize instead that they are due to scattering at splash products at the sea surface. This hypothesis is supported by radar backscattering measurements carried out in the laboratory and from a shore-based platform, which show that, at C-and X-band, strong rain can give rise to strong radar returns also at cross-polarization.

A noise-robust water segmentation method based on synthetic aperture radar images combined with automatic sample collection

ABSTRACT 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 L NR-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.

ISAR Image Registration Based on Line Features

Inverse synthetic aperture radar (ISAR) image registration enables the analysis of target dynamics by comparing registered images from different viewpoints. However, it faces significant challenges due to various factors, such as the complex scattering characteristics of the target, limited availability of information, and additive noise in ISAR images. This paper proposes a novel ISAR image registration method based on line features. It integrates information from both dominant scatterers and the target’s outer contour to detect lines. According to the consistency principles of multiple lines in rotation and translation, line features from different ISAR images are matched. Simultaneously, the results of the feature matching are utilized to guide the parameter configuration for optimizing the image registration process. Comparative experiments illustrate the advantages of the proposed method in both feature extraction and registration feasibility.

PP-ISEA: An Efficient Algorithm for High-Resolution Three-Dimensional Geometry Reconstruction of Space Targets Using Limited Inverse Synthetic Aperture Radar Images.

As the variety of space targets expands, two-dimensional (2D) ISAR images prove insufficient for target recognition, necessitating the extraction of three-dimensional (3D) information. The 3D geometry reconstruction method utilizing energy accumulation of ISAR image sequence (ISEA) facilitates superior reconstruction while circumventing the laborious steps associated with factorization methods. Nevertheless, ISEA's neglect of valid information necessitates a high quantity of images and elongated operation times. This paper introduces a partitioned parallel 3D reconstruction method utilizing sorted-energy semi-accumulation with ISAR image sequences (PP-ISEA) to address these limitations. The PP-ISEA innovatively incorporates a two-step search pattern-coarse and fine-that enhances search efficiency and conserves computational resources. It introduces a novel objective function 'sorted-energy semi-accumulation' to discern genuine scatterers from spurious ones and establishes a redundant point exclusion module. Experiments on the scatterer model and simulated electromagnetic model demonstrate that the PP-ISEA reduces the minimum image requirement from ten to four for high-quality scatterer model reconstruction, thereby offering superior reconstruction quality in less time.

DNN-PolSAR: Urban Image Segmentation and Classification using Polarimetric SAR based on DNNs

Synthetic Aperture Radar (SAR) image segmentation and classification is a popular technique for learning and detection of objects such as buildings, trees, monuments, crops water-bodies, hills, etc. SAR technique is being used for urban development and city-planning, building control of municipal objects, searching best locations, detection of changes in the existing systems, etc. using polarimetry based on Deep Neural Networks. In this paper, weproposed a technique for Urban Image Segmentation and Classification using Polarimetric SAR based on Deep NeuralNetworks (DNN-PolSAR). In our proposed DNN-PolSAR technique, we useMask-RCNN, LinkNet, FPN, and PSP- Net as model architectures, whereas ResNet50, ResNet101, ResNet152, and VGG-19 are used as backbone networks.We first apply polarimetric decomposition on airborne Uninhabited Aerial Vehicle Synthetic Aperture (UAVSAR) im- ages of urban areas and then the decomposed images are fed to DNNs for segmentation and classification. We then simulate DNN-PolSAR considering different hyper-parameters and compare the obtained scores of hyper-parametersagainst used model architectures and backbone networks. In comparison, it is found that DNN-PolSAR based on FPN model with ResNet152 performed the best for segmentation and classification. The mean Average Precision (mAP) score of the DNN-PolSAR based on FPN with a pixel accuracy of 90.9% is 0.823, which outperforms other Deep Learning models.

Target recognition in diverse synthetic aperture radar image datasets with low size weight and power processing hardware

Target recognition in diverse synthetic aperture radar image datasets with low size weight and power processing hardware