Synthetic Aperture Radar Images Research Articles

Mangrove mapping in China using Gaussian mixture model with a novel mangrove index (SSMI) derived from optical and SAR imagery

As an important shoreline vegetation and highly productive ecosystem, mangroves play an essential role in the protection of coastlines and ecological diversity. Accurate mapping of the spatial distribution of mangroves is crucial for the protection and restoration of mangrove ecosystems. Supervised classification methods rely on large sample sets and complex classifiers and traditional thresholding methods that require empirical thresholds, given the problems that limit the feasibility and stability of existing mangrove identification and mapping methods on large scales. Thus, this paper develops a novel mangrove index (spectral and SAR mangrove index, SSMI) and Gaussian mixture model (GMM) mangrove mapping method, which does not require training samples and can automatically and accurately map mangrove boundaries by utilizing only single-scene Sentinel-1 and single-scene Sentinel-2 images from the same time period. The SSMI capitalizes on the fact that mangroves are differentiated from other land cover types in terms of optical characteristics (greenness and moisture) and backscattering coefficients of SAR images and ultimately highlights mangrove forest information through the product of three expressions (f(S) = red egde/SWIR1, f(B) = 1/(1 + e-VH), f(W)=(NIR-SWIR1)/(NIR+SWIR1)). The proposed SSMI was tested in six typical mangrove distribution areas in China where climatic conditions and mangrove species vary widely. The results indicated that the SSMI was more capable of mapping mangrove forests than the other mangrove indices (CMRI, NDMI, MVI, and MI), with overall accuracys (OA) higher than 0.90 and F1 scores as high as 0.93 for the other five areas except for the Maowei Gulf (S5). Moreover, the mangrove maps generated by the SSMI were highly consistent with the reference maps (HGMF_2020、LASAC_2018 and IMMA). In addition, the SSMI achieves stable performance, as shown by the mapping results of the other two classification methods (K-means and Otsu’s algorithm). Mangrove mapping in six typical mangrove distribution areas in China for five consecutive years (2019–2023) and experiments in three Southeast Asian countries with major mangrove distributions (Thailand, Vietnam, and Indonesia) demonstrated that the SSMIs constructed in this paper are highly stable across time and space. The SSMI proposed in this paper does not require reference samples or predefined parameters; thus, it has great flexibility and applicability in mapping mangroves on a large scale, especially in cloudy areas.

Enhanced Prototypical Network with Customized Region-Aware Convolution for Few-Shot SAR ATR

With the prosperous development and successful application of deep learning technologies in the field of remote sensing, numerous deep-learning-based methods have emerged for synthetic aperture radar (SAR) automatic target recognition (ATR) tasks over the past few years. Generally, most deep-learning-based methods can achieve outstanding recognition performance on the condition that an abundance of labeled samples are available to train the model. However, in real application scenarios, it is difficult and costly to acquire and to annotate abundant SAR images due to the imaging mechanism of SAR, which poses a big challenge to existing SAR ATR methods. Therefore, SAR target recognition in the situation of few-shot, where only a scarce few labeled samples are available, is a fundamental problem that needs to be solved. In this paper, a new method named enhanced prototypical network with customized region-aware convolution (CRCEPN) is put forward to specially tackle the few-shot SAR ATR tasks. To be specific, a feature-extraction network based on a customized and region-aware convolution is first developed. This network can adaptively adjust convolutional kernels and their receptive fields according to each SAR image’s own characteristics as well as the semantical similarity among spatial regions, thus augmenting its capability to extract more informative and discriminative features. To achieve accurate and robust target identity prediction under the few-shot condition, an enhanced prototypical network is proposed. This network can improve the representation ability of the class prototype by properly making use of training and test samples together, thus effectively raising the classification accuracy. Meanwhile, a new hybrid loss is designed to learn a feature space with both inter-class separability and intra-class tightness as much as possible, which can further upgrade the recognition performance of the proposed method. Experiments performed on the moving and stationary target acquisition and recognition (MSTAR) dataset, the OpenSARShip dataset, and the SAMPLE+ dataset demonstrate that the proposed method is competitive with some state-of-the-art methods for few-shot SAR ATR tasks.

Identifying the spatio-temporal distribution characteristics of offshore wind turbines in China from Sentinel-1 imagery using deep learning

ABSTRACT Offshore wind power is a crucial clean energy source for coastal countries and advances blue economies. Accurate spatial mapping of offshore wind turbines supports energy assessment and the sustainable development of marine resources. Offshore wind power has developed rapidly in China, but the spatiotemporal distribution characteristics of offshore wind turbines and farms remain poorly understood. Extracting nearshore small targets from Sentinel-1 synthetic aperture radar images remain challenging due to noise from the rough sea surface background and the diverse substrate types in offshore wind turbine (OWT) distributions. Here, to address these issues, we first created a dataset of China’s offshore wind turbines (COWTs) that contained typical substrate types (e.g. seawater and tidal flats). Then, a deep learning model integrating an attention mechanism and receptive fields was used to accurately detect COWTs. The well-trained model was used to detect the changes in COWTs from 2015 to 2022. Our model detected an increase in the number of offshore wind farms in China from 7 to 114 (clusters counted as an offshore wind farm), while COWTs increased from 305 to 6451 from 2015 to 2022. The Taiwan Strait exhibited the highest wind speeds, yet the Jiangsu and Guangdong regions had the most installed turbines. Over 90% of COWTs were located in areas with offshore wind energy resources less than 500 W/m3, indicating a mismatch between COWT installation and wind resource distribution. This study demonstrated temporal and geographical generalizability, which is promising for global wind power detection. Our analysis of spatiotemporal variation in COWTs facilitates informed decision-making for future field establishment and sustainable marine planning.

Multi-Domain Joint Synthetic Aperture Radar Ship Detection Method Integrating Complex Information with Deep Learning

With the flourishing development of deep learning, synthetic aperture radar (SAR) ship detection based on this method has been widely applied across various domains. However, most deep-learning-based detection methods currently only use the amplitude information from SAR images. In fact, phase information and time-frequency features can also play a role in ship detection. Additionally, the background noise and the small size of ships also pose challenges to detection. Finally, satellite-based detection requires the model to be lightweight and capable of real-time processing. To address these difficulties, we propose a multi-domain joint SAR ship detection method that integrates complex information with deep learning. Based on the imaging mechanism of line-by-line scanning, we can first confirm the presence of ships within echo returns in the eigen-subspace domain, which can reduce detection time. Benefiting from the complex information of single-look complex (SLC) SAR images, we transform the echo returns containing ships into the time-frequency domain. In the time-frequency domain, ships exhibit distinctive features that are different from noise, without the limitation of size, which is highly advantageous for detection. Therefore, we constructed a time-frequency SAR image dataset (TFSID) using the images in the time-frequency domain, and utilizing the advantages of this dataset, we combined space-to-depth convolution (SPDConv) and Inception depthwise convolution (InceptionDWConv) to propose Efficient SPD-InceptionDWConv (ESIDConv). Using this module as the core, we proposed a lightweight SAR ship detector (LSDet) based on YOLOv5n. The detector achieves a detection accuracy of 99.5 with only 0.3 M parameters and 1.2 G operations on the dataset. Extensive experiments on different datasets demonstrated the superiority and effectiveness of our proposed method.

A HRWS SAR Motion Compensation Method with Multichannel Phase Correction

The multichannel synthetic aperture radar (SAR) possesses the capability to acquire high-resolution, wide-swath SAR imagery, which has great potential for application. However, similar to traditional single-channel SAR systems, it suffers from imaging quality degradation due to motion errors. Many motion compensation algorithms have been used to improve the quality of single-channel SAR images, while fewer studies have been conducted on multichannel SAR motion compensation methods. The sub-image motion compensation method utilizes the single channel motion errors to perform multichannel motion errors compensation, considering that multiple channels have the same phase errors. To improve the quality of multichannel SAR imaging when multiple channel motion errors are inconsistent, this paper proposes a motion compensation method with multichannel phase correction for HRWS SAR. First, the method derives the phase errors estimation model via maximum sharpness to simultaneously estimate multichannel phase. Then, it compensates for the motion errors of all channels during backprojection imaging. The inconsistent motion errors of multiple channels can be compensated by estimating the phase errors of all channels, improving the image quality. The channel phase errors can be corrected while compensating for the motion errors. Moreover, the experimental results of point targets and complex scenes validate the effectiveness of the proposed method.

Robust target recognition in synthetic aperture radar images via correlation analysis and joint decision of multi-level bidimensional intrinsic mode functions by bidimensional variational mode decomposition

Robust target recognition in synthetic aperture radar images via correlation analysis and joint decision of multi-level bidimensional intrinsic mode functions by bidimensional variational mode decomposition

A Robust Translational Motion Compensation Method for Moving Target ISAR Imaging Based on Phase Difference-Lv’s Distribution and Auto-Cross-Correlation Algorithm

Translational motion compensation constitutes a pivotal and essential procedure in inverse synthetic aperture radar (ISAR) imaging. Many researchers have previously proposed their methods to address this requirement. However, conventional methods may struggle to produce satisfactory results when dealing with non-stationary moving targets or operating under conditions of low signal-to-noise ratios (SNR). Aiming at this challenge, this article proposes a parametric non-search method that contains two main stages. The radar echoes can be modeled as polynomial phase signals (PPS). In the initial stage, the energy of the received two-dimensional signal is coherently integrated into a peak point by leveraging phase difference (PD) and Lv’s distribution (LVD), from which the high-order polynomial coefficients can be obtained accurately. The estimation of the first-order coefficients is conducted during the second stage. The auto-cross-correlation function for range profiles is introduced to enhance the accuracy and robustness of estimation. Subsequently, a novel mathematical model for velocity estimation is proposed, and its least squares solution is derived. Through this model, a sub-resolution solution can be obtained without requiring interpolation. By employing all the estimated polynomial coefficients, the non-stationary motion of the target can be fully compensated, yielding the acquisition of a finely focused image. Finally, the experimental findings validate the superiority and robustness of the proposed method in comparison to state-of-the-art approaches.

Unveiling spatial pattern in slope units-based landslides using a data-driven GAMM and InSAR techniques

ABSTRACT Spatial slope deformation pattern is crucial for landslide early warning and management. Quantitative modelling of the effects of multiple driving factors on the landslide spatial pattern has not yet been fully explored. To address this, this study utilizes data-driven models and Interferometric Synthetic Aperture Radar (InSAR) techniques to unveil the spatial pattern of landslides based on SUs (slope units). Specifically, we propose a methodology that establishes a correlation between InSAR deformation and topographic/meteorological factors by constructing a Generalized Additive Mixed Model (GAMM). Using this method, we analysed ALOS-2 ascending SAR images from April 2020 to September 2021 and successfully obtained cumulative minimum deformation during the period. Through the GAMM, we linked the observed cumulative minimum deformation in each SU to various factors, including topographic and meteorological factors. In terms of accuracy, there is an excellent agreement between the observed and fitted deformation, with a Pearson Correlation Coefficient (PCC) of 0.83. In terms of interpretability, our method can well explain the contribution of each factor to landslide deformation. We conclude that using the data-driven GAMM and InSAR techniques can effectively reveal the spatial pattern of landslides based on SUs. Furthermore, based on the method proposed in this study, it is possible to predict landslide deformation in the future.

Drone Swarm Classification from ISAR Imaging

In today's technology, the use of drones has become very popular for they can be easily purchased over the Internet and can be easily developed. With drones that have wide usage areas, swarm structures have become popular. However, this has brought about some problems. The issue of drone detection has emerged in order to prevent the uncontrolled use of drone swarms in the airspace. Drone swarm detection is important to prevent dangerous accidents or criminal acts. In this study, a new classification algorithm is proposed with deep learning using inverse synthetic aperture radar (ISAR) images of drone swarms based on various formation swarm types. ISAR images are created using ANSYS simulation. Additionally, high frequency structural simulator (HFSS) - shooting bouncing ray (SBR+) solver is used for high-speed computation. Radar and simulation parameters to obtain ISAR images are discussed. Especially, down-range and cross-range resolution parameters are taken into account to achieve high resolution. ISAR images are classified using deep learning methods in terms of formation. Formation types include Line, Square, Cross, and Triangle. The convolutional neural network (CNN) model is used to solve classification problems. The model consists of train, validation, and test steps. Classification performance results are presented with high accuracy. The developed method can be used for anti-drone technologies.

The Impacts of Phenological Stages within the Annual Cycle on Mapping Forest Stock Volume Using Multi-Band Dual-Polarization SAR Images in Boreal Forests

SAR images with two polarizations show strong potential for mapping forest stock volume (FSV) combined with limited samples. However, accurately mapping FSV still presents challenges in selecting the optimal acquisition date to obtain the SAR images during specific phenological stages within the annual forest cycle (growth and dormant stages). To clarify the impacts of phenological stages within the annual cycle on FSV mapping, SAR images with various polarization models and bands (Sentinel-1(S), GaoFen-3(GF-3 (G)) and ALOS-2(A)) were acquired within the growth and dormant stages of an annual cycle in a boreal evergreen coniferous forest (Chinese pine) and a deciduous coniferous forest (Larch). Subsequently, single-band (G, S, and A) and multi-band combined alternative variable sets (A + G, A + S, S + G, and A + S + G) were extracted within the same stage, respectively. Finally, the forward selection approach was utilized in conjunction with four different models (MLR, KNN, RF, and SVR) to obtain the most suitable variable sets and generate FSV mapping. The results demonstrated a strong correlation between the intensity of backscattering coefficients and the phenological stages of the forest. Within the dormant stage, there was a significant decrease in the gaps of backscattering coefficients obtained from the same polarization compared to those within the growth stage. Furthermore, the results also revealed that more signals from inside the canopy could be detected during the dormant stage in both evergreen coniferous forests and deciduous coniferous forests. Subsequently, the accuracy in mapping FSV obtained from single-band SAR images within the dormant stage are slightly higher than that within the growth stage, and the accuracy was still significantly affected by both overestimation and underestimation. Moreover, the combined effects of different bands significantly improve the reliability of mapped FSV. The rRMSE values in four multi-band combinations ranged from 22.37% to 29.40% for Chinese pine forests and from 21.27% to 34.38% for Larch forests, and the optimal result was observed from combinations of A + S + G acquired within the dormant stage. It is confirmed that SAR signal and their sensitivity to FSV depends on the stages of forest annual growth cycle. In comparison to the growth period, dual-polarization SAR data acquired during the dormant stage is more suitable for estimating FSV in boreal forests.

Ship Wake Detection in a Single SAR Image via a Modified Low-Rank Constraint

Ship Wake Detection in a Single SAR Image via a Modified Low-Rank Constraint

Ship target detection in SAR images based on SimAM attention YOLOv8

AbstractDeep learning has been widely applied in ship detection in synthetic aperture radar (SAR) imagery due to their powerful feature representation capabilities. However, YOLOv8 models treat all regions of the image equally during convolutional feature processing, resulting in less‐than‐ideal outcomes. To address this limitation, this study proposes a simple, parameter‐free attention module (SimAM) attention‐based YOLOv8 algorithm for ship detection in SAR images. The proposed algorithm first passes through a backbone network, which incorporates SimAM attention modules. The SimAM attention mechanism successfully allocates the convolutional neural network's 3D weights effectively using an energy function method, without introducing additional parameters. This mechanism enables the network to automatically emphasize key features in the image, enhancing its ability to represent target areas and suppress background interference. Subsequently, deep features are upsampled and fused with relatively shallow features to extract features at three different scales and achieve target detection, ultimately outputting classification and positional information of the targets. The effectiveness of the model on the SAR‐ship‐dataset is experimentally validated achieving an mAP50 value of 97.72% and an mAP50‐95 value of 68.99%, confirming the superiority of the proposed model.

Forest Canopy Height Estimation Combining Dual-Polarization PolSAR and Spaceborne LiDAR Data

Forest canopy height data are fundamental parameters of forest structure and are critical for understanding terrestrial carbon stock, global carbon cycle dynamics and forest productivity. To address the limitations of retrieving forest canopy height using conventional PolInSAR-based methods, we proposed a method to estimate forest height by combining single-temporal polarimetric synthetic aperture radar (PolSAR) images with sparse spaceborne LiDAR (forest height) measurements. The core idea of our method is that volume scattering energy variations which are linked to forest canopy height occur during radar acquisition. Specifically, our methodology begins by employing a semi-empirical inversion model directly derived from the random volume over ground (RVoG) formulation to establish the relationship between forest canopy height, volume scattering energy and wave extinction. Subsequently, PolSAR decomposition techniques are used to extract canopy volume scattering energy. Additionally, machine learning is employed to generate a spatially continuous extinction coefficient product, utilizing sparse LiDAR samples for assistance. Finally, with the derived inversion model and the resulting model parameters (i.e., volume scattering power and extinction coefficient), forest canopy height can be estimated. The performance of the proposed forest height inversion method is illustrated with L-band NASA/JPL UAVSAR from AfriSAR data conducted over the Gabon Lope National Park and airborne LiDAR data. Compared to high-accuracy airborne LiDAR data, the obtained forest canopy height from the proposed approach exhibited higher accuracy (R2 = 0.92, RMSE = 6.09 m). The results demonstrate the potential and merit of the synergistic combination of PolSAR (volume scattering power) and sparse LiDAR (forest height) measurements for forest height estimation. Additionally, our approach achieves good performance in forest height estimation, with accuracy comparable to that of the multi-baseline PolInSAR-based inversion method (RMSE = 5.80 m), surpassing traditional PolSAR-based methods with an accuracy of 10.86 m. Given the simplicity and efficiency of the proposed method, it has the potential for large-scale forest height estimation applications when only single-temporal dual-polarization acquisitions are available.

YOLOSAR-Lite: a lightweight framework for real-time ship detection in SAR imagery

ABSTRACT Real-time ship detection using SAR images is crucial for maritime operations but remains challenging due to issues such as sidelobes, complex background interferences, and target defocusing. Studies have shown that deep learning can address these problems, current deep learning algorithms often suffer from achieving the necessary accuracy due to complex noise and background characteristics. Additionally, advanced algorithms excelling in feature extraction usually require significant computational resources. This paper proposes a novel deep learning approach based on YOLOv8 to improve accuracy while reducing model complexity and computational burden. The approach innovatively integrates a knowledge distillation module into the YOLOv8. We further enhance the model by replacing the original YOLOv8 backbone, neck, and head with lightweight alternatives: HGNetv2, SlimNeck, and a newly designed decoupled head. The enhanced model, YOLOSAR-Lite, demonstrated exceptional performance on both accuracy and model complexity metrics. Rigorous evaluation on the Official-SSDD dataset and the SAR-Ship-Dataset revealed recognition accuracies of 95.32% and 96.06%, respectively. Furthermore, YOLOSAR-Lite achieved significant reduction in complexity compared to the YOLOv8: 4.48G FLOPs (45.30% reduction), 2.05M parameters (31.89% reduction), and 3.91MB model size (31.76% reduction). These improvements demonstrate YOLOSAR-Lite achieves higher accuracy with minimal complexity. Consequently, YOLOSAR-Lite is a promising solution for real-time ship detection.

AFMSFFNet: An Anchor-Free-Based Feature Fusion Model for Ship Detection

This paper aims to improve a small-scale object detection model to achieve detection accuracy matching or even surpassing that of complex models. Efforts are made in the module design phase to minimize parameter count as much as possible, thereby providing the potential for rapid detection of maritime targets. Here, this paper introduces an innovative Anchor-Free-based Multi-Scale Feature Fusion Network (AFMSFFNet), which improves the problems of missed detection and false positives, particularly in inshore or small target scenarios. Leveraging the YOLOX tiny as the foundational architecture, our proposed AFMSFFNet incorporates a novel Adaptive Bidirectional Fusion Pyramid Network (AB-FPN) for efficient multi-scale feature fusion, enhancing the saliency representation of targets and reducing interference from complex backgrounds. Simultaneously, the designed Multi-Scale Global Attention Detection Head (MGAHead) utilizes a larger receptive field to learn object features, generating high-quality reconstructed features for enhanced semantic information integration. Extensive experiments conducted on publicly available Synthetic Aperture Radar (SAR) image ship datasets demonstrate that AFMSFFNet outperforms the traditional baseline models in detection performance. The results indicate an improvement of 2.32% in detection accuracy compared to the YOLOX tiny model. Additionally, AFMSFFNet achieves a Frames Per Second (FPS) of 78.26 in SSDD, showcasing superior efficiency compared to the well-established performance networks, such as faster R-CNN and CenterNet, with efficiency improvement ranging from 4.7 to 6.7 times. This research provides a valuable solution for efficient ship detection in complex backgrounds, demonstrating the efficacy of AFMSFFNet through quantitative improvements in accuracy and efficiency compared to existing models.

Mini-Satellite Fucheng 1 SAR: Interferometry to Monitor Mining-Induced Subsidence and Comparative Analysis with Sentinel-1

Mining-induced subsidence poses a serious hazard to the surrounding environment and infrastructure, necessitating the detection of such subsidence for effective disaster mitigation and the safeguarding of local residents. Fucheng 1 is the first high-resolution mini-satellite interferometric Synthetic Aperture Radar (SAR) launched by China in June 2023. In this study, we used Fucheng 1 SAR images to analyze mining-induced subsidence in Karamay by InSAR Stacking and D-InSAR. The findings were compared with Sentinel-1A imagery to evaluate the effectiveness of Fucheng 1 in monitoring subsidence and its interferometric performance. Analysis revealed significant mining-induced subsidence in Karamay, and the results from Fucheng 1 closely corresponded with those from Sentinel-1A, particularly regarding the extent of the subsidence. It is indicated that the precision of Fucheng 1 SAR imagery has reached leading standards. In addition, due to its higher resolution, the maximum detectable deformation gradient (MDDG) of Fucheng 1 is 2.15 times higher than that of Sentinel images. This study provides data support for the monitoring of mining-induced subsidence in the Karamay and give a theoretical basis for the application of Fucheng 1 in the field of Geohazard monitoring.

Super-resolution water body mapping with a feature collaborative CNN model by fusing Sentinel-1 and Sentinel-2 images

Mapping water bodies from remotely sensed imagery is crucial for understanding hydrological and biogeochemical processes. The identification of water extent is mainly dependent on optical and synthetic aperture radar (SAR) images. However, the use of remote sensing for water body mapping is often undermined by the mixed pixel dilemma inherent to traditional hard classification approaches. At the same time, the presence of clouds in optical imagery and speckle noise in SAR imagery, coupled with the difficulty in differentiating between water-like surfaces and actual water bodies, significantly compromise the accuracy of water body identification. This paper proposes a DEEP feature collaborative convolutional neural network (CNN) for Water Super-Resolution Mapping based on Optical and SAR images (DeepOSWSRM), which collaboratively leverages Sentinel-1 and Sentinel-2 imagery to address the challenges of missing data and mixed pixels. The Sentinel-1 image provides complementary water distribution information for the cloudy areas of the Sentinel-2 image, while the Sentinel-2 image enhances the perception capabilities for small water bodies in the Sentinel-1 image. Using PlanetScope imagery as the true reference data, the effectiveness of the proposed method was assessed through two experimental scenarios: one utilizing synthetic coarse-resolution imagery degraded from Sentinel-1 and Sentinel-2 data and another using actual Sentinel-1 and Sentinel-2 data, encompassing both simulated and real cloud conditions. A comparative analysis was conducted against three state-of-the-art CNN-based water mapping methods and two CNN SRM methods. The findings demonstrate that the proposed DeepOSWSRM method successfully produces accurate, fine-resolution water body maps, with its performance mainly benefiting from the fusion of SAR and optical images.

SAR-NTV-YOLOv8: A Neural Network Aircraft Detection Method in SAR Images Based on Despeckling Preprocessing

Monitoring aircraft using synthetic aperture radar (SAR) images is a very important task. Given its coherent imaging characteristics, there is a large amount of speckle interference in the image. This phenomenon leads to the scattering information of aircraft targets being masked in SAR images, which is easily confused with background scattering points. Therefore, automatic detection of aircraft targets in SAR images remains a challenging task. For this task, this paper proposes a framework for speckle reduction preprocessing of SAR images, followed by the use of an improved deep learning method to detect aircraft in SAR images. Firstly, to improve the problem of introducing artifacts or excessive smoothing in speckle reduction using total variation (TV) methods, this paper proposes a new nonconvex total variation (NTV) method. This method aims to ensure the effectiveness of speckle reduction while preserving the original scattering information as much as possible. Next, we present a framework for aircraft detection based on You Only Look Once v8 (YOLOv8) for SAR images. Therefore, the complete framework is called SAR-NTV-YOLOv8. Meanwhile, a high-resolution small target feature head is proposed to mitigate the impact of scale changes and loss of depth feature details on detection accuracy. Then, an efficient multi-scale attention module was proposed, aimed at effectively establishing short-term and long-term dependencies between feature grouping and multi-scale structures. In addition, the progressive feature pyramid network was chosen to avoid information loss or degradation in multi-level transmission during the bottom-up feature extraction process in Backbone. Sufficient comparative experiments, speckle reduction experiments, and ablation experiments are conducted on the SAR-Aircraft-1.0 and SADD datasets. The results have demonstrated the effectiveness of SAR-NTV-YOLOv8, which has the most advanced performance compared to other mainstream algorithms.

Automatic Water Body Extraction from SAR Images Based on MADF-Net

Water extraction from synthetic aperture radar (SAR) images has an important application value in wetland monitoring, flood monitoring, etc. However, it still faces the problems of low generalization, weak extraction ability of detailed information, and weak suppression of background noises. Therefore, a new framework, Multi-scale Attention Detailed Feature fusion Network (MADF-Net), is proposed in this paper. It comprises an encoder and a decoder. In the encoder, ResNet101 is used as a solid backbone network to capture four feature levels at different depths, and then the proposed Deep Pyramid Pool (DAPP) module is used to perform multi-scale pooling operations, which ensure that key water features can be captured even in complex backgrounds. In the decoder, a Channel Spatial Attention Module (CSAM) is proposed, which focuses on feature areas that are critical for the identification of water edges by fusing attention weights in channel and spatial dimensions. Finally, the high-level semantic information is effectively fused with the low-level edge features to achieve the final water detection results. In the experiment, Sentinel-1 SAR images of three scenes with different characteristics and scales of water body are used. The PA and IoU of water extraction by MADF-Net can reach 92.77% and 89.03%, respectively, which obviously outperform several other networks. MADF-Net carries out water extraction with high precision from SAR images with different backgrounds, which could also be used for the segmentation and classification of other tasks from SAR images.

OS-PSO: A Modified Ratio of Exponentially Weighted Averages-Based Optical and SAR Image Registration.

Optical and synthetic aperture radar (SAR) images exhibit non-negligible intensity differences due to their unique imaging mechanisms, which makes it difficult for classical SIFT-based algorithms to obtain sufficiently correct correspondences when processing the registration of these two types of images. To tackle this problem, an accurate optical and SAR image registration algorithm based on the SIFT algorithm (OS-PSO) is proposed. First, a modified ratio of exponentially weighted averages (MROEWA) operator is introduced to resolve the sudden dark patches in SAR images, thus generating more consistent gradients between optical and SAR images. Next, we innovatively construct the Harris scale space to replace the traditional difference in the Gaussian (DoG) scale space, identify repeatable key-points by searching for local maxima, and perform localization refinement on the identified key-points to improve their accuracy. Immediately after that, the gradient location orientation histogram (GLOH) method is adopted to construct the feature descriptors. Finally, we propose an enhanced matching method. The transformed relation is obtained in the initial matching stage using the nearest neighbor distance ratio (NNDR) and fast sample consensus (FSC) methods. And the re-matching takes into account the location, scale, and main direction of key-points to increase the number of correctly corresponding points. The proposed OS-PSO algorithm has been implemented on the Gaofen and Sentinel series with excellent results. The superior performance of the designed registration system can also be applied in complex scenarios, including urban, suburban, river, farmland, and lake areas, with more efficiency and accuracy than the state-of-the-art methods based on the WHU-OPT-SAR dataset and the BISTU-OPT-SAR dataset.