High-resolution Polarimetric Synthetic Aperture Radar Research Articles

BSDSNet: Dual-Stream Feature Extraction Network Based on Segment Anything Model for Synthetic Aperture Radar Land Cover Classification

Land cover classification using high-resolution Polarimetric Synthetic Aperture Radar (PolSAR) images obtained from satellites is a challenging task. While deep learning algorithms have been extensively studied for PolSAR image land cover classification, the performance is severely constrained due to the scarcity of labeled PolSAR samples and the limited domain acceptance of models. Recently, the emergence of the Segment Anything Model (SAM) based on the vision transformer (VIT) model has brought about a revolution in the study of specific downstream tasks in computer vision. Benefiting from its millions of parameters and extensive training datasets, SAM demonstrates powerful capabilities in extracting semantic information and generalization. To this end, we propose a dual-stream feature extraction network based on SAM, i.e., BSDSNet. We change the image encoder part of SAM to a dual stream, where the ConvNext image encoder is utilized to extract local information and the VIT image encoder is used to extract global information. BSDSNet achieves an in-depth exploration of semantic and spatial information in PolSAR images. Additionally, to facilitate a fine-grained amalgamation of information, the SA-Gate module is employed to integrate local–global information. Compared to previous deep learning models, BSDSNet’s impressive ability to represent features is akin to a versatile receptive field, making it well suited for classifying PolSAR images across various resolutions. Comprehensive evaluations indicate that BSDSNet achieves excellent results in qualitative and quantitative evaluation when performing classification tasks on the AIR-PolSAR-Seg dataset and the WHU-OPT-SAR dataset. Compared to the suboptimal results, our method improves the Kappa metric by 3.68% and 0.44% on the AIR-PolSAR-Seg dataset and the WHU-OPT-SAR dataset, respectively.

Polarimetric Synthetic Aperture Radar Speckle Filter Based on Joint Similarity Measurement Criterion

Polarimetric Synthetic Aperture Radar (PolSAR) data is inherently characterized by speckle noise, which significantly deteriorates certain aspects of the quality of the PolSAR data processing, including the polarimetric decomposition and target interpretation. With the rapid increase in PolSAR resolution, SAR images in complex natural and artificial scenes exhibit non-homogeneous characteristics, which creates an urgent demand for high-resolution PolSAR filters. To address these issues, a new adaptive PolSAR filter based on joint similarity measure criterion (JSMC) is proposed in this paper. Firstly, a scale-adaptive filtering window is established in order to preserve the texture structure based on a multi-directional ratio edge detector. Secondly, the JSMC is proposed in order to accurately select homogeneous pixels; it describes pixel similarity based on both space distance and polarimetric distance. Thirdly, the homogeneous pixels are filtered based on statistical averaging. Finally, the airborne and spaceborne real data experiment results validate the effectiveness of our proposed method. Compared with other filters, the filter proposed in this paper provides a better outcome for PolSAR data in speckle suppression, edge texture, and the preservation of polarimetric properties.

A Statistical Rule of the Stokes Parameters of Pol-SAR for Identifying Flat Surface in PSR

Finding a flat area is critical for choosing the landing site, especially in the permanently shadowed region (PSR) without reference of optical images. The digital elevation model (DEM) data are not enough to describe the complex lunar topography in the landing mission. High-resolution polarimetric synthetic aperture radar (pol-SAR) can acquire the polarized scattering map of lunar surface. In this paper, Chandrayaan-2 pol-SAR images of flat and rocky regions at non-PSR are selected with assistance of optical image. The statistics of the Stokes parameters can tell the difference between flat and rocky surfaces. Enhanced backscattering with a large variance from rough surfaces with rocks can be distinguished from the smooth and flat areas. Large value of the first Stokes parameter I1 with extendedly distributed probability distribution function is related with the scatterings from rough surface and rocks. A statistics rule of the Stokes parameters is established, and is applied to identifying the flat regions in the PSR of the crater Shoemaker.

MP-ResNet: Multipath Residual Network for the Semantic Segmentation of High-Resolution PolSAR Images

There are limited studies on the semantic segmentation of high-resolution Polarimetric Synthetic Aperture Radar (PolSAR) images due to the scarcity of training data and the inference of speckle noises. The Gaofen contest has provided open access of a high-quality PolSAR semantic segmentation dataset. Taking this chance, we propose a Multi-path ResNet (MP-ResNet) architecture for the semantic segmentation of high-resolution PolSAR images. Compared to conventional U-shape encoder-decoder convolutional neural network (CNN) architectures, the MP-ResNet learns semantic context with its parallel multi-scale branches, which greatly enlarges its valid receptive fields and improves the embedding of local discriminative features. In addition, MP-ResNet adopts a multi-level feature fusion design in its decoder to make the best use of the features learned from its different branches. Ablation studies show that the MPResNet has significant advantages over its baseline method (FCN with ResNet34). It also surpasses several classic state-of-the-art methods in terms of overall accuracy (OA), mean F1 and fwIoU, whereas its computational costs are not much increased. This CNN architecture can be used as a baseline method for future studies on the semantic segmentation of PolSAR images. The code is available at: https://github.com/ggsDing/SARSeg.

Low-Resolution Fully Polarimetric SAR and High-Resolution Single-Polarization SAR Image Fusion Network

The data fusion technology aims to aggregate the characteristics of different data and to obtain products with multiple data advantages. To solve the problem of reduced resolution of polarimetric synthetic aperture radar (PolSAR) images due to system limitations, we propose a fully PolSAR images and single-polarization synthetic aperture radar (SinSAR) images fusion network to generate high-resolution PolSAR (HR-PolSAR) images. To take advantage of the polarimetric information of the low-resolution PolSAR (LR-PolSAR) images and the spatial information of the high-resolution single-polarization SAR (HR-SinSAR) images, we propose a fusion framework for joint LR-PolSAR images and HR-SinSAR images and design a cross-attention mechanism to extract features from the joint input data. Besides, based on the physical imaging mechanism, we designed the PolSAR polarimetric loss functions for constrained network training. The experimental results confirm the superiority of the fusion network over traditional algorithms. The average peak signal-to-noise ratio (PSNR) is increased by more than 3.6 dB, and the average mean absolute error (MAE) is reduced to less than 0.07. Experiments on polarimetric decomposition and polarimetric signature show that it maintains polarimetric information well.

Applying Particle Swarm Intelligence in PolSAR Image Clustering

The clustering problem of polarimetric SAR image is an optimization problem with high dimension and large amount of data. Aiming at the problem that the classical unsupervised classification methods for High Resolution Polarimetric SAR images are difficult to find the global optimal solution. The Particle Swarm Optimization (PSO) algorithm was proposed in High Resolution PolSAR images clustering. For the first beginning, the scattering eigenvalues of PolSAR data were used for initial classification, and then followed by the computation of clustering center and initialization of PSO algorithm, finally the particle swarm are introduced in the iterative steps to reduce noise effect and improve the classification results. The performance of this novel method is demonstrated in experiments using L-Band PolSAR image of San Francisco Bay.

Random Region Matting for the High-Resolution PolSAR Image Semantic Segmentation

Polarimetric synthetic aperture radar (PolSAR) imagery can provide more intuitive and detailed SAR polarization information, and it is widely used in the classification and semantic segmentation of remote sensing. To bridge the PolSAR data and application, the 2020 Gaofen Challenge on Automated High-Resolution Earth Observation Image Interpretation provides a set of high-quality PolSAR semantic segmentation dataset. A series of preprocessing methods is first used to analyze the PolSAR images to improve the semantic segmentation performance of the PolSAR imagery. A special polarimetric decomposition method is used to extract the features, and the filter and the data truncation are implemented to enhance local and global information of images. And the random region matting method is proposed to expand the training samples. Finally, the DeepLabV3+ method with the ResNet101-V2 is employed to achieve the semantic segmentation. A variety of comparison experiments verifies the effectiveness of our methods. Simultaneously, compared with the classification methods of other groups in the competition, our methods have obvious advantages in the inference time and semantic segmentation accuracy. The proposed method achieved a frequency weighted intersection over union of 75.29% in the contest.

3-D Characterization of Urban Areas Using High-Resolution Polarimetric SAR Tomographic Techniques and a Minimal Number of Acquisitions

This article addresses the 3-D reconstruction of urban areas using a minimal number of Synthetic Aperture Radar (SAR) acquisitions, that is, a set of three images, characterized by intermediate spatial resolution features. In such extreme conditions, conventional tomographic techniques reveal unadapted to refined 3-D imaging purposes, either due to the resulting intrinsic coarse vertical resolution, or to the low dimensionality of the data set, that prevents any separation of complex mixed scattering patterns. A new high-resolution (HR) tomographic estimator, based on a polarimetric signal subspace fitting criterion, is proposed to overcome these limitations, as this method adapts to the statistical behavior of the backscattered signals using robust metrics. The optimization of the corresponding focusing criterion is led through a new polarimetric alternating projection algorithm, characterized by a low computational cost, which may also be used to optimize the polarimetric the deterministic maximum likelihood criterion. The proposed polarimetric signal subspace fitting technique is shown to outperform the other studied HR techniques over both simulated signals and data acquired by the DLR’s ESAR sensor at L-band over Dresden city, Germany. Finally full-rank polarimetric tomographic estimators are proposed that generalize nonparametric polarimetric estimators, and permit to estimate second-order polarimetric representations in 3-D, instead of unitary rank target vector with their conventional versions. This approach makes it possible to characterize polarimetric scattering mechanisms in 3-D.

Research on an Urban Building Area Extraction Method with High-Resolution PolSAR Imaging Based on Adaptive Neighborhood Selection Neighborhoods for Preserving Embedding

Feature extraction of an urban area is one of the most important directions of polarimetric synthetic aperture radar (PolSAR) applications. A high-resolution PolSAR image has the characteristics of high dimensions and nonlinearity. Therefore, to find intrinsic features for target recognition, a building area extraction method for PolSAR images based on the Adaptive Neighborhoods selection Neighborhood Preserving Embedding (ANSNPE) algorithm is proposed. First, 52 features are extracted by using the Gray level co-occurrence matrix (GLCM) and five polarization decomposition methods. The feature set is divided into 20 dimensions, 36 dimensions, and 52 dimensions. Next, the ANSNPE algorithm is applied to the training samples, and the projection matrix is obtained for the test image to extract the new features. Lastly, the Support Vector machine (SVM) classifier and post processing are used to extract the building area, and the accuracy is evaluated. Comparative experiments are conducted using Radarsat-2, and the results show that the ANSNPE algorithm could effectively extract the building area and that it had a better generalization ability; the projection matrix is obtained using the training data and could be directly applied to the new sample, and the building area extraction accuracy is above 80%. The combination of polarization and texture features provide a wealth of information that is more conducive to the extraction of building areas.

Independent Target Detection of PolSAR Image Joint Polarimetric and Spatial Features Based on Adaptive Convolution Sparse Representation

Target detection is of great significance for polarimetric SAR (PolSAR) image applications, and independent target detection in a large scene can be considered as a sparse problem. In high-resolution PolSAR, the independent targets generally present a cluster of similar pixels, and therefore, the detection principle should include not only the internal characteristics but also the spatial information. This letter proposes an unsupervised adaptive convolution sparse representation (ACSR) method for PolSAR image independent target detection. The proposed method updates the dictionary to realize adaptive spatial information injection into polarimetric features, and then the independent target can be detected through the iteration. Two sets of high-resolution, full PolSAR images of unmanned aerial vehicle synthetic aperture radar (UAVSAR) system are used to validate the performance of the proposed method. The results indicate the potential of the proposed method in target detection of PolSAR image.

Polarimetric SAR Image Terrain Classification

In practical applications of automated terrain classification from high-resolution polarimetric synthetic aperture radar (PolSAR) imagery, different terrain types may inherently contain a high level of internal variability, as when a broadly defined class (e.g., “trees”) contains elements arising from multiple subclasses (pine, oak, and willow). In addition, real-world factors such as the time of year of a collection, the moisture content of the scene, the imaging geometry, and the radar system parameters can all increase the variability observed within each class. Such variability challenges the ability of classifiers to maintain a high level of sensitivity in recognizing diverse elements that are within-class, without sacrificing their selectivity in rejecting out-of-class elements. In an effort to gauge the degree to which classifiers respond robustly in the presence of intraclass variability and generalize to untrained scenes and conditions, we compare the performance of a suite of classifiers across six broad terrain categories from a large set of polarimetric synthetic aperture radar (PolSAR) image sets. The main contributions of this article are as follows: 1) an analysis of the robustness of a variety of current state-of-the art classification algorithms to intraclass variability found in PolSAR image sets, and 2) the associated PolSAR image and feature data that Sandia is releasing to the research community with this publication. The analysis of the classification algorithms we provide will serve as a benchmark of performance for the future PolSAR terrain classification algorithm research and development enabled by the image sets and data provided. By sharing our analysis and high-resolution fully polarimetric Sandia data with the research community, we enable others to develop and assess a new generation of robust terrain classification algorithms for PolSAR.

SIRV-Based High-Resolution PolSAR Image Speckle Suppression via Dual-Domain Filtering

This paper presents a spherically invariant random vector (SIRV)-based dual-domain filter for polarimetric synthetic aperture radar (PolSAR) speckle suppression. The SIRV model is a simple clutter decomposition model designed for heterogeneous scenes that decomposes the clutter into two independent domains: the texture domain and the polarimetric domain (also known as the speckle domain). The dual-domain filter takes full advantage of the SIRV model to decompose the task of PolSAR speckle suppression into polarimetric domain filtering and texture domain filtering. The polarimetric domain filtering involves estimating a stable state of the normalized coherency matrix with similar samples via fixed-point iteration. For this purpose, the nonlocal patch matching distance measure and the cutoff threshold in the SIRV model case are defined to select similar samples. The texture domain filtering involves reconstructing a sparse texture image without the effect of speckle. For this purpose, patch ordering-based SAR image despeckling via transform-domain filtering (SAR-POTDF), which is based on simultaneous sparse coding (SSC), is applied for the texture filtering. After the dual-domain filtering, a speckle-free PolSAR image can be reconstructed by the product of the texture and the normalized coherency matrix. The effectiveness and robustness of the proposed method was demonstrated by experiments undertaken with CETC-38 X-band, DLR F-SAR S-band, and IECAS C-band high-resolution PolSAR images. The DLR E-SAR L-band medium-resolution data were also used for comparison. The results showed that the dual-domain filtering is effective for the speckle suppression of high-resolution PolSAR images with heterogeneous and detailed scenes.

Unsupervised Hierarchical Land Classification Using Self-Organizing Feature Codebook for Decimeter-Resolution PolSAR

In this paper, we propose a hierarchical polarization feature generation using a self-organizing codebook to realize unsupervised land classification that fully utilizes the detailed polarization information contained in high-resolution polarimetric synthetic aperture radar (PolSAR) data. PolSAR has reached a decimeter-level high resolution. In general, conventional methods lower the resolution of the PolSAR data to 10–20 m in the real-space distance to classify observation regions into land classes such as farm, forest, and town. However, lowering resolution prevents us from discovering new land classes potentially enabled by the resolution enhancement. The hierarchical method we propose here not only classifies observation regions successfully into land classes such as farm, forest, and town that humans can naturally distinguish but also discovers new land subclasses findable only in high-resolution PolSAR data. We explain these two types of our achievements (classification/discovery) through experimental results for Japan Aerospace Exploration Agency’s polarimetric and interferometric airborne SAR-L2 data having decimeter resolution.

High-resolution polarimetric SAR image decomposition of urban areas based on a POA correction method

ABSTRACTPolarimetric decomposition in urban areas is important for monitoring the speed of city expansion and studying its ecological environmental influence. Using fully Polarimetric Synthetic Aperture Radar (PolSAR) is a way to keep large range urban change under observation. PolSAR images allow us to detect the scattering mechanisms using polarimetric target decomposition methods. Much of the research on decomposition methods has stated that the polarization orientation angle (POA) would affect the result of decomposition. The available deorientation methods cannot rotate built-up areas with large POAs. As a result, some urban areas are still misjudged as volume scattering dominant. In this paper, an algorithm for decomposition of high-resolution urban area images based on a POA correction method is proposed. First, for high-resolution images of built-up areas, the POA changes radically pixel by pixel. An approximate assessment of urban areas can be done using POA randomness. Then, in order to search out the true POAs of large dominant POA areas (most built-up regions), the linear approximation method is used to locate POAs that can minimize the cross-polarized term. Thereby, the inaccurate decomposition that occurs by the deviation of POA can be fixed, and the accuracy of the results is improved.

High-Resolution PolSAR Image Interpretation Based on Human Image Cognition Mechanism

Since the all-day and all-weather working characteristics, polarimetric synthetic aperture radar (PolSAR) has been widely used in many application for earth observation. PolSAR system can acquire targets scattering information under different polarization patterns and obtain detailed information, such as physical characteristics, spatial distribution and direction of the targets, etc. However, in high-resolution PolSAR images, the scattering characteristics of targets are very sophisticated, and different parts of the same target may present different scattering mechanisms. As a result, traditional interpretation technologies are sometimes not applicable in a high-resolution PolSAR scenario. Recently, many artificial intelligence algorithms that simulate the scheme of human cognition have been developed and they have great significance for efficient, intelligent, and accurate images interpretation. In this paper, based on human cognition theory and taking into account the characteristics of targets in PolSAR images, an algorithm for the interpretation of various targets in high-resolution PolSAR images is presented. In this paper, a “perception–reasoning–decision” cognition model is built following the cognition process of professional PolSAR image interpreters. In addition, with the help of theories such as multilevel image segmentation, fuzzy logic, neural network, and the context semantic characteristics, the mathematical model is established and implemented. The proposed method is performed on several sets of PolSAR images obtained under different imaging conditions and is validated by comparing with existing classification methods. The results show that the proposed method can effectively improve the identification of targets from PolSAR images.

Iceberg Detection in Open and Ice-Infested Waters Using C-Band Polarimetric Synthetic Aperture Radar

Icebergs can cause a significant threat to shipping, offshore oil and gas production facilities, and subsea pipelines. Synthetic aperture radar (SAR) is a well-established tool for detecting and monitoring sea-ice objects in the often dark and cloud-covered polar regions. However, detection of small icebergs floating in nonhomegeous sea clutter environments is still a challenging task. We propose a new methodology for automatic identification of potential icebergs in high-resolution polarimetric SAR images. The algorithm adopts to various sea-ice conditions and it tackles high iceberg density situations and heterogeneous background conditions in the marginal ice zone. Results from a time series of RADARSAT-2 data containing numerous icebergs broken off from glaciers in Kongsfjorden on Svalbard demonstrate that the approach is viable.

Level Set Segmentation Algorithm for High-Resolution Polarimetric SAR Images Based on a Heterogeneous Clutter Model

In this paper, a heterogeneous clutter model named ${\mathcal L}$ distribution is introduced into the level set method for segmentation of high-resolution polarimetric synthetic aperture radar (PolSAR) images. Level set methods are robust and effective techniques for segmentation. However, traditional level set methods for PolSAR data are based on the complex Wishart distribution, which is not an applicable model to high-resolution PolSAR images and heterogeneous regions such as forest and urban areas. The ${\mathcal L}$ distribution is proved to be a highly flexible model for multilook PolSAR data, which is based on the product model with a generalized-gamma-distributed texture component. The ${\mathcal L}$ -model-based level set segmentation method is assessed using C-band, X-band, and L-band PolSAR data acquired by RADARSAT-2, TerraSAR-X, and ESAR sensors, respectively. Experimental results show that the ${\mathcal L}$ -distribution model has a better capacity for characterizing high-resolution PolSAR data, especially for extremely heterogeneous regions. Compared with the Wishart-model-based and Kummer-U-model-based level set methods and Markov random field based methods, it is observed that the proposed level set algorithm can obtain more precise segmentation results.

Effects of BOW Model With Affinity Propagation and Spatial Pyramid Matching on Polarimetric SAR Image Classification

The bag of visual words (BOW) model has attracted a great deal of attention in high-resolution remote sensing image classification. Related studies based on polarimetric synthetic aperture radar (PolSAR) image classification are also still challenging. This paper presents a new method of PolSAR image classification based on the BOW model and supplemented by an affinity propagation clustering algorithm (AP), as well as a spatial pyramid matching (SPM) technique (AP_SPM_BOW). Its classification effect is analyzed compared to those of BOW models with other clustering algorithms (e.g., K -means, meanshift, Gaussian mixture model), as well as the ML classification method by using two types of PolSAR images with different resolutions collected from EMISAR and RADARSAT-2 sensors. These results demonstrate that the proposed AP_SPM_BOW model is not only a stable and effective tool for PolSAR image classification but also more suitable for the analysis of high-resolution PolSAR images. In addition, the use of polarimetric information in BOW models can effectively improve the classification accuracy.

Multi-Feature Segmentation for High-Resolution Polarimetric SAR Data Based on Fractal Net Evolution Approach

Segmentation techniques play an important role in understanding high-resolution polarimetric synthetic aperture radar (PolSAR) images. PolSAR image segmentation is widely used as a preprocessing step for subsequent classification, scene interpretation and extraction of surface parameters. However, speckle noise and rich spatial features of heterogeneous regions lead to blurred boundaries of high-resolution PolSAR image segmentation. A novel segmentation algorithm is proposed in this study in order to address the problem and to obtain accurate and precise segmentation results. This method integrates statistical features into a fractal net evolution algorithm (FNEA) framework, and incorporates polarimetric features into a simple linear iterative clustering (SLIC) superpixel generation algorithm. First, spectral heterogeneity in the traditional FNEA is substituted by the G0 distribution statistical heterogeneity in order to combine the shape and statistical features of PolSAR data. The statistical heterogeneity between two adjacent image objects is measured using a log likelihood function. Second, a modified SLIC algorithm is utilized to generate compact superpixels as the initial samples for the G0 statistical model, which substitutes the polarimetric distance of the Pauli RGB composition for the CIELAB color distance. The segmentation results were obtained by weighting the G0 statistical feature and the shape features, based on the FNEA framework. The validity and applicability of the proposed method was verified with extensive experiments on simulated data and three real-world high-resolution PolSAR images from airborne multi-look ESAR, spaceborne single-look RADARSAT-2, and multi-look TerraSAR-X data sets. The experimental results indicate that the proposed method obtains more accurate and precise segmentation results than the other methods for high-resolution PolSAR images.

Discovery of Semantic Relationships in PolSAR Images Using Latent Dirichlet Allocation

We propose a multilevel semantics discovery approach for bridging the semantic gap when mining high-resolution polarimetric synthetic aperture radar (PolSAR) remote sensing images. First, an Entropy/Anisotropy/Alpha-Wishart classifier is employed to discover low-level semantics as classes representing the physical scattering properties of targets (e.g., low-entropy/surface scattering/high anisotropy). Then, the images are tiled into patches and each patch is modeled as a bag-of-words, a histogram of the class labels. Next, latent Dirichlet allocation is applied to discover their higher level semantics as a set of topics. Our results demonstrate that topic semantics are close to human semantics used for basic land-cover types (e.g., grassland). Therefore, using the topic description (bag-of-topics) of PolSAR images leads to a narrower semantic gap in image mining. In addition, a visual exploration of the topic descriptions helps to find semantic relationships, which can be used for defining new semantic categories (e.g., mixed land-cover types) and designing rule-based categorization schemes.