Synthetic Aperture Radar Image Interpretation Research Articles

SAR image segmentation with structure tensor based hierarchical student's t-mixture model

Synthetic aperture radar (SAR) plays an important role in Satellite IoT, due to its remarkable capability of all-weather monitoring and information acquisition under complicated conditions. It is well-known that SAR image interpretation usually requires accurate segmentation. However, SAR image segmentation inevitably encounters speckle noise because of the unique imaging mechanism of SAR. In order to address the problem, we proposed SAR images segmentation method by combined a hierarchical Student’s t-mixture model (HSMM) with an anisotropic mean template, which can divide the global SAR image segmentation into several sub-clustering-issues efficiently resolved using classical algorithm. With the aid of a non-linear structure tensor for image contents analysis, the adaptive template can explore more spatial correlations between pixels for the purpose of improving HSMM robustness and segmentation accuracy. Experiments results both synthetic and real SAR images demonstrate that our proposed HSMM is more robust to speckle noise and obtains more accurate segmented images.

Fast texture classification of denoised SAR image patches using GLCM on Spark

Classification of a synthetic aperture radar (SAR) image is an essential process for SAR image analysis and interpretation. Recent advances in imaging technologies have allowed data sizes to grow, and a large number of applications in many areas have been generated. However, analysis of high-resolution SAR images, such as classification, is a time-consuming process and high-speed algorithms are needed. In this study, classification of high-speed denoised SAR image patches by using Apache Spark clustering framework is presented. Spark is preferred due to its powerful open-source cluster-computing framework with fast, easy-to-use, and in-memory analytics. Classification of SAR images is realized on patch level by using the supervised learning algorithms embedded in the Spark machine learning library. The feature vectors used as the classifier input are obtained using gray-level cooccurrence matrix which is chosen to quantitatively evaluate textural parameters and representations. SAR image patches used to construct the feature vectors are first applied to the noise reduction algorithm to obtain a more accurate classification accuracy. Experimental studies were carried out using naive Bayes, decision tree, and random forest algorithms to provide comparative results, and significant accuracies were achieved. The results were also compared with a state-of-the-art deep learning method. TerraSAR-X images of high-resolution real-world SAR images were used as data.

A SAR-to-Optical Image Translation Method Based on Conditional Generation Adversarial Network (cGAN)

Synthetic Aperture Radar (SAR) remote sensing images with all-day and all-weather advantages are attracting ever-increasing attention in various areas. However, the interpretation of SAR images is quite challenging and not adapted to the non-expert people. To improve the interpretation of SAR images, a SAR-to-optical image translation method based on a modified conditional generation adversarial network (cGAN) is proposed. In the proposed method, the characteristics of the SAR and optical images are comprehensively considered. To generate better results, a modified strongly constrained cGAN, with the generator and the discriminator networks, based on structure similarity index measure (SSIM) and L1 norm was constructed. The generator network aims to generate an artificial optical image by SAR, and the discriminator network aims to force the generated image to be close to the real optical image. The proposed method was verified and compared with several state-of-the-art methods, and the experimental results show the superiority of the proposed method.

Discriminative Sketch Topic Model With Structural Constraint for SAR Image Classification

Synthetic aperture radar (SAR) image classification is an important part in the understanding and interpretation of SAR images. Each patch in SAR images has a scene category, but usually contains multiple land-cover classes or latent properties, which can be represented by topics in the probabilistic topic model (PTM). The representation and selection of discriminative features in PTM have a large impact on the classification results. Most of the existing feature learning methods do not make full use of high-level structure feature and the feature correlation within similar images to mine discriminative features. Therefore, this article proposes a discriminative sketch topic model with structural constraint (C-SSTM) for SAR image classification. In the proposed model, each image patch is characterized by structural and texture features. In particular, the sketch structural feature is based on the sketch map to represent the image local structure pattern. Then, the local image manifold information is preserved in terms of structure and texture. In the structural constraint, the texture and structure of each image patch are combined to learn discriminative latent semantic topics between image patches. Finally, each image patch is quantified by discriminative latent semantic topics instead of low-level representation. The experimental results tested on synthetic and real SAR images demonstrate that the proposed C-SSTM is able to learn effective structural feature representation from SAR images. Compared with other related approaches, C-SSTM produces competitive classification accuracies with high time efficiency.

A Novel Attention Fully Convolutional Network Method for Synthetic Aperture Radar Image Segmentation

As an important step of synthetic aperture radar image interpretation, synthetic aperture radar image segmentation aims at segmenting an image into different regions in terms of homogeneity. Because of the deficiency of the labeled samples and the existence of speckling noise, synthetic aperture radar image segmentation is a challenging task. We present a new method for synthetic aperture radar image segmentation in this article. Due to the large size of the original synthetic aperture radar image, we first divide the input image into small slices. Then the image slices are input to the attention-based fully convolutional network for obtaining the segmentation results. Finally, the fully connected conditional random field is adopted for improving the segmentation performance of the network. The innovations of our method are as follows: 1) The attention-based fully convolutional network is embedded with the multiscale attention network which is capable of enhancing the extraction of the image features through three strategies, namely, multiscale feature extraction, channel attention extraction, and spatial attention extraction. 2) We design a new loss function for the attention fully convolutional network by combining Lovasz-Softmax and cross-entropy losses. The new loss allows us to simultaneously optimize the intersection over union and the pixel classification accuracy of the segmentation results. The experiments are performed on two airborne synthetic aperture radar image databases. It has been proved that our method is superior to other state-of- the-art image segmentation approaches.

OpenSARUrban: A Sentinel-1 SAR Image Dataset for Urban Interpretation

The Sentinel-1 mission provides a freely accessible opportunity for urban image interpretation based on synthetic aperture radar (SAR) data with a specific resolution, which is of paramount importance for Earth observation. In parallel, with the rapid development of advanced technologies, especially deep learning, we urgently need a large-scale SAR dataset supporting urban image interpretation. This article presents OpenSARUrban : a Sentinel-1 dataset dedicated to the content-related interpretation of urban SAR images, including a well-defined hierarchical annotation scheme, data collection, well-established procedures for dataset compilation and organization as well as properties, visualizations, and applications of this dataset. Particularly, our OpenSARUrban collection provides 33 358 image patches of urban SAR scenes, covering 21 major cities of China, including 10 different target area categories, 4 kinds of data formats, 2 kinds of polarization modes, and owning 5 essential properties: large-scale coverage, diversity, specificity, reliability, and sustainability. These properties guarantee the achievement of several goals for OpenSARUrban . The first one is to support urban target characterization. The second one is to help develop well-applicable and advanced algorithms for Sentinel-1 urban target classification. The third one is to explore content-based image retrieval for these kinds of data. In addition, dataset visualization is implemented from the perspective of manifolds to give an intuitive understanding. Besides a detailed description and visualization of the dataset, we present results of some benchmarking algorithms, demonstrating that this dataset is practical and challenging. Notably, developing algorithms to enhance the classification performance on the whole dataset and considering the data imbalance are especially demanding.

Covariance of Textural Features: A New Feature Descriptor for SAR Image Classification

Synthetic aperture radar (SAR) image land-cover classification is an important research topic in SAR image interpretation. However, speckle, inherent of active coherent imaging systems, diminishes the performances of feature extraction techniques and classifiers. In this article, we firstly present a simple yet powerful feature extraction method based on the covariance matrix of textural feature. In comparison with the original textural feature, the use of second-order statistics makes the proposed covariance descriptor more distinguishable for various land covers and robust to speckle noise. Considering that the covariance descriptors are symmetric positive definite (SPD) matrices which form a Riemannian manifold, we then propose to map the manifold with Log-Euclidean metric in a high dimensional reproducing kernel Hilbert space by means of the positive definite kernel defined on manifolds. As a result, the kernel not only allows us to make full use of algorithms developed for linear spaces but also preserves geometric structure of SPD matrices. Our experiments demonstrate that the proposed method can provide promising results in terms of classification accuracy and label consistency.

Despeckling and Analysing Land Use Land Cover of Synthetic Aperture Radar Image

In Image processing, Synthetic Aperture Radar(SAR) images are inherently affected by speckle noise, which visually degrades the appearance of the images and may severely affect the quality of SAR image interpretation tasks like object detection or target detection, instance segmentation and image analysis. Hence SAR image Despeckling becomes a hot research issue. In this paper, the proposed method Total Variation (TV) Denoising is used to address this issue. It is applied to SAR imagery to decrease the noise. It is a filtering method which works efficiently. The process of decreasing the speckle noise is known as Despeckling. When there is noise in the image the actual data is affected. The actual meaning of noise is unwanted signal. Noise is an undesirable by-product in an image that disturbs the original image. On removal of noise, it results in the noise free SAR image. The Land Use Land Cover (LULC) analysis of a SAR image can be accurate when there is no noise in the SAR image. Therefore the main aim of this paper is to analyze the land use and land cover (LULC) in the despeckled high resolution image.

Registration method for GIS vector road data and SAR image

Road detection of synthetic aperture radar (SAR) image is an important part of SAR image interpretation. Thus, in the influence of sophisticated background and speckle noise, it is hard to directly extract the road information from the SAR image. So a SAR image registration method using existing geographic information system (GIS) road data is proposed to assist road detection for SAR image in this paper. By making full use of road geometry structure information, a road structure support is defined. Then all the positions in the SAR image are searched to find the biggest structure support as the optimum matching centre. The image registration for SAR image and road vector data is finally realised. Experimental results show that the registration accuracy of the proposed method is ∼2 pixels, which can be used for shoreline extraction, target positioning and so on.

Random cross-observation intensity consistency method for large-scale SAR images mosaics: An example of Gaofen-3 SAR images covering China

Large-scale microwave remote sensing research requires multiple synthetic aperture radar (SAR) image mosaics. However, owing to multiple factors, different SAR images frequently exhibit intensity differences. This leads to poor intensity continuity after mosaicking, making SAR image interpretation extremely difficult. The existing method assumes that SAR images have been subject to strict radiometric calibration and mainly focuses on intensity balancing to minimize the intensity difference between adjacent images. This method is not effective when the intensity difference between SAR images is large. In this paper, a novel random cross-observation intensity consistency (RCOIC) algorithm is proposed to eliminate the intensity differences for SAR image mosaics. The proposed algorithm is based on global intensity constraints. First, source images are downsampled. Then, the gain correction coefficients of each image in a study area are accurately estimated through random cross-observation. Further, multi-view local intensity correction of each image in the study area is performed using the overlapping areas of adjacent images, and a low-resolution intensity reference map covering the study area is generated. Finally, the intensities of high-resolution source images are corrected by constructing local independent linear models of images based on the low-resolution intensity reference map. Gaofen-3 satellite images with large ranges and large time differences, covering most of China's land area, were selected to test the effect of the newly developed algorithm. After processing, the overall visual effect was good after mosaicking, the intensity transitions at the edges of the images were smooth, the intensity distributions corresponded better to actual ground objects, and the statistical result of adjacent images overlapping regions was good. This validates the reliability of the algorithm.

Analysis and identification of continuous line target in SAR echo based on sidelobe features

There are a large number of continuous line targets in the synthetic aperture radar (SAR) images of land scenes. The traditional SAR processing, which makes the continuous line targets becomes two point targets in traditional SAR images, cannot reveal the complete features of continuous line targets and cause enormous difficulties for SAR image interpretation. This study analyses the reason why traditional SAR processing cannot embody the edge characteristics of continuous line targets based on the sidelobe features of the SAR echo. The authors propose a new identification method based on the difference of the echo amplitude between continuous line target and point targets, which will help to interpret the continuous line targets in SAR images. Numerical simulation experiment based on the classical strip SAR validates the effectiveness of the proposed method.

How Can Despeckling and Structural Features Benefit to Change Detection on Bitemporal SAR Images?

Change detection on bitemporal synthetic aperture radar (SAR) images is a key branch of SAR image interpretation. However, it is challenging due to speckle and unavoidable registration errors within bitemporal SAR images. A key issue is whether and how despeckling and structural features can improve accuracy. In this paper, we investigate how despeckling and structural features can benefit change detection for SAR images. Several change detection methods were performed on both input images and the corresponding despeckled images, where despeckling was achieved by different methods. The comparisons demonstrate that despeckling methods that preserve the structures can suppress noise in difference images and can improve the accuracy of change detection. We also developed a sparse model to exploit structural features from the difference images while reducing the influence of misalignment between bitemporal SAR images. The comparisons were performed on five datasets of bitemporal SAR images, and the experimental results show that our proposed sparse model outperforms other traditional methods, demonstrating the advantages of change detection.

Ship Segmentation in SAR Images by Improved Nonlocal Active Contour Model.

Synthetic aperture radar (SAR) has been widely used in ocean surveillance. As an important part of shipping management and military applications, ship monitoring is a study hotspot in SAR image interpretation; hence, many researches focus on ship targets. Among these studies, ship segmentation is a basic work, but still remains challenging due to the speckle noise and the complicated backscattering phenomenology in SAR images. To solve the problems, this paper proposes a new method for ship segmentation by nonlocal processing. Firstly, the proposed nonlocal energy describes the nonlocal comparison of patches and optimizes regions with spatially-varying features. Secondly, we rewrite the energy functional by introducing a ratio distance defined with respect to the probability density functions of regions to overcome the influence of the multiplicative noise. Finally, the integral histogram is introduced into the pairwise interactions to fasten the speed of convergence. Several rounds of comparative experiments are implemented on real SAR data with different resolutions and bands. The results demonstrate that the proposed method is robust to the speckle noise and intensity variations and could achieve refined segmentation for ship targets.

Scheme for Characterizing Clutter Statistics in SAR Amplitude Images by Combining Two Parametric Models

Accurate knowledge of the statistical properties of synthetic aperture radar (SAR) amplitude data is essential for the processing and interpretation of SAR images. During the last few decades, parametric models have been extensively investigated for use in characterizing the statistics of SAR amplitude data. Among them, the generalized gamma distribution ( $\text{G}\Gamma \text{D}$ ) has been widely applied in many fields of SAR image processing, as it has been demonstrated to be appropriate for describing the statistical behaviors of both SAR land and sea clutter. However, the $\text{G}\Gamma \text{D}$ is not effective when the third-order sample log-cumulant of SAR data is close to zero as it cannot match the distribution of actual data. In this case, the recently proposed ${\mathcal{ G}}_{\textrm {AO}} $ model shows good performance for fitting the distribution of the actual SAR data. In this paper, we theoretically derive the analytical conditions of the ${\mathcal{ G}}_{\textrm {AO}} $ model applicability. Based on the derivations, a scheme describing the statistical behaviors of SAR pixel amplitudes is given by combining the $\text{G}\Gamma \text{D}$ and the ${\mathcal{ G}}_{\textrm {AO}} $ model. Experiments performed on C-band RADARSAT-2 and L-band ALOS-PALSAR SAR data verify the effectiveness of the proposed scheme. The usefulness of the ${\mathcal{ G}}_{\textrm {AO}} $ model for extremely heterogeneous land clutter, for example, from urban regions, is also demonstrated.

Patch-Sorted Deep Feature Learning for High Resolution SAR Image Classification

Synthetic aperture radar (SAR) image classification is a fundamental process for SAR image understanding and interpretation. The traditional SAR classification methods extract shallow and handcrafted features, which cannot subtly depict the abundant modal information in high resolution SAR image. Inspired by deep learning, an effective feature learning tool, a novel method called patch-sorted deep neural network (PSDNN) to implement unsupervised discriminative feature learning is proposed. First, the randomly selected patches are measured and sorted by the meticulously designed patch-sorted strategy, which adopts instance-based prototypes learning. Then the sorted patches are delivered to a well-designed dual-sparse autoencoder to obtain desired weights in each layer. Convolutional neural network is followed to extract high-level spatial and structural features. At last, the features are fed to a linear support vector machine to generate predicted labels. The experimental results in three broad SAR images of different satellites confirm the effectiveness and generalization of our method. Compared with three traditional feature descriptors and four unsupervised deep feature descriptors, the features learned in PSDNN appear powerful discrimination and the PSDNN achieves desired classification accuracy and a good visual appearance.

Discrimination of Algal-Bloom Using Spaceborne SAR Observations of Great Lakes in China

Although optical remote sensing can intuitively detect algal bloom, it is limited by the weather conditions. Synthetic aperture radar (SAR) is not affected by inadequate weather conditions. According to visual interpretation of SAR images and comparisons of quasi-synchronized optical images, the gathering areas of algal bloom present as “dark regions” on SAR images. It is shown that using SAR to monitor the water surface is workable. However, dark regions may also be caused by other factors, such as low wind speeds. This challenges with SAR monitoring of algal bloom on the water surface. In this study, an improved K-means algorithm, combined with multi-Otsu thresholding algorithm, was proposed to segment the dark regions. After feature analysis and extraction of Sentinel-1A images, an algal bloom recognition model with a support vector machine (SVM) was applied to discriminate the algal bloom dark regions from the low wind dark regions. According the experimental results, the overall accuracy achieved 74.00% in Taihu Lake. Additionally, this method was also validated in Chaohu Lake and Danjiangkou Reservoir. Therefore, it can be concluded that SAR can provide a new technical means for monitoring algal bloom of inland lakes, particularly when it is cloudy and unsuitable for optical remote sensing. To obtain more information about algal bloom, multi-band and multi-polarization SAR images can be considered for future.

SAR Image Classification via Deep Recurrent Encoding Neural Networks

Synthetic aperture radar (SAR) image classification is a fundamental process for SAR image understanding and interpretation. With the advancement of imaging techniques, it permits to produce higher resolution SAR data and extend data amount. Therefore, intelligent algorithms for high-resolution SAR image classification are demanded. Inspired by deep learning technology, an end-to-end classification model from the original SAR image to final classification map is developed to automatically extract features and conduct classification, which is named deep recurrent encoding neural networks (DRENNs). In our proposed framework, a spatial feature learning network based on long–short-term memory (LSTM) is developed to extract contextual dependencies of SAR images, where 2-D image patches are transformed into 1-D sequences and imported into LSTM to learn the latent spatial correlations. After LSTM, nonnegative and Fisher constrained autoencoders (NFCAEs) are proposed to improve the discrimination of features and conduct final classification, where nonnegative constraint and Fisher constraint are developed in each autoencoder to restrict the training of the network. The whole DRENN not only combines the spatial feature learning power of LSTM but also utilizes the discriminative representation ability of our NFCAE to improve the classification performance. The experimental results tested on three SAR images demonstrate that the proposed DRENN is able to learn effective feature representations from SAR images and produce competitive classification accuracies to other related approaches.

Simulation-Based Interpretation and Alignment of High-Resolution Optical and SAR Images

The successful alignment of optical and synthetic aperture radar (SAR) satellite data requires that we account for the effects of sensor-specific geometric distortion, which is a consequence of the different imaging concepts of the sensors. This paper introduces SimGeoI, a simulation framework for the object-related interpretation of optical and SAR images, as a solution to this problem. Using metainformation from the images and a digital surface model as input, the processor follows the steps of scene definition, ray tracing, image generation, geocoding, interpretation layer generation, and image part extraction. Thereby, for the first time, object-related sections of optical and SAR images are automatically identified and extracted in world coordinates under consideration of three-dimensional object shapes. A case study for urban scenes in Munich and London, based on WorldView-2 images and high-resolution TerraSAR-X data, confirms the potential of SimGeoI in the context of a perspective-independent and object-focused analysis of high-resolution satellite data.

Learning from label proportions for SAR image classification

Synthetic aperture radar (SAR) image classification plays a key role in SAR interpretation. Due to the cost and difficulty of truth labeling for SAR images, the newly labeled samples available for image classification are very limited. This paper focuses on defining a new sample labeling method to solve the problem of truth acquisition for training data in SAR image classification. An efficient classification framework for high-resolution SAR images is presented in this paper, which is built on learning from uncertain labels. We use grid labeling for rapid training data acquisition by assigning a label to a group of neighboring pixels at a time. A novel SVM-based learning model is proposed to optimize the uncertain training data within the constraints of label proportions in each group and then predict the label of each sample for the test data based on the optimized training set. This work intends to explore a rapid labeling method called grid labeling for efficient training set definition and apply it to large-scale SAR image classification. The model demonstrates good performance in both accuracy and efficiency for scene interpretation of high-resolution SAR images.

High-Performance Anti-Retransmission Deception Jamming Utilizing Range Direction Multiple Input and Multiple Output (MIMO) Synthetic Aperture Radar (SAR).

Retransmission deception jamming seriously degrades the Synthetic Aperture Radar (SAR) detection efficiency and can mislead SAR image interpretation by forming false targets. In order to suppress retransmission deception jamming, this paper proposes a novel multiple input and multiple output (MIMO) SAR structure range direction MIMO SAR, whose multiple channel antennas are vertical to the azimuth. First, based on the multiple channels of range direction MIMO SAR, the orthogonal frequency division multiplexing (OFDM) linear frequency modulation (LFM) signal was adopted as the transmission signal of each channel, which is defined as a sub-band signal. This sub-band signal corresponds to the transmission channel. Then, all of the sub-band signals are modulated with random initial phases and concurrently transmitted. The signal form is more complex and difficult to intercept. Next, the echoes of the sub-band signal are utilized to synthesize a wide band signal after preprocessing. The proposed method will increase the signal to interference ratio and peak amplitude ratio of the signal to resist retransmission deception jamming. Finally, well-focused SAR imagery is obtained using a conventional imaging method where the retransmission deception jamming strength is degraded and defocused. Simulations demonstrated the effectiveness of the proposed method.