Large Synthetic Aperture Radar Research Articles

Density-based ship detection in SAR images: Extension to a self-similarity perspective

Nonlocal self-similarity is an important property of Synthetic Aperture Radar (SAR) images to characterize the repetitiveness of features embodied by SAR images within nonlocal areas and has been used for enhancement of SAR images. Existing SAR ship detectors often independently handle small sub-images cropped from a large marine SAR image and do not exploit the nonlocal self-similarity therein. In this paper, we propose a new ship detector from the perspective of nonlocal self-similarity in SAR images to improve the ship detection performance, basically including three stages: prescreening, intra-cue calculation, and inter-cue calculation. In the pre-screening stage, we design a new Histogram-based Density (HD) feature to rapidly select candidate sub-images potentially containing ship targets from a large SAR image. In the intra-cue calculation stage, target cues within a single candidate sub-image are extracted. In the inter-cue calculation stage, thanks to the nonlocal self-similarity among different candidate sub-images in terms of density features, we innovatively extract a weighted superpixel-HD map to obtain accumulated intra-cues across all the candidate sub-images. Finally, for each candidate sub-image, we fuse its inter-cue and intra-cue to obtain final detection results. Experimental results based on real SAR images show that our newly proposed method provides a better target-to-clutter contrast and ship detection performance than those of other state-of-the-art detection approaches.

Deceptive Jamming Algorithm against Synthetic Aperture Radar in Large Squint Angle Mode Based on Non-Linear Chirp Scaling and Low Azimuth Sampling Reconstruction

Due to the complex range migration characteristics of large squint angle synthetic aperture radar (SAR), it is difficult for traditional SAR deceptive jamming algorithms to balance focusing ability and computational efficiency. There is an urgent demand for proposing a deceptive jamming algorithm against large squint angle SAR in the field of SAR jamming. This article proposes a deceptive jamming algorithm against SAR with large squint angles based on non-linear chirp scaling and low azimuth sampling reconstruction (NLCSR). The NLCSR algorithm uses a high-order approximation of a high-precision model to accurately construct the jammer’s frequency response (JFR) function. In line with the notion of low azimuth sampling processing of the transformation domain, the construction of the space-variant azimuth modulation phase item is completed using the non-linear chirp scaling method. Compared with the traditional deceptive jamming algorithms against the large squint angle SAR, the NLCSR algorithm only needs Fourier transform and complex multiplication while ensuring the focusing ability, which is easier to implement on an efficient parallel digital signal processor based on fast Fourier transform (FFT). Simulation results prove the superior property of the NLCSR algorithm in focusing ability and computational efficiency. Compared to the existing large squint angle SAR deceptive jamming algorithm, the focusing ability of the NLCSR algorithm is almost the same, and the calculation efficiency is improved by at least 52.1%.

Observing Near-Real-Time Volcanoes Deformation Using InSAR – Time Series of LiCSBAS in Indonesia

The Synthetic Aperture Radar (SAR) sensors onboard satellites are leading in advance to monitor physical changes of volcano edifices such as deformation. A few days of temporal resolution imageries provided by Sentinel-1 SAR make it possible to monitor volcanoes worldwide in near-real-time monitoring. The Sentinel-1 constellation produces a large amount of SAR images valuable for monitoring volcano deformation and hazard mitigation. However, processing the Interferometric SAR (InSAR) for a large SAR dataset is time-consuming and requires high-performance computers. Overcoming the problem, the Center for Observation & Modelling of Earthquakes, Volcanoes & Tectonics (COMET) created a program called Looking Inside the Continents from Space of Sentinel Aperture Radar (LiCSAR) integrated to InSAR time-series analysis of LiCSBAS InSAR to process and calculate the interferogram deformation in time series efficiently. This study presents the LiCSBAS observation for volcanic activities at Mts. Sinabung and Agung in North Sumatra and Bali, respectively. We have observed the activities of both volcanoes from 2015 to 2022. We have analyzed 1824 and 848 interferograms of Mt. Sinabung and Mt. Agung, respectively. The LiCSBAS InSAR time-series analysis was successfully processed for Mts. Sinabung and Agung despite being located under tropics. Velocity deformation of Mt. Sinabung is dominated by uplift around the summit, consistent with measurement by the Center for Volcanology and Geological Hazard Mitigation (CVGHM). In the period of January to March 2017, the lava dome measurement using laser distance meter showed an increase from 1.2 Mm3 to 1.7 Mm3, while LiCSBAS detected an uplift movement of 14 mm. On the contrary, the deformation of Mt. Agung at the summit of lava dome is 3 mm/yr indicate uplift, while subsidence is observed around the crater.

CCNR: Cross-regional context and noise regularization for SAR image segmentation

Semantic segmentation, a fundamental research direction in synthetic aperture radar (SAR) image interpretation, has significant application value for multiple sectors. However, noise, multi-style terrains, geometric distortion, and shadows make SAR image segmentation challenging. Although existing deep learning algorithms tend to mine the semantic relationship between pixels within individual images, they disregard the global context of the training data and the semantic relationship between different regions from different images. Moreover, the noise resistance of the networks is not strong enough. All of these factors degrade the performance of SAR image segmentation algorithms. In this study, a semantic segmentation algorithm using a cross-regional context and noise regularization (CCNR) for SAR images is proposed. CCNR has three heads to output segmentation results, representation features of each pixel, and reconstructed images. The self-attention and contrastive learning are adopted to explore region-level semantic relations and pixel-level semantic relations between different images, achieving the aggregation of pixels with the same class. Furthermore, to improve the robustness of the network, noise regularization is applied to impose additional constraints on the encoder and segmentation results. The results of experiments conducted on three large scene SAR images prove the efficacy of CCNR. Compared with other comparative algorithms, the proposed CCNR achieves more optimal performance and increased robustness.

Utilising Sentinel-1's Orbital Stability for Efficient Pre-Processing of Radiometric Terrain Corrected Gamma Nought Backscatter.

Radiometric Terrain Corrected (RTC) gamma nought backscatter, which was introduced around a decade ago, has evolved into the standard for analysis-ready Synthetic Aperture Radar (SAR) data. While working with RTC backscatter data is particularly advantageous over undulated terrain, it requires substantial computing resources given that the terrain flattening is more computationally demanding than simple orthorectification. The extra computation may become problematic when working with large SAR datasets such as the one provided by the Sentinel-1 mission. In this study, we examine existing Sentinel-1 RTC pre-processing workflows and assess ways to reduce processing and storage overheads by considering the satellite's high orbital stability. By propagating Sentinel-1's orbital deviations through the complete pre-processing chain, we show that the local contributing area and the shadow mask can be assumed to be static for each relative orbit. Providing them as a combined external static layer to the pre-processing workflow, and streamlining the transformations between ground and orbit geometry, reduces the overall processing times by half. We conducted our experiments with our in-house developed toolbox named wizsard, which allowed us to analyse various aspects of RTC, specifically run time performance, oversampling, and radiometric quality. Compared to the Sentinel Application Platform (SNAP) this implementation allowed speeding up processing by factors of 10-50. The findings of this study are not just relevant for Sentinel-1 but for all SAR missions with high spatio-temporal coverage and orbital stability.

HRLE-SARDet: A Lightweight SAR Target Detection Algorithm Based on Hybrid Representation Learning Enhancement

In recent years, deep learning has been widely used in remote sensing, especially in the field of synthetic aperture radar (SAR) image target detection. However, all of these deep learning models continue increasing the network’s depth and width without maintaining a good balance between accuracy and speed. Therefore, in this paper, we propose a hybrid representation learning-enhanced SAR target detection algorithm based on the unique features of SAR images from a lightweight perspective called HRLE-SARDet. First, we design a lightweight and scattering feature extraction backbone that is more suitable for SAR image data. Second, for the multiscale feature discrepancy, we design a new multiscale feature fusion neck. Next, to better extract the scattering information from small targets of SAR images and improve the detection accuracy, we design a lightweight hybrid representation learning enhancement module. Finally, to better fit target detection for SAR image datasets, we redesign a more flexible loss function, which allows for an easy adjustment of the importance of polynomial bases according to the target task and dataset. Extensive experimental results on three SAR image ship target datasets (SSDD, AIR-SARShip-2.0, and HRSID) and a newly released large multiclass target SAR dataset (MSAR-1.0) show that our HRLE-SARDet achieves 98.4%, 79.2%, 92.5%, and 88.4% mAPs (mean Average Precision) with only 1.09 M parameters and 2.5 G floating-point operations (FLOPs) on the SSDD, AIR-SARShip-2.0, HRSID, and MSAR-1.0 datasets, which is an excellent performance.

A Robust Two-Stage Registration Algorithm for Large Optical and SAR Images

Accurate optical and synthetic aperture radar (SAR) image registration is crucial to multisensor remote sensing applications. Though several algorithms have been proposed, the practical implementation that directly matches large and high-resolution optical and SAR images remains underexplored by the community. Equipped with the satellite positioning parameters, optical and SAR images can be roughly registered based on geographic coordinates. However, the relative positioning accuracy is still dozens, even hundreds of pixels due to the inaccuracies of sensor parameters and elevation. Consequently, we propose a robust registration algorithm, which consists of two stages, where the horizontal positioning errors can be reduced in the first stage, and then, we fine-tune the correspondences in the second stage. Specifically, we propose a novel template matching method based on the dilated convolutional feature (DCF) and epipolar-oriented phase correlation. DCF is constructed by a depthwise-separable dilated convolution with multichannel gradients, which are generated by the Sobel operator for optical images and a ratio of exponentially weighted averages (ROEWA) operator for SAR images. Due to the large reception field of dilated convolution, DCF can retain invariance even for large relative positioning errors. The epipolar-oriented rectangle template, which stretches along the epipolar line, is then proposed to capture more overlapping areas compared to square templates. Furthermore, the outlier removal is implemented in the coordinate system of optical images to avoid the effect of range compression in SAR images. Inliers are finally used to refine the rational polynomial coefficients (RPCs) based on the bundle adjustment technique. Experimental results on high-resolution optical and SAR image products of various scenarios demonstrate the effectiveness of the proposed registration framework. The relative positioning errors of the refined RPCs can be reduced from hundreds of pixels to the subpixel level.

CFAR-Guided Dual-Stream Single-Shot Multibox Detector for Vehicle Detection in SAR Images

Recently, deep neural network has achieved enormous success on target detection but how to better apply these methods to synthetic aperture radar (SAR) images is still a challenge due to the following two reasons: the characteristics of SAR images are not utilized sufficiently; SAR images contain a smaller number of targets, and thus detectors face more serious foreground-background class imbalance problem during training. In this paper, we propose a constant false alarm rate (CFAR) guided dual stream single shot multibox detector to solve the above problems. Firstly, based on the proposed dual sub-networks and the interactive channel-spatial attention fusion (ICSAF) module, we effectively utilize the strong scattering characteristic of SAR target to improve the feature extraction ability. Then we integrate the CFAR indicative map into focal loss to alleviate the foreground-background class imbalance problem. In addition, a non-maximum suppression method based on area-ratio (AR-NMS) is proposed for problem caused by dividing large scene SAR image into blocks. The whole framework can realize end-to-end joint training, and the experimental results on miniSAR real data demonstrate the effectiveness of the proposed method.

Attentional Feature Refinement and Alignment Network for Aircraft Detection in SAR Imagery

Aircraft detection in Synthetic Aperture Radar (SAR) imagery is a challenging task in SAR Automatic Target Recognition (SAR ATR) areas due to aircraft's extremely discrete appearance, obvious intraclass variation, small size and serious background's interference. In this paper, a single-shot detector namely Attentional Feature Refinement and Alignment Network (AFRAN) is proposed for detecting aircraft in SAR images with competitive accuracy and speed. Specifically, three significant components including Attention Feature Fusion Module (AFFM), Deformable Lateral Connection Module (DLCM) and Anchor-guided Detection Module (ADM), are carefully designed in our method for refining and aligning informative characteristics of aircraft. To represent characteristics of aircraft with less interference, low-level textural and high-level semantic features of aircraft are fused and refined in AFFM throughly. The alignment between aircraft's discrete back-scatting points and convolutional sampling spots is promoted in DLCM. Eventually, the locations of aircraft are predicted precisely in ADM based on aligned features revised by refined anchors. To evaluate the performance of our method, a self-built SAR aircraft sliced dataset and a large scene SAR image are collected. Extensive quantitative and qualitative experiments with detailed analysis illustrate the effectiveness of the three proposed components. Furthermore, the topmost detection accuracy and competitive speed are achieved by our method compared with other domain-specific,e.g., DAPN, PADN, and general CNN-based methods,e.g., FPN, Cascade R-CNN, SSD, RefineDet and RPDet.

Revisiting SLIC: Fast Superpixel Segmentation of Marine SAR Images Using Density Features

The simple linear iterative clustering (SLIC) has been shown as an efficient and widely used superpixel-based algorithm for segmenting marine synthetic aperture radar (SAR) images. However, SLIC does not consider the fact that the density of ship target pixels is significantly lower than that of sea clutter pixels, leading to a waste of computational cost and memory resources on lots of pure clutter areas and to the degradation of the compactness of superpixels. To address the aforementioned issues, we develop a new density-based SLIC (DSLIC) method for the superpixel-based segmentation of marine SAR images. In the initialization stage of our DSLIC, all the subimages in a large marine SAR image are rapidly prescreened via a new density-driven classifier, where most of the subimages only occupied by clutter pixels with comparatively high density are discarded and do not need to be segmented in the subsequent local clustering stage. The retained subimages contain both the clutter and potential target areas. This prescreening operation results in higher computation efficiency and memory savings. In the local clustering stage of DSLIC, besides the intensity proximity and the spatiality proximity (used in SLIC), the sparsity proximity (measured by density distances) is considered to reduce the coexistence of sparse target pixels with low density and nonsparse clutter pixels with high density within superpixels. Our theoretical and experimental results show that the proposed DSLIC method is faster and requires less memory than SLIC and other state-of-the-art superpixel-based segmentation methods for marine SAR images with similar or better segmentation accuracy.

Explore Better Network Framework for High-Resolution Optical and SAR Image Matching

To fully explore the complementary information from optical and synthetic aperture radar (SAR) imageries, they need first to be coregistered with high accuracy. Due to the vast radiometric and geometric disparity, the problem to match high-resolution optical and SAR images is quite challenging. The present deep learning-based methods have shown advantages over the traditional approaches, but the performance increment is not significant. In this article, we explore a better network framework for high-resolution optical and SAR image matching from three aspects. First, we propose an effective multilevel feature fusion method, which helps to take advantage of both the low-level fine-grained features for precious feature location and the high-level semantic features for better discriminative ability. Second, a feature channel excitation procedure is conducted using a novel multifrequency channel attention module, which is able to make image features of different types and multiple levels effectively collaborate with each other and produce image matching features with high diversity. Third, the self-adaptive weighting loss is introduced, with which, each sample is assigned with an adaptive weighting factor, and therefore, information buried in all nearby samples can be better exploited. Under a pseudo-Siamese architecture, the proposed optical and SAR image matching network (OSMNet) is trained and tested on a large and diverse high-resolution optical and SAR dataset. Extensive experiments demonstrate that each component of the proposed deep framework helps to improve the matching accuracy. Also, the OSMNet shows overwhelming superior to the state-of-the-art handcrafted approaches on imageries of different land-cover types.

A Benchmark Sentinel-1 SAR Dataset for Airport Detection

Airports are important targets in both military and civilian tasks. Synthetic aperture radar (SAR) based airport detection has received increasing attention in recent years. However, due to the high cost of the SAR imaging and annotation process, there is no publicly available SAR dataset for airport detection, which leads to the fact that deep learning methods have not been applied to airport detection and no unified evaluation benchmark for traditional methods. To provide a benchmark for airport detection studies in SAR images, a large SAR airport dataset (SAD) is presented in this paper. It contains 624 SAR images from Sentinel 1B, covering 104 instances of airports with different scales, orientations and shapes, which can realistically reflect the real world. Experiments with four deep learning-based methods and three traditional methods on this dataset demonstrate its effectiveness and challenge. It serves for the development of state-of-the-art airport detection algorithms or other related tasks. In addition, we found that airport runways in SAR images always have some parallel line segments. Inspired by this, the fusion of deep features and line segments is achieved by designing a line segment detector branch, which further improves the accuracy of airport detection.

Tackling SAR Imagery Ship Classification Imbalance via Deep Convolutional Generative Adversarial Network

Shipping constitutes the majority of the world trade, and Synthetic Aperture Radar (SAR) imagery is the primary involuntary all-condition ship monitoring and classification approach. However, large SAR datasets for deep learning are difficult to curate, usually leading to imbalanced classes. Herein, conventional methods such as weighted cost function or over-sampling are shown to be insufficient for our application. Therefore, we propose to utilize a Deep Convolutional Generative Adversarial Network (DCGAN) to be trained on the minority class, and generate new SAR chips to balance the training dataset. A case study is consequently presented that utilizes our methodology, and a base classifier is devised to evaluate its performance. The investigation of the classification metrics confirms that DCGAN is an effective alternative for tackling imbalanced ship SAR data for deep learning applications.

TanDEM-X: 10 Years of Formation Flying Bistatic SAR Interferometry

On June 21, 2010, the TanDEM-X mission was launched and opened a new era in spaceborne radar remote sensing. The first formation flying radar system was built by extending the TerraSAR-X synthetic aperture radar (SAR) mission by a second, TerraSAR-X-like satellite TanDEM-X. The resulting large single-pass SAR interferometer features flexible baseline selection, enabling the acquisition of highly accurate cross-track interferograms not impacted by temporal decorrelation and atmospheric disturbances. The primary objective of the mission was the generation of a global Digital Elevation Model (DEM) with unprecedented accuracy (12-m horizontal resolution and 2-m relative height accuracy). The main mission phase for DEM data acquisition has been finished in 2014; the processing of the global TanDEM-X DEM was concluded in September 2016. The final DEM product is well within specifications and features an extremely low percentage of void areas. It is of fundamental importance for a wide range of commercial and scientific applications. But the scientific exploitation of TanDEM-X is not limited to the DEM. TanDEM-X has unique capabilities, including along-track interferometry, and new bistatic and multistatic SAR techniques, that support numerous secondary mission objectives. Indeed, some of these experiments were directly performed during the DEM acquisition phase, when suitable satellite formation geometries were available. Moreover, regular acquisitions over selected super test sites enabled multitemporal analyses. A dedicated science phase after the DEM acquisitions included up to 4 km cross-track baselines, operation in the so-called Dual-Receive Antenna mode, as well as a period in pursuit monostatic flight formation. Comparisons of the TanDEM-X DEM with that of SRTM, or among multitemporal TanDEM-X data, revealed dramatic, ongoing, changes in Earth's topography, especially over ice and forests. In the last 3.5 years the mission has further acquired data for a global change layer showing the height changes relative to the first global DEM dataset. The so-called “Change DEM” is planned for release in 2021. Despite being well beyond their design lifetime, both satellites are still fully functional and have enough consumables for several additional years. Therefore, bistatic operations continue with a focus on changes in the cryosphere, biosphere, and densely populated urban areas.

A Gamma Distribution-Based Fuzzy Clustering Approach for Large Area SAR Image Segmentation

Synthetic aperture radar (SAR) image segmentation is a challenge due to its inherent speckle. Gamma distribution is believed to be an appropriate statistical model to describe the characteristics of speckle in SAR images. In this letter, a fuzzy clustering algorithm based on gamma distribution for SAR image segmentation is proposed, in which the Stirling equation is used to approach the gamma function in the dominator of gamma distribution under the assumption of mean field theory to make the shape parameter <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> derivable. Then, the value range of the estimated <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> is demonstrated to meet the requirement of gamma distribution by Jensen’s inequality. Experimental results show that the proposed method gives promising results in SAR image (including large area SAR image) segmentation and effectively suppresses the influence of speckle.

Very High Resolution SAR Imaging With DGPS-Supported Airborne X-Band Data

High spatial resolution imaging in synthetic-aperture radar (SAR) can provide accurate monitoring capacity and has been gaining great attention recently in the fields of military and civilian. Apparently, the slant range resolution of the SAR system depends on the radar operating bandwidth. Currently, the large bandwidth signal synthesizing technology of the stepped frequency chirp signal waveform is highly practical for achieving high spatial resolution. However, the system structure and the corresponding signal processing technology become more complex. In order to verify the feasibility and operability of the large full-bandwidth system, a 3.6-GHz full-bandwidth airborne experimental SAR system operating at X-band, featured by full-bandwidth transmitting and receiving, has been designed by the Department of Space Microwave Remote Sensing System, Institute of Electronics, Chinese Academy of Sciences, as a test bed for the development and implementation of the future spaceborne realizations. For this large full-bandwidth SAR system, in addition to the hardware resource, the motion compensation (MOCO) is an urgent problem. The improvement of spatial resolution will aggravate the effect of motion errors. In order to focus the SAR images accurately, this article presents a technical approach by utilizing the differential global positioning system (DGPS) technology to improve the position accuracy of the inertial measurement unit device. Meanwhile, considering the significant deviation of range cell migration correction (RCMC) due to the residual range-variant errors, this article proposes an accurate MOCO strategy with DGPS-supported to implement the second-order MOCO, space-variant residual range envelope, and space-variant residual phase error in azimuth before RCMC. Finally, this article presents the outfield experiment and reports the corresponding analysis and processing results of an outfield flight experiment successfully conducted in March 2019.

A novel biologically-inspired target detection method based on saliency analysis for synthetic aperture radar (SAR) imagery

Saliency Object Detection (SOD) models driven by the biologically-inspired Focus of Attention (FOA) mechanism can result in highly accurate saliency maps. However, their application in high-resolution Synthetic Aperture Radar (SAR) images entails a number of intractable problems due to complex backgrounds. In this paper, we propose a novel hierarchical self-diffusion saliency (HSDS) method for detecting vehicle targets in large scale SAR images. To reduce the influence of cluttered returns on saliency analysis, we learn a weight vector from the training set to capture optimal initial saliency of the superpixels during saliency diffusion. By accounting for the multiple sizes of background objects, the saliency analysis is implemented in multi-scale space, and a saliency fusion strategy employed to integrate the multi-scale saliency maps. Simulation experiments demonstrate that our proposed method can produce a more accurate and stable detection performance, with decreased false alarms, compared to benchmark approaches.

Arbitrary-oriented target detection in large scene sar images

Target detection in the field of synthetic aperture radar (SAR) has attracted considerable attention of researchers in national defense technology worldwide, owing to its unique advantages like high resolution and large scene image acquisition capabilities of SAR. However, due to strong speckle noise and low signal-to-noise ratio, it is difficult to extract representative features of target from SAR images, which greatly inhibits the effectiveness of traditional methods. In order to address the above problems, a framework called contextual rotation region-based convolutional neural network (RCNN) with multilayer fusion is proposed in this paper. Specifically, aimed to enable RCNN to perform target detection in large scene SAR images efficiently, maximum sliding strategy is applied to crop the large scene image into a series of sub-images before RCNN. Instead of using the highest-layer output for proposal generation and target detection, fusion feature maps with high resolution and rich semantic information are constructed by multilayer fusion strategy. Then, we put forwards rotation anchors to predict the minimum circumscribed rectangle of targets to reduce redundant detection region. Furthermore, shadow areas serve as contextual features to provide extraneous information for the detector identify and locate targets accurately. Experimental results on the simulated large scene SAR image dataset show that the proposed method achieves a satisfactory performance in large scene SAR target detection.

Large-Scale Automatic Vessel Monitoring Based on Dual-Polarization Sentinel-1 and AIS Data

This research addresses the use of dual-polarimetric descriptors for automatic large-scale ship detection and characterization from synthetic aperture radar (SAR) data. Ship detection is usually performed independently on each polarization channel and the detection results are merged subsequently. In this study, we propose to make use of the complex coherence between the two polarization channels of Sentinel-1 and to perform vessel detection in this domain. Therefore, an automatic algorithm, based on the dual-polarization coherence, and applicable to entire large scale SAR scenes in a timely manner, is developed. Automatic identification system (AIS) data are used for an extensive and also large scale cross-comparison with the SAR-based detections. The comparative assessment allows us to evaluate the added-value of the dual-polarization complex coherence, with respect to SAR intensity images in ship detection, as well as the SAR detection performances depending on a vessel’s size. The proposed methodology is justified statistically and tested on Sentinel-1 data acquired over two different and contrasting, in terms of traffic conditions, areas: the English Channel the and Pacific coastline of Mexico. The results indicate a very high SAR detection rate, i.e., &gt;80%, for vessels larger than 60 m and a decrease of detection rate up to 40 % for smaller size vessels. In addition, the analysis highlights many SAR detections without corresponding AIS positions, indicating the complementarity of SAR with respect to cooperative sources for detecting dark vessels.

SAR-to-Optical Image Translation Using Supervised Cycle-Consistent Adversarial Networks

Optical remote sensing (RS) data suffer from the limitation of bad weather and cloud contamination, whereas synthetic aperture radar (SAR) can work under all weather conditions and overcome this disadvantage of optical RS data. However, due to the imaging mechanism of SAR and the speckle noise, untrained people are difficult to recognize the land cover types visually from SAR images. Inspired by the excellent image-to-image translation performance of Generative Adversarial Networks (GANs), a supervised Cycle-Consistent Adversarial Network (S-CycleGAN) was proposed to generate large optical images from the SAR images. When the optical RS data are unavailable or partly unavailable, the generated optical images can be alternative data that aid in land cover visual recognition for untrained people. The main steps of SAR-to-optical image translation were as follows. First, the large SAR image was split to small patches. Then S-CycleGAN was used to translate the SAR patches to optical image patches. Finally, the optical image patches were stitched to generate the large optical image. A paired SAR-optical image dataset which covered 32 Chinese cities was published to evaluate the proposed method. The dataset was generated from Sentinel-1 (SEN-1) SAR images and Sentinel-2 (SEN-2) multi-spectral images. S-CycleGAN was applied to two experiments, which were SAR-to-optical image translation and cloud removal, and the results showed that S-CycleGAN could keep both the land cover and structure information well, and its performance was superior to some famous image-to-image translation models.