Search Results (846)

Icebergs pose significant risks to shipping, offshore oil exploration, and underwater pipelines. Detecting and monitoring icebergs in the North Atlantic Ocean, where darkness and cloud cover are frequent, is particularly challenging. Synthetic aperture radar (SAR) serves as a powerful tool to overcome these difficulties. In this paper, we propose a method for automatically detecting and classifying icebergs in various sea conditions using C-band dual-polarimetric images from the RADARSAT Constellation Mission (RCM) collected throughout 2022 and 2023 across different seasons from the east coast of Canada. This method classifies SAR imagery into four distinct classes: open water (OW), which represents areas of water free of icebergs; open water with target (OWT), where icebergs are present within open water; sea ice (SI), consisting of ice-covered regions without any icebergs; and sea ice with target (SIT), where icebergs are embedded within sea ice. Our approach integrates statistical features capturing subtle patterns in RCM imagery with high-dimensional features extracted using a pre-trained Vision Transformer (ViT), further augmented by climate parameters. These features are classified using XGBoost to achieve precise differentiation between these classes. The proposed method achieves a low false positive rate of 1% for each class and a missed detection rate ranging from 0.02% for OWT to 0.04% for SI and SIT, along with an overall accuracy of 96.5% and an area under curve (AUC) value close to 1. Additionally, when the classes were merged for target detection (combining SI with OW and SIT with OWT), the model demonstrated an even higher accuracy of 98.9%. These results highlight the robustness and reliability of our method for large-scale iceberg detection along the east coast of Canada. Full article

Aboveground biomass (AGB) serves as an important indicator for assessing the productivity of forest ecosystems and exploring the global carbon cycle. However, accurate estimation of forest AGB remains a significant challenge, especially when integrating multi-source remote sensing data, and the effects of different feature combinations for AGB estimation results are unclear. In this study, we proposed a method for estimating forest AGB by combining Gao Fen 7 (GF-7) stereo imagery with data from Sentinel-1 (S1), Sentinel-2 (S2), and the Advanced Land Observing Satellite digital elevation model (ALOS DEM), and field survey data. The continuous tree height (TH) feature was derived using GF-7 stereo imagery and the ALOS DEM. Spectral features were extracted from S1 and S2, and topographic features were extracted from the ALOS DEM. Using these features, 15 feature combinations were constructed. The recursive feature elimination (RFE) method was used to optimize each feature combination, which was then input into the extreme gradient boosting (XGBoost) model for AGB estimation. Different combinations of features used to estimate forest AGB were compared. The best model was selected for mapping AGB distribution at 30 m resolution. The outcomes showed that the forest AGB model was composed of 13 features, including TH, topographic, and spectral features extracted from S1 and S2 data. This model achieved the best prediction performance, with a determination coefficient (R²) of 0.71 and a root mean square error (RMSE) of 18.11 Mg/ha. TH was found to be the most important predictive feature, followed by S2 optical features, topographic features, and S1 radar features. Full article

This study investigates the use of the radar vegetation index (RVI) derived from Sentinel-1 synthetic aperture radar (SAR) data for landslide detection. Traditional landslide detection methods often rely on the Normalized Difference Vegetation Index (NDVI) derived from optical imagery, which is susceptible to limitations imposed by weather conditions (clouds, rain) and nighttime. In contrast, SAR data, acquired by Sentinel-1, provides all-weather, day-and-night coverage. To leverage this advantage, we propose a novel approach utilizing RVI, a vegetation index calculated from SAR data, to identify non-vegetated areas, which often indicate potential landslide zones. To enhance the accuracy of non-vegetated area classification, we employ the high-performing EfficientNetV2 deep learning model. We evaluated the classification performance of EfficientNetV2 using RVI derived from Sentinel-1 SAR data with VV and VH polarizations. Experiments were conducted on SAR imagery of the Iburi district in Hokkaido, Japan, severely impacted by an earthquake in 2018. Our findings demonstrate that the classification performance using RVI with both VV and VH polarizations significantly surpasses that of using VV and VH polarizations alone. These results highlight the effectiveness of RVI for identifying non-vegetated areas, particularly in landslide detection scenarios. The proposed RVI-based method has broader applications beyond landslide detection, including other disaster area assessments, agricultural field monitoring, and forest inventory. Full article

Inverse Synthetic Aperture Radar (ISAR) images of space targets and their key components are very important. However, this method suffers from numerous drawbacks, including a low Signal-to-Noise Ratio (SNR), blurred edges, significant variations in scattering intensity, and limited data availability, all of which constrain its recognition capabilities. The terahertz (THz) regime has reflected excellent capacity for space detection in terms of showing the details of target structures. However, in ISAR images, as the observation aperture moves, the imaging features of the extended structures (ESs) undergo significant changes, posing challenges to the subsequent recognition performance. In this paper, a parabolic antenna is taken as the research object. An innovative approach for identifying this component is proposed by using the advantages of the Component Prior and Imaging Characteristics (CPICs) effectively. In order to tackle the challenges associated with component identification in satellite ISAR imagery, this study employs the Improved-YOLOv8 model, which was developed by incorporating the YOLOv8 algorithm, an adaptive detection head known as the Dynamic head (Dyhead) that utilizes an attention mechanism, and a regression box loss function called Wise Intersection over Union (WIoU), which addresses the issue of varying sample difficulty. After being trained on the simulated dataset, the model demonstrated a considerable enhancement in detection accuracy over the five base models, reaching an mAP50 of 0.935 and an mAP50-95 of 0.520. Compared with YOLOv8n, it improved by 0.192 and 0.076 in mAP50 and mAP50-95, respectively. Ultimately, the effectiveness of the suggested method is confirmed through the execution of comprehensive simulations and anechoic chamber tests. Full article

This paper introduces a high-resolution 77–81 GHz mmWave Synthetic Aperture Radar (SAR) imaging methodology integrating low-cost hardware with modified radar signal characteristics specifically for NDT applications. The system is optimized to detect minimal defects in materials, including low-reflectivity ones. In contrast to the existing studies, by optimizing key system parameters, including frequency slope, sampling interval, and scanning aperture, high-resolution SAR images are achieved with reduced computational complexity and storage requirements. The experiments demonstrate the effectiveness of the system in detecting optically undetectable minimal surface defects down to 0.4 mm, such as bonded adhesive lines on low-reflectivity materials with 2500 measurement points and sub-millimeter features on metallic targets at a distance of 30 cm. The results show that the proposed system achieves comparable or superior image quality to existing high-cost setups while requiring fewer data points and simpler signal processing. Low-cost, low-complexity, and easy-to-build mmWave SAR imaging is constructed for high-resolution SAR imagery of targets with a focus on detecting defects in low-reflectivity materials. This approach has significant potential for practical NDT applications with a unique emphasis on scalability, cost-effectiveness, and enhanced performance on low-reflectivity materials for industries such as manufacturing, civil engineering, and 3D printing. Full article

The long-term overextraction of groundwater in the Beijing–Tianjin–Hebei region has led to the formation of the world’s largest groundwater depression cone and the most extensive land subsidence zone, posing a potential threat to the operational safety of high-speed railways in the region. As a critical transportation hub connecting Beijing and the Xiong’an New Area, the Beijing–Xiong’an Intercity Railway traverses geologically complex areas with significant ground subsidence issues. Monitoring and analyzing the causes of land subsidence along the railway are essential for ensuring its safe operation. Using Sentinel-1A radar imagery, this study applies PS-InSAR technology to extract the spatiotemporal evolution characteristics of ground subsidence along the railway from 2016 to 2022. By employing a buffer zone analysis and profile analysis, the subsidence patterns at different stages (pre-construction, construction, and operation) are revealed, identifying the major subsidence cones along the Yongding River, Yongqing, Daying, and Shengfang regions, and their impacts on the railway. Furthermore, the XGBoost model and SHAP method are used to quantify the primary influencing factors of land subsidence. The results show that changes in confined water levels are the most significant factor, contributing 34.5%, with strong interactions observed between the compressible layer thickness and confined water levels. The subsidence gradient analysis indicates that the overall subsidence gradient along the Beijing–Xiong’an Intercity Railway currently meets safety standards. This study provides scientific evidence for risk prevention and the control of land subsidence along the railway and holds significant implications for ensuring the safety of high-speed rail operations. Full article

Deep-learning-based ship detection methods in Synthetic Aperture Radar (SAR) imagery are a current research hotspot. However, these methods rely on high-quality images as input, and in practical applications, SAR images are interfered with by speckle noise, leading to a decrease in image quality and thus affecting detection accuracy. To address this problem, we propose a unified framework for ship detection that incorporates a despeckling module into the object detection network. This integration is designed to enhance the detection performance, even with low-quality SAR images that are affected by speckle noise. Secondly, we propose a Multi-Scale Window Swin Transformer module. This module is adept at improving image quality by effectively capturing both global and local features of the SAR images. Additionally, recognizing the challenges associated with the scarcity of labeled data in practical scenarios, we employ an unlabeled distillation learning method to train our despeckling module. This technique avoids the need for extensive manual labeling and making efficient use of unlabeled data. We have tested the robustness of our method using public SAR datasets, including SSDD and HRSID, as well as a newly constructed dataset, the RSSDD. The results demonstrate that our method not only achieves a state-of-the-art performance but also excels in conditions with low signal-to-noise ratios. Full article

The deep learning object detection algorithm has been widely applied in the field of synthetic aperture radar (SAR). By utilizing deep convolutional neural networks (CNNs) and other techniques, these algorithms can effectively identify and locate targets in SAR images, thereby improving the accuracy and efficiency of detection. In recent years, achieving real-time monitoring of regions has become a pressing need, leading to the direct completion of real-time SAR image target detection on airborne or satellite-borne real-time processing platforms. However, current GPU-based real-time processing platforms struggle to meet the power consumption requirements of airborne or satellite applications. To address this issue, a low-power, low-latency deep learning SAR object detection algorithm accelerator was designed in this study to enable real-time target detection on airborne and satellite SAR platforms. This accelerator proposes a Process Engine (PE) suitable for multidimensional convolution parallel computing, making full use of Field-Programmable Gate Array (FPGA) computing resources to reduce convolution computing time. Furthermore, a unique memory arrangement design based on this PE aims to enhance memory read/write efficiency while applying dataflow patterns suitable for FPGA computing to the accelerator to reduce computation latency. Our experimental results demonstrate that deploying the SAR object detection algorithm based on Yolov5s on this accelerator design, mounted on a Virtex 7 690t chip, consumes only 7 watts of dynamic power, achieving the capability to detect 52.19 512 × 512-sized SAR images per second. Full article

The Aburrá Valley, located in the northwestern region of Colombia, has undergone significant land cover changes and urban expansion in recent decades, driven by rapid population growth and infrastructure development. This region, known for its steep topography and dense urbanization, faces considerable environmental challenges. Monitoring these transformations is essential for informed territorial planning and sustainable development. This study leverages Synthetic Aperture Radar (SAR) imagery from the Sentinel-1 mission, covering 2017–2024, to propose a methodology for the multitemporal analysis of land cover dynamics and urban expansion in the valley. The novel proposed methodology comprises several steps: first, monthly SAR images were acquired for every year under study from 2017 to 2024, ensuring the capture of surface changes. These images were properly calibrated, rescaled, and co-registered. Then, various multitemporal fusions using statistics operations were proposed to detect and find different phenomena related to land cover and urban expansion. The methodology also involved statistical fusion techniques—median, mean, and standard deviation—to capture urbanization dynamics. The kurtosis calculations highlighted areas where infrequent but significant changes occurred, such as large-scale construction projects or sudden shifts in land use, providing a statistical measure of surface variability throughout the study period. An advanced clustering technique segmented images into distinctive classes, utilizing fuzzy logic and a kernel-based method, enhancing the analysis of changes. Additionally, Pearson correlation coefficients were calculated to explore the relationships between identified land cover change classes and their spatial distribution across nine distinct geographic zones in the Aburrá Valley. The results highlight a marked increase in urbanization, particularly along the valley’s periphery, where previously vegetated areas have been replaced by built environments. Additionally, the visual inspection analysis revealed areas of high variability near river courses and industrial zones, indicating ongoing infrastructure and construction projects. These findings emphasize the rapid and often unplanned nature of urban growth in the region, posing challenges to both natural resource management and environmental conservation efforts. The study underscores the need for the continuous monitoring of land cover changes using advanced remote sensing techniques like SAR, which can overcome the limitations posed by cloud cover and rugged terrain. The conclusions drawn suggest that SAR-based multitemporal analysis is a robust tool for detecting and understanding urbanization’s spatial and temporal dynamics in regions like the Aburrá Valley, providing vital data for policymakers and planners to promote sustainable urban development and mitigate environmental degradation. Full article

This study addresses the issue of atmospheric delay correction for the rational polynomial coefficient (RPC) model associated with spaceborne synthetic aperture radar (SAR) imagery under conditions lacking ephemeris data, proposing a novel approach to enhance the geometric positioning accuracy of RPC models. A satellite position inversion method based on the vector-autonomous intersection technique was developed, incorporating ionospheric delay and neutral atmospheric delay models to derive atmospheric delay errors. Additionally, an RPC model reconstruction approach, which integrates atmospheric correction, is proposed. Validation experiments using GF-3 satellite imagery demonstrated that the atmospheric delay values obtained by this method differed by only 0.0001 m from those derived using the traditional ephemeris-based approach, a negligible difference. The method also exhibited high robustness in long-strip imagery. The reconstructed RPC parameters improved image-space accuracy by 18–44% and object-space accuracy by 19–32%. The results indicate that this approach can fully replace traditional ephemeris-based methods for atmospheric delay extraction under ephemeris-free conditions, significantly enhancing the geometric positioning accuracy of SAR imagery RPC models, with substantial application value and development potential. Full article

Cropland is fundamental to food security, and monitoring cropland non-agriculturalization through satellite enforcement can effectively manage and protect cropland. However, existing research primarily focuses on optical imagery, and there are problems such as low data processing efficiency and long updating cycles, which make it difficult to meet the needs of large-scale rapid monitoring. To comprehensively and accurately obtain cropland change information, this paper proposes a method based on the Google Earth Engine (GEE) cloud platform, combining optical imagery and synthetic aperture radar (SAR) data for quick and accurate detection of cropland non-agriculturalization. The method uses existing land-use/land cover (LULC) products to quickly update cropland mapping, employs change vector analysis (CVA) for detecting non-agricultural changes in cropland, and introduces vegetation indices to remove pseudo-changes. Using Shanwei City, Guangdong Province, as a case study, the results show that (1) the cropland map generated in this study aligns well with the actual distribution of cropland, achieving an accuracy of 90.8%; (2) compared to using optical imagery alone, the combined optical and SAR data improves monitoring accuracy by 22.7%, with an overall accuracy of 73.65%; (3) in the past five years, cropland changes in Shanwei followed a pattern of an initial increase followed by a decrease. The research in this paper can provide technical reference for the rapid monitoring of cropland non-agriculturalization on a large scale, so as to promote cropland protection and rational utilization of cropland. Full article

The concept of synthetic aperture radar (SAR) has the advantage of being able to obtain high-quality images even when the target area is at night or covered with obstacles such as clouds or fog. These imaging capabilities have led to a rapid increase in demand for space SAR imagery across a variety of sectors, including government, military, and commercial sectors. The SAR-based deployable reflector antenna was developed in this series of paper. The satellite performance is influenced by the aperture size of an antenna. To improve the image acquisition performance, the SAR antenna has the configuration of several foldable CFRP reflectors. In this paper, the experimental investigation of the Structural-thermal model deployable reflector antenna is performed. During the launch condition, the satellite and payload are subjected to the dynamic load. In the STM phase, the acoustic test was conducted to evaluate the structural stability of the deployable reflector antenna within the acoustic environment. The sinusoidal vibration test was implemented to investigate the fundamental frequency for inplane/normal directions and evaluate the structural stability of reflector antenna. By using experimental data obtained from the thermal-balance test, the well-correlated thermal analysis model was established to execute the orbital thermal analysis. The experimental results of the environmental test in STM phase show that the deployable reflector antenna has structural stability for the structural/thermal environments. The configuration of the deployable reflector antenna determined in STM phase can be applied to the qualification model. Full article

Timely access to soil moisture distribution is critical for agricultural production. As an in-orbit L-band synthetic aperture radar (SAR), SAOCOM offers high penetration and full polarization, making it suitable for agricultural soil moisture estimation. In this study, based on the single-temporal coupled water cloud model (WCM) and Oh model, we first modified the WCM (MWCM) to incorporate bare soil effects on backscattering using SAR data, enhancing the scattering representation during crop growth. Additionally, the Oh model was revised to enable retrieval of both the surface layer (0–5 cm) and underlying layer (5–10 cm) soil moisture. SAOCOM data from 19 June 2022, and 23 June 2023 in Bei’an City, China, along with Sentinel-2 imagery from the same dates, were used to validate the coupled MWCM-Oh model individually. The enhanced vegetation index (EVI), normalized difference vegetation index (NDVI), and leaf area index (LAI), together with the radar vegetation index (RVI) served as vegetation descriptions. Results showed that surface soil moisture estimates were more accurate than those for the underlying layer. LAI performed best for surface moisture (RMSE = 0.045), closely followed by RVI (RMSE = 0.053). For underlying layer soil moisture, RVI provided the most accurate retrieval (RMSE = 0.038), while LAI, EVI, and NDVI tended to overestimate. Overall, LAI and RVI effectively capture surface soil moisture, and RVI is particularly suitable for underlying layers, enabling more comprehensive monitoring. Full article

The exchange of water, sediment, and nutrients in wetlands occurs through a complex network of channels and overbank flow. Although optical sensors can map channels at high resolution, they fail to identify narrow intermittent channels colonized by vegetation. Here we demonstrate an innovative application of rapid-repeat interferometric synthetic aperture radar (InSAR) to study hydrologic connectivity and tidal influences in Louisiana’s coastal wetlands, which can provide valuable insights into water flow dynamics, particularly in vegetation-covered and narrow channels where traditional optical methods struggle. Data used were from the airborne UAVSAR L-band sensor acquired for the Delta-X mission. We applied interferometric techniques to rapid-repeat (~30 min) SAR imagery of the southern Atchafalaya basin acquired during two flights encompassing rising-to-high tides and ebbing-to-low tides. InSAR coherence is used to identify and differentiate permanent open water channels from intermittent channels in which flow occurs underneath the vegetation canopy. The channel networks at rising and ebbing tides show significant differences in the extent of flow, with vegetation-filled small channels more clearly identified at rising-to-high tide. The InSAR phase change is used to identify locations on channel banks where overbank flow occurs, which is a critical component for modeling wetland hydrodynamics. This is the first study to use rapid-repeat InSAR to monitor tidal impacts on water flow dynamics in wetlands. The results show that the InSAR method outperforms traditional optical remote sensing methods in monitoring water flow in vegetation-covered wetlands, providing high-resolution data to support hydrodynamic models and critical support for wetland protection and management. Full article

Synthetic Aperture Radar (SAR) is a crucial remote sensing technology with significant advantages. Ship detection in SAR imagery has garnered significant attention. However, existing ship detection methods often overlook feature extraction, and the unique imaging mechanisms of SAR images hinder the direct application of conventional natural image feature extraction techniques. Moreover, oriented bounding box-based detection methods often prioritize accuracy excessively, leading to increased parameters and computational costs, which in turn elevate computational load and model complexity. To address these issues, we propose a novel two-stage detector, Burgs-rooted vertex offset encoding scheme (BurgsVO), for detecting rotated ships in SAR images. BurgsVO consists of two key modules: the Burgs equation heuristics module, which facilitates feature extraction, and the average diagonal vertex offset (ADVO) encoding scheme, which significantly reduces computational costs. Specifically, the Burgs equation module integrates temporal information with spatial data for effective feature aggregation, establishing a strong foundation for subsequent object detection. The ADVO encoding scheme reduces parameters through anchor transformation, leveraging geometric similarities between quadrilaterals and triangles to further reduce computational costs. Experimental results on the RSSDD and RSDD benchmarks demonstrate that the proposed BurgsVO outperforms the state-of-the-art detectors in both accuracy and efficiency. Full article