Real Synthetic Aperture Radar Images Research Articles

SAR Image Despeckling Based on Denoising Diffusion Probabilistic Model and Swin Transformer

The speckle noise inherent in synthetic aperture radar (SAR) imaging has long posed a challenge for SAR data processing, significantly affecting image interpretation and recognition. Recently, deep learning-based SAR speckle removal algorithms have shown promising results. However, most existing algorithms rely on convolutional neural networks (CNN), which may struggle to effectively capture global image information and lead to texture loss. Besides, due to the different characteristics of optical images and synthetic aperture radar (SAR) images, the results of training with simulated SAR data may bring instability to the real-world SAR data denoising. To address these limitations, we propose an innovative approach that integrates swin transformer blocks into the prediction noise network of the denoising diffusion probabilistic model (DDPM). By harnessing DDPM’s robust generative capabilities and the Swin Transformer’s proficiency in extracting global features, our approach aims to suppress speckle while preserving image details and enhancing authenticity. Additionally, we employ a post-processing strategy known as pixel-shuffle down-sampling (PD) refinement to mitigate the adverse effects of training data and the training process, which rely on spatially uncorrelated noise, thereby improving its adaptability to real-world SAR image despeckling scenarios. We conducted experiments using both simulated SAR image datasets and real SAR image datasets, evaluating our algorithm from subjective and objective perspectives. The visual results demonstrate significant improvements in noise suppression and image detail restoration. The objective results demonstrate that our method obtains state-of-the-art performance, which outperforms the second-best method by an average peak signal-to-noise ratio (PSNR) of 0.93 dB and Structural Similarity Index (SSIM) of 0.03, affirming the effectiveness of our approach.

Matching UAV-Based Simulated High-Resolution SAR and Real SAR Images

Abstract. Our experiment focuses on generating simulated high-resolution (HR) Synthetic Aperture Radar (SAR) images from the photogrammetric 3D model derived from Unmanned Aerial Vehicle (UAV) imagery. The goal is to match these simulated images with real SAR data of complex scenes containing vegetation, buildings, roads, and the construction of receiving antennas. Simulated SAR reflectivity map was based on ray tracing techniques.Two feature matching algorithms, namely Scale-Invariant Feature Transform (SIFT) and Oriented FAST and Rotated BRIEF (ORB) were employed for co-registration of the simulated to the real SAR images. The effectiveness of these algorithms is evaluated by their matching rates. The experiment results of the matching process highlight the superior performance of ORB over SIFT, attributed to its robustness against rotation variations and noise. ORB emerges as a versatile choice for high-resolution real and simulated SAR registration tasks.

SAR Image Generation Method Using DH-GAN for Automatic Target Recognition.

In recent years, target recognition technology for synthetic aperture radar (SAR) images has witnessed significant advancements, particularly with the development of convolutional neural networks (CNNs). However, acquiring SAR images requires significant resources, both in terms of time and cost. Moreover, due to the inherent properties of radar sensors, SAR images are often marred by speckle noise, a form of high-frequency noise. To address this issue, we introduce a Generative Adversarial Network (GAN) with a dual discriminator and high-frequency pass filter, named DH-GAN, specifically designed for generating simulated images. DH-GAN produces images that emulate the high-frequency characteristics of real SAR images. Through power spectral density (PSD) analysis and experiments, we demonstrate the validity of the DH-GAN approach. The experimental results show that not only do the SAR image generated using DH-GAN closely resemble the high-frequency component of real SAR images, but the proficiency of CNNs in target recognition, when trained with these simulated images, is also notably enhanced.

Contrastive Learning-Based Dual Dynamic GCN for SAR Image Scene Classification.

As a typical label-limited task, it is significant and valuable to explore networks that enable to utilize labeled and unlabeled samples simultaneously for synthetic aperture radar (SAR) image scene classification. Graph convolutional network (GCN) is a powerful semisupervised learning paradigm that helps to capture the topological relationships of scenes in SAR images. While the performance is not satisfactory when existing GCNs are directly used for SAR image scene classification with limited labels, because few methods to characterize the nodes and edges for SAR images. To tackle these issues, we propose a contrastive learning-based dual dynamic GCN (DDGCN) for SAR image scene classification. Specifically, we design a novel contrastive loss to capture the structures of views and scenes, and develop a clustering-based contrastive self-supervised learning model for mapping SAR images from pixel space to high-level embedding space, which facilitates the subsequent node representation and message passing in GCNs. Afterward, we propose a multiple features and parameter sharing dual network framework called DDGCN. One network is a dynamic GCN to keep the local consistency and nonlocal dependency of the same scene with the help of a node attention module and a dynamic correlation matrix learning algorithm. The other is a multiscale and multidirectional fully connected network (FCN) to enlarge the discrepancies between different scenes. Finally, the features obtained by the two branches are fused for classification. A series of experiments on synthetic and real SAR images demonstrate that the proposed method achieves consistently better classification performance than the existing methods.

Studies on High-Resolution Airborne Synthetic Aperture Radar Image Formation with Pseudo-Random Agility of Interpulse Waveform Parameters

By means of alteration of the transmitted linear frequency modulation (LFM) signal waveform parameters, such as pulse width or chirp rate, initial phase, pulse repetition interval (PRI), and chirp rate polarity at every position of synthetic apertures, the pseudo-random agility technology of interpulse waveform parameters in airborne Synthetic Aperture Radar (SAR) actively increases the complexity and uncertainty of radar waveforms. This technology confuses jamming interception receivers, thus improving its anti-interference ability for active coherent jamming, which is one of the main research interests of airborne SAR technology. But the pseudo-random agility technology for interpulse waveform parameters faces certain challenges of large computation and complex system design, which need to be further studied and solved. To address these issues, a processing scheme of high-resolution SAR image formation which is appropriate for agile interpulse waveform parameters is proposed in this paper. This method can deal with multiple agile parameters, not only single ones as in most existing literature. Its computation load is nearly comparable to that of traditional SAR image formation with constant waveform parameters. The high-resolution SAR imaging results obtained by processing SAR raw data with agile interpulse waveform parameters demonstrate the effectiveness of the proposed method. In addition, real SAR images with resolutions of 0.5 m and 0.15 m, which are rarely found in the public literature, are shown under the circumstance of randomly changing the transmitted wideband LFM signal pulse parameters one by one.

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.

Statistical Modeling of Shadows in SAR Imagery

Shadows are a special distortion in synthetic aperture radar (SAR) imaging. They often hamper proper image understanding and target recognition but also offer useful information, and therefore, the statistical modeling of SAR image shadows is imperative. In this endeavor, we systematically deduced the statistical models of shadows in multimodal SAR images, including single-look intensity and amplitude images and multilook intensity and amplitude images in a real domain and complex domain, respectively. In particular, for the filtered SAR image shadow, we introduced the generalized extreme value (GEV) distribution to characterize its statistics. We carried out an experiment on shadows in a real SAR image and conducted chi-square goodness-of-fit tests on the deduced models. Furthermore, we compared the deduced statistical models of shadows with state-of-the-art statistical models of SAR imagery. Finally, suggestions are given for selecting the optimal statistical model of shadows in multimodal SAR images.

Non-Local SAR Image Despeckling Based on Sparse Representation

Speckle noise is an inherent problem of synthetic aperture radar (SAR) images, which not only seriously affects the acquisition of SAR image information, but also greatly reduces the efficiency of image segmentation and feature classification. Therefore, research on how to effectively suppress speckle noise while preserving SAR image content information as much as possible has received increasing attention. Based on the non-local idea of SAR image block-matching three-dimensional (SAR-BM3D) algorithm and the concept of sparse representation, a novel SAR image despeckling algorithm is proposed. The new algorithm uses K-means singular value decomposition (K-SVD) to learn the dictionary to distinguish valid information and speckle noise and constructs a block filter based on K-SVD for despeckling, so as to avoid strong point diffusion problem in SAR-BM3D and achieve better speckle noise suppression with stronger adaptability. The experimental results on real SAR images show that the proposed algorithm achieves better comprehensive effect of speckle noise suppression in terms of evaluation indicators and information preservation of SAR images compared with several existing algorithms.

CBIR-SAR System Using Stochastic Distance.

This article proposes a system for Content-Based Image Retrieval (CBIR) using stochastic distance for Synthetic-Aperture Radar (SAR) images. The methodology consists of three essential steps for image retrieval. First, it estimates the roughness (α^) and scale (γ^) parameters of the GI0 distribution that models SAR data in intensity. The parameters of the model were estimated using the Maximum Likelihood Estimation and the fast approach of the Log-Cumulants method. Second, using the triangular distance, CBIR-SAR evaluates the similarity between a query image and images in the database. The stochastic distance can identify the most similar regions according to the image features, which are the estimated parameters of the data model. Third, the performance of our proposal was evaluated by applying the Mean Average Precision (MAP) measure and considering clippings from three radar sensors, i.e., UAVSAR, OrbiSaR-2, and ALOS PALSAR. The CBIR-SAR results for synthetic images achieved the highest MAP value, retrieving extremely heterogeneous regions. Regarding the real SAR images, CBIR-SAR achieved MAP values above 0.833 for all polarization channels for image samples of forest (UAVSAR) and urban areas (ORBISAR). Our results confirmed that the proposed method is sensitive to the degree of texture, and hence, it relies on good estimates. They are inputs to the stochastic distance for effective image retrieval.

Level Sets Guided by SoDEF-Fitting Energy for River Channel Detection in SAR Images

To achieve river channel detection in SAR (synthetic aperture radar) images, we developed a level-set-based model (LSBM) guided by a designed data-fitting energy which is called the SoDEF (sum of dual exponential functions)-fitting energy. Firstly, we designed a function by computing the sum of dual exponential functions to substitute for the quadratic function, and used it to construct the data-fitting energy. Secondly, the adaptive area-fitting centers (AFCs) were computed based on two kinds of grayscale characteristics, which are more accurate and more stable. Thirdly, the Dirac function in gradient descent flow was displaced by an edge indicator function to help the evolving level sets stop at the target edges. Moreover, some regularized terms were incorporated into the objective function to guarantee the model’s stability. The river channel detection experiments conducted with real SAR images indicated that the developed model is superior to the related state-of-the-art methods in its detection accuracy and efficiency.

Integrated Quantitative Evaluation Method of SAR Filters

An excellent quantitative evaluation method of SAR de-speckling filters needs to contain a comprehensive evaluation of both noise smoothing and edge preservation. However, most existing evaluation models only evaluate a single aspect, while a few comprehensive indicators lack robustness. For this reason, a novel integrated quantitative evaluation method of de-speckling filters is proposed. The proposed evaluation method is weighted by two sub-indicators: the coherent equivalent number of looks and edge preservation evaluation. The evaluation indicator of the coherent equivalent number of looks is built to evaluate the noise-smoothing ability of de-speckling filters, whereas the indicator of edge preservation evaluation is built to evaluate the edge-preserving performance of filtered image. Six filters with an excellent performance, five real synthetic aperture radar images with three bands, four polarization modes, four resolutions, and five common evaluation indexes were used in the experiment. The experimental results show that the evaluation results of the proposed evaluation method were consistent with the visual effect and other indicators, and its feasibility was verified.

Interpretation of Latent Codes in InfoGAN with SAR Images

Generative adversarial networks (GANs) can synthesize abundant photo-realistic synthetic aperture radar (SAR) images. Some modified GANs (e.g., InfoGAN) are even able to edit specific properties of the synthesized images by introducing latent codes. It is crucial for SAR image synthesis since the targets in real SAR images have different properties due to the imaging mechanism. Despite the success of the InfoGAN in manipulating properties, there still lacks a clear explanation of how these latent codes affect synthesized properties; thus, editing specific properties usually relies on empirical trials, which are unreliable and time-consuming. In this paper, we show that latent codes are almost disentangled to affect the properties of SAR images in a nonlinear manner. By introducing some property estimators for latent codes, we are able to decompose the complex causality between latent codes and different properties. Both qualitative and quantitative experimental results demonstrate that the property value can be computed by the property estimators; inversely, the required latent codes can be computed given the desired properties. Unlike the original InfoGAN, which only provides the visual trend between properties and latent codes, the properties of SAR images can be manipulated numerically by latent codes as users expect.

SAR Image Classification Using Markov Random Fields with Deep Learning

Classification algorithms integrated with convolutional neural networks (CNN) display high accuracies in synthetic aperture radar (SAR) image classification. However, their consideration of spatial information is not comprehensive and effective, which causes poor performance in edges and complex regions. This paper proposes a Markov random field (MRF)-based algorithm for SAR image classification which fully considers the spatial constraints between superpixel regions. Firstly, the initialization of region labels is obtained by the CNN. Secondly, a probability field is constructed to improve the distribution of spatial relationships between adjacent superpixels. Thirdly, a novel region-level MRF is employed to classify the superpixels, which combines the intensity field and probability field in one framework. In our algorithm, the generation of superpixels reduces the misclassification at the pixel level, and region-level misclassification is rectified by the improvement of spatial description. Experimental results on simulated and real SAR images confirm the efficacy of the proposed algorithm for classification.

Synthetic aperture radar image despeckling neural network based on maximum a posteriori probability estimation

ABSTRACT Automatic despeckling from synthetic aperture radar (SAR) imagery plays a significant role in many urban applications. Recently, owing to the impressive performance of deep learning, various SAR despeckling methods based on the convolutional neural network (CNN) have been proposed for optical SAR images. However, existing CNN-based despeckling methods and spatial-domain-based methods exhibit certain limitations. The introduction of prior knowledge is not considered in the designing of classical denoising and certain CNN-based methods, thus leading to the loss of edge information and artefacts in the results. To address this issue, an a priori-based deep neural network SAR image despeckling method (SAR-PBDNN) was proposed. In this approach, the Bayesian formula is modelled to calculate the objective function suitable for the despeckling problem and is optimized using the alternating direction method of multipliers algorithm and deep learning method. The introduction of prior information effectively improved the denoising effect of SAR images. The SAR-PBDNN was evaluated using both simulated and real SAR images. The comparisons demonstrate the effectiveness of the proposed method. We conclude that the SAR-PBDNN has the potential to automatically despeckle SAR images with an accuracy that renders it a useful tool for practical application scenarios.

Contrastive Domain Adaptation-Based Sparse SAR Target Classification under Few-Shot Cases

Due to the imaging mechanism of synthetic aperture radar (SAR), it is difficult and costly to acquire abundant labeled SAR images. Moreover, a typical matched filtering (MF) based image faces the problems of serious noise, sidelobes, and clutters, which will bring down the accuracy of SAR target classification. Different from the MF-based result, a sparse image shows better quality with less noise and higher image signal-to-noise ratio (SNR). Therefore, theoretically using it for target classification will achieve better performance. In this paper, a novel contrastive domain adaptation (CDA) based sparse SAR target classification method is proposed to solve the problem of insufficient samples. In the proposed method, we firstly construct a sparse SAR image dataset by using the complex image based iterative soft thresholding (BiIST) algorithm. Then, the simulated and real SAR datasets are simultaneously sent into an unsupervised domain adaptation framework to reduce the distribution difference and obtain the reconstructed simulated SAR images for subsequent target classification. Finally, the reconstructed simulated images are manually labeled and fed into a shallow convolutional neural network (CNN) for target classification along with a small number of real sparse SAR images. Since the current definition of the number of small samples is still vague and inconsistent, this paper defines few-shot as less than 20 per class. Experimental results based on MSTAR under standard operating conditions (SOC) and extended operating conditions (EOC) show that the reconstructed simulated SAR dataset makes up for the insufficient information from limited real data. Compared with other typical deep learning methods based on limited samples, our method is able to achieve higher accuracy especially under the conditions of few shots.

SAR Image Classification Using Gated Channel Attention Based Convolutional Neural Network

Algorithms combining CNN (Convolutional Neural Network) and super-pixel based smoothing have been proposed in recent years for Synthetic Aperture Radar (SAR) image classification. However, the smoothing may lead to the damage of details. To solve this problem the feature fusion strategy is utilized, and a novel adaptive fusion module named Gated Channel Attention (GCA) is designed in this paper. In this module, the relevance between channels is embedded into the conventional gated attention module to emphasize the variation in contribution on classification results between channels of feature-maps, which is not well considered by the conventional gated attention module. A GCA-CNN network is then constructed for SAR image classification. In this network, feature-maps corresponding to the original image and the smoothed image are extracted, respectively, by feature-extraction layers and adaptively fused. The fused features are used to obtain the results. Classification can be performed by the GCA-CNN in an end-to-end way. By the adaptive feature fusion in GCA-CNN, the smoothing of misclassification and the detail keeping can be realized at the same time. Experiments have been performed on one elaborately designed synthetic image and three real world SAR images. The superiority of the GCA-CNN is demonstrated by comparing with the conventional algorithms and the relative state-of-the-art algorithms.

Blind Super-Resolution for SAR Images with Speckle Noise Based on Deep Learning Probabilistic Degradation Model and SAR Priors

As an active microwave coherent imaging technology, synthetic aperture radar (SAR) images suffer from severe speckle noise and low-resolution problems due to the limitations of the imaging system, which cause difficulties in image interpretation and target detection. However, the existing SAR super-resolution (SR) methods usually reconstruct the images by a determined degradation model and hardly consider multiplicative speckle noise, meanwhile, most SR models are trained with synthetic datasets in which the low-resolution (LR) images are down-sampled from their high-resolution (HR) counterparts. These constraints cause a serious domain gap between the synthetic and real SAR images. To solve the above problems, this paper proposes an unsupervised blind SR method for SAR images by introducing SAR priors in a cycle-GAN framework. First, a learnable probabilistic degradation model combined with SAR noise priors was presented to satisfy various SAR images produced from different platforms. Then, a degradation model and a SR model in a unified cycle-GAN framework were trained simultaneously to learn the intrinsic relationship between HR–LR domains. The model was trained with real LR and HR SAR images instead of synthetic paired images to conquer the domain gap. Finally, experimental results on both synthetic and real SAR images demonstrated the high performance of the proposed method in terms of image quality and visual perception. Additionally, we found the proposed SR method demonstrates the tremendous potential for target detection tasks by reducing missed detection and false alarms significantly.

Building profile reconstruction based on double subspace TomoSAR

ABSTRACT Synthetic aperture radar (SAR) tomography (TomoSAR) is an attractive technology for the three-dimensional (3D) reconstruction of buildings through multiple coherent SAR images. This technique can overcome the limitations of interferometry in multi-scatterer resolution and play a significant role in the analysis of urban areas. The robustness of the many advanced nonparametric spectral estimation approaches is founded on a high number of observations. In this work, we evaluate the abilities of five nonparametric spectral estimation methods, including double subspace (DS), maximum entropy (ME), minimum norm (MN), Capon, and beamforming (BF), to separate scattering contributions with a small number of SAR images in the tomographic reconstruction of urban environments. DS is a high-resolution estimator based on the existence of non-uniform noise, which assesses the noise subspace via eigendecomposition of some appropriately designed matrix without comprehending the noise covariance matrix. The study results indicate that DS can distinguish the scatterers in one azimuth-range resolution and satisfactorily dissuade the appearance of unpleasant side lobes. To investigate the effectiveness of DS and compare that with other expressed methods, both simulated and real SAR datasets were acquired from TerraSAR-X strip-map images of the city of Tehran, Iran. The experimental results on real SAR images demonstrate that the DS can remarkably enhance building reconstruction in the urban environment. The estimated height of the scatters utilizing the DS technique is similar to the data obtained from ground observations.

What Catch Your Attention in SAR Images: Saliency Detection Based on Soft-Superpixel Lacunarity Cue

In existing superpixel-wise saliency detection algorithms, superpixel generation often is an isolated preprocessing step. The performance of saliency maps is determined by the accuracy of superpixels to a certain extent. However, it is still a challenge to develop a stable superpixel generation method. In this article, we attempt to incorporate the superpixel generation and saliency calculation steps into an end-to-end trainable deep network. First, we employ a recently proposed differentiable superpixel generation method to over-segment the synthetic aperture radar (SAR) images, which outputs the possibility that the pixels assigned to neighbor superpixels (soft superpixel). In saliency calculation part, as one of our main contributions, we propose a differentiable and computationally simple saliency model, i.e., lacunarity cue. It is inspired by the fact that generally the backscattering intensity of regions of interest (ROIs) in SAR images irregularly fluctuates, while the areas with consistent pixels are often ignored as the clusters. We improve the pixelwise box differential dimension algorithm to measure the irregularity of scattering points in a superpixel. The superpixel generation and saliency calculation can be implemented under a unified deep network. Hence, the shapes of the superpixels can be iteratively adjusted according to the saliency maps until the ROIs are correctly detected. Experiments on real SAR images with different sizes and scenes show that the saliency maps can effectively highlight the target areas, thus outperforming the state-of-the-art saliency detection models.

A Novel Unsupervised Evaluation Metric Based on Heterogeneity Features for SAR Image Segmentation

The segmentation of Synthetic aperture radar (SAR) images is vital and fundamental in SAR image processing, so evaluating segmentation results without ground truth (GT) is an essential part in segmentation algorithms comparison, parameters selection, and optimization. In this study, we first extracted the heterogeneous features (HF) of SAR images to adequately describe the SAR image targets, which were extracted by the proposed intensity feature extractor (IFEE) based on edge-hold and two fruitful methods. Then we proposed a novel and effective unsupervised evaluation (UE) metric <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">G</i> to evaluate the SAR image segmentation results, which metric based on HF using the global intra-segment homogeneity ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GHO</i> ), global inter-segment heterogeneity ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GHE</i> ), and edge validity index ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EVI</i> ) as local segmentation measures. The effectiveness of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GHO</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GHE</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EVI,</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">G</i> was revealed by visual interpretation as qualitative analysis and supervised evaluation (SE) as quantitative analysis. In experiments, four segmentation algorithms are used to segment plenty of synthetic and real SAR images as the evaluation objects, and four widely used metrics are utilized for comparison. The results show the effectiveness, and superiority of the proposed metric. Moreover, the mean correlation between the proposed UE metric and the SE metric is more than 0.67 and 0.99, which indicates that the proposed metric helps in choosing parameters of segmentation algorithms without GT.