Search

Dielectric constant for the earth and planetary surfaces has been estimated using reflection coefficients in the past. A recent trend is to use model-based decomposition for dielectric constant retrieval from polarimetric synthetic aperture radar (polSAR) data. We examine the reported literature in this regard and propose a unique dielectric constant estimation (UDCE) algorithm using three-component decomposition technique. In UDCE, the dielectric constant is obtained directly from one of the elements of the measured coherency matrix in a single step. The dielectric constant estimate from the UDCE is independent of the volume scattering model when single-bounce or double-bounce scattering is dominant. This avoids error propagation from overestimation of volume scattering to the copolarization ratios, and in turn, to the dielectric constant, inherent in reported algorithms that use model-based decomposition. Consequently, a unique solution is obtained. We also demonstrate that the solution from the UDCE is unaffected by using a higher-order model-based decomposition. We evaluate the performance of the proposed UDCE algorithm over three Apollo 12, Apollo 15, and Apollo 17 landing sites on the lunar surface using Chandrayaan- 2 dual-frequency synthetic aperture radar (DFSAR) datasets. An excellent convergence rate for dielectric constant estimation is maintained over all three test sites. Using the proposed UDCE algorithm, the dielectric constant maps are produced for the lunar surface using full polSAR data for the first time. We observe that the generated dielectric constant maps capture all the ground truth features, previously unseen with such clarity. [ABSTRACT FROM AUTHOR]

Influenced by the imaging mechanism, the occurrence of interference clutter in synthetic aperture radar (SAR) renders the identification of false alarms using detectors challenging. Polarimetric SAR has the potential to improve ship detection performance owing to its distinctive polarization characteristics. The present study proposes a dualistic cascade convolutional neural network (DCCNN) algorithm driven by polarization characteristics for ship detection with fully PolSAR data. First, the new characterizations of fully PolSAR data—Optimized SPAN (OSPAN) and 6-D polarization vector (P 6), were mined and defined based on the polarization coherence matrix to introduce more intact information of targets. Then, a backbone feature extraction network with parallel dualistic cascade architecture, basic geometric feature extraction network (BGFENet), and polarization feature enhancement network (PFENet) was specifically constructed, which provided the comprehensive feature representation of targets via feature fusion. Finally, the classification and regression tasks were accomplished in the fully convolutional detection subnetwork relying on extracted multi-scale fusion feature maps, while focusing on target location regression, leading to a decrease in the cost of detection efficiency caused by redundant PFENet. In addition, the corresponding training strategy according to the special architecture of DCCNN was designed to overcome the problem of labeled fully PolSAR data insufficiency and migrate refined geometric knowledge of targets in the single-polarization SAR amplitude image. Experimental results on the established fully polarized SAR dataset show that DCCNN outperforms the competitive CNN-based target detection methods by at least 6.79% in terms of average precision. Moreover, experimental results on the typical large scenes show that DCCNN is at least 0.88% higher than the well-known conventional methods in terms of F 1 score. [ABSTRACT FROM AUTHOR]

This paper addresses the unsupervised classification problem for multilook polarimetric synthetic aperture radar (PolSAR) images via proposing a new spatially variant G p 0 mixture model based on the extended variational inference (SV G p 0 MM-EVI). Specifically, an SV G p 0 MM is constructed by associating data points with their own mixing coefficient vectors rather than sharing the same vector, which provides the additional flexibility to incorporate the spatial context and further effectively deal with the more heterogeneous areas than G p 0 MM. Then, an extended variational inference (EVI) algorithm is derived to facilitate the assignment of the conjugate prior distributions and accurately estimate the underlying parameters, in which two "help" functions are specially designed to solve the intractable variational lower bound. Meanwhile, dual spatial constraints based on the Bartlett distance and the mean template are also explored on polarimetric matrix and the hyperparameter in the posteriori distribution of mixing coefficients, respectively, thus adequately capturing the local correlation from the polarimetric matrix and the geometry perspectives. Furthermore, the learning algorithm with all the closed-form updates is developed and the cluster number could be automatically determined. Five PolSAR data sets (including two labeled data sets), which are obtained by different sensors, are used to verify the effectiveness of the proposed model. Experimental results illustrate that the proposed SV G p 0 MM-EVI is beneficial to classification task of PolSAR images, and achieves higher accuracy (e.g., 95.65% and 97.41% OA values) compared with some widely-used methods (i.e., H/ α -Wishart, Chernoff–Wishart, G p 0 MM, SVWMM and CK-HDRF methods). [ABSTRACT FROM AUTHOR]

Polarimetric synthetic aperture radar (PolSAR) images containing polarimetric, scattering and contextual features are useful radar data for ground surface classification. Appropriate feature extraction and fusion by using a small set of available labeled samples is an important and challenging task. Several transformers with self-attention mechanism have recently achieved great success for PolSAR image classification. While almost all methods just exploit the self-attention features from the PolSAR cube, the feature fusion method proposed in this work, which is called attention based scattering and contextual (ASC) network, utilizes the polarimetric self-attention beside two cross-attention blocks. The cross-attention blocks extract the polarimetric-scattering dependencies and polarimetric-contextual interactions, individually. The proposed ASC network uses three inputs: the PolSAR cube, the scattering feature maps obtained by clustering of the entropy-alpha features, and the segmentation maps obtained by a super-pixel generation algorithm. The features extracted by self- and cross-attention blocks are fused together, and the residual learning improves the feature learning. While transformers and attention-based networks usually need large training sets, the proposed ASC network shows high efficiency with relatively low number of training samples in various real and synthetic PolSAR images. For example, in the Flevoland PolSAR image containing 15 classes acquired by AIRSAR in L-band, with using 100 training samples per class (less than 1% of labeled samples), the ASC network achieves the overall accuracy of 99.51, which is statistically preferred than the self-attention-based network according to the McNemars test. • An attention-based feature fusion network is proposed for polarimetric SAR image classification. • A self-attention block is suggested to highlight the high important polarimetric features. • Methods for unsupervised generation of scattering and contextual feature maps are introduced. • A cross-attention block is proposed to explore the polarimetric-scattering dependencies. • A cross-attention block is proposed to explore the polarimetric-contextual relationships. [ABSTRACT FROM AUTHOR]

The Chinese Gaofen-3 satellite was launched into space in 2016. Based on the corner reflector (CR) evaluation, it was found that the co-pol channel imbalance phase may exceed the maximum tolerance in the distributed standard product, i.e., ± 10°. The ground-deployed CRs are the most precise way to calibrate the phase error. However, the Mongolia calibration site of Gaofen-3 is a temporal calibration field, and the calibration time is limited for the periodic evaluation over the 8-year satellite life. In this paper, to accelerate the Gaofen-3 calibration mission, we propose an improved unitary zero helix (IUZH) framework to fully replace the use of CRs when estimating the co-pol channel imbalance phase by the use of natural media. Compared with our previous works (Shi et al., 2014; Shi et al., 2020), this novel algorithm presents three main contributions to remote sensing society. Firstly, we believe that this natural media calibration technique could replace the use of CRs, to satisfy the radar design specification, and that this natural media calibration technique could be more than just a substitute in an emergency event when there are no CRs deployed in-scene. Secondly, the IUZH framework is less dependent on choosing the particular reference targets, such as isolated point-like objects or Amazon forest. As a result, it will help radar researchers to relax the constraint of selecting references, and will make the non-CR calibration task easier in satellite or airborne cases. Last but not least, the proposed technique simultaneously derives the calibration phase after the sensor has completed a long-strip observation, which will significantly accelerate the Gaofen-3 external calibration mission, at little cost and time-consuming. The experiments with more than 500 images demonstrate that the phase bias of the Gaofen-3 standard product can be robustly calibrated, even when CRs are unavailable in the scene. The HH and VV imbalance phase with the IUZH method is generally found to be less than 4°. The potential phase error of the Gaofen-3 product is thus significantly suppressed, and the calibration accuracy satisfies the 10° design specification. [ABSTRACT FROM AUTHOR]

Abstract Polarimetric calibration is a critical step to suppress the potential system distortion before implementing any applications for polarimetric synthetic aperture radar (PolSAR). Among all the distortion elements, the crosstalk and cross-pol channel imbalance are generally estimated by the use of natural media, and the co-pol channel imbalance is traditionally solved by the use of corner reflectors (CRs). However, the deployment of ground CRs is costly and may even be impossible in some areas. Many bright point targets, such as poles, lamps, and corner points of structures, are commonly found in manmade regions. In particular, if the object orientation is parallel or perpendicular to the radar flight direction, some points will present similar polarimetric responses to trihedral or dihedral CRs. These points, which are referred to here as "CR-like targets", can be treated as a supplement to approximately solve the system distortion elements when CRs are unavailable. In this paper, we propose a novel step-by-step algorithm to determine the CR-like targets and estimate the co-pol channel imbalance phase in uncalibrated PolSAR imagery. Chinese X-band airborne and C-band satellite PolSAR data were used to test the proposed method. Compared with the CR-derived co-pol channel imbalance phase, the solution errors of the CR-like targets were 1.305° and 0.03° for the X- and C-band experiments, respectively. The results of the experiments confirm that the solutions of the CR-like targets are very close to those of ground-deployed CRs, and the proposed method can be considered as an effective way to calibrate PolSAR images when sufficient CR-like point targets are detected in manmade regions. [ABSTRACT FROM AUTHOR]

Highlights • Spatial correlation of objects in PolSAR image is helpful for target detection. • Globally weighted Perturbation Filters with saliency outperform single PNF. • Two initializations of GWPF achieve different performance, but both acceptable. Abstract In this paper, a saliency detection for Polarimetric Synthetic Aperture Radar (PolSAR) images is proposed based on weighted perturbation filters. Auxiliary data is demanded to identify polarimetric vector of targets, for a canonical perturbation filter. Only if the target signature was available and accurate, it would be satisfiable to apply the filter in practice. Besides, not every target can usually be detected by an individual filter, because of variant polarimetric characteristics of targets with respect to different aspects or shapes. To overcome these drawbacks, several perturbation filters are combined in the proposed method. By initializing with different parameters, these filters decompose PolSAR data into their index maps. Then, aiming to find out filters of interest, i.e., ones related to target pixels, we assume that targets to detect are sparse in PolSAR image. Thus, saliency weights are assigned to the filters, based on Jaccard distances of their index maps. Therein, the spatial sparseness between objects and their surrounding derives high weights for corresponding filters. And then, after globally fusion of refined filtering responses with the weights, saliency map is generated for every local pattern in PolSAR image. Finally, the target regions are extracted from this map, by thresholding and morphological operation. Experiments performed on real and simulated PolSAR data verify the performance of this method, in comparison with several common PolSAR detectors. Also, the proposed method finds out most targets in ground truth, without auxiliary polarimetric information provided. [ABSTRACT FROM AUTHOR]

Wetlands provide a wide variety of environmental services globally and detailed wetland inventory maps are always necessary to determine the conservation strategies and effectively monitor these productive ecosystems. During the last two decades, satellite remote sensing data have been extensively used for wetland mapping and monitoring worldwide. Polarimetric Synthetic Aperture Radar (PolSAR) imagery is a complex and multi-dimensional data, which has high potential to discriminate different land cover types. However, despite significant improvements to both information content in PolSAR imagery and advanced classification approaches, wetland classification using PolSAR data may not provide acceptable classification accuracy. This is because classification accuracy using PolSAR imagery strongly depends on the polarimetric features that are incorporated into the classification scheme. In this paper, a novel feature weighting method for PolSAR imagery is proposed to increase the classification accuracy of complex land cover. Specifically, a new coefficient is determined for each element of the coherency matrix by integration of Fisher Linear Discriminant Analysis (FLDA) and physical interpretation of the PolSAR data. The proposed methodology was applied to multi-temporal polarimetric C-band RADARSAT-2 data in the Avalon Peninsula, Deer Lake, and Gros Morne pilot sites in Newfoundland and Labrador, Canada. Different combinations of input features, including original PolSAR features, polarimetric decomposition features, and modified coherency matrix were used to evaluate the capacity of the proposed method for improving the classification accuracy using the Random Forest (RF) algorithm. The results demonstrated that the modified coherency matrix obtained by the proposed method, Van Zyl, and Freeman-Durden decomposition features were the most important features for wetland classification. The fine spatial resolution maps obtained in this study illustrate the distribution of terrestrial and aquatic habitats for the three wetland pilot sites in Newfoundland using the modified coherency matrix and other polarimetric features. The classified maps provide valuable baseline data for effectively monitoring climate and land cover changes, and support further scientific research in this area. [ABSTRACT FROM AUTHOR]

This publication presents a computer method allowing river channels to be segmented based on SAR polarimetric images. Solutions have been proposed which are based on a morphological approach using the watershed segmentation and combining regions by maximising the average contrast. The image processing methods were developed so that their computational complexity is as low as possible, which is of particular importance in analysing high resolution SAR/polarimetric SAR images, where it has a measurable impact on the total segmentation time. What is more, compared to the existing solutions known from the literature review: (1) in the proposed approach, there is no need to execute further steps necessary to eliminate objects (i.e. background components) located outside the river channel from the image as a result of the segmentation carried out, (2) there is no need to sample the entire image and carry out a pixel–wise classification to prepare the segmentation process. If the steps listed in items (1) – (2) are performed, they can, unfortunately, extend the segmentation time. The experiments completed on images acquired from the ALOS PALSAR satellite for different regions of the world have shown a high quality of the segmentations carried out and a high computational efficiency compared to state–of–the art methods. Consequently, the proposed method can be used as a useful tool for monitoring changes in river courses and adopted in expert and intelligent systems used for analysing remote sensing data. [ABSTRACT FROM AUTHOR]

Wetlands are important ecosystems around the world, although they are degraded due both to anthropogenic and natural process. Newfoundland is among the richest Canadian province in terms of different wetland classes. Herbaceous wetlands cover extensive areas of the Avalon Peninsula, which are the habitat of a number of animal and plant species. In this study, a novel hierarchical object-based Random Forest (RF) classification approach is proposed for discriminating between different wetland classes in a sub-region located in the north eastern portion of the Avalon Peninsula. Particularly, multi-polarization and multi-frequency SAR data, including X-band TerraSAR-X single polarized (HH), L-band ALOS-2 dual polarized (HH/HV), and C-band RADARSAT-2 fully polarized images, were applied in different classification levels. First, a SAR backscatter analysis of different land cover types was performed by training data and used in Level-I classification to separate water from non-water classes. This was followed by Level-II classification, wherein the water class was further divided into shallow- and deep-water classes, and the non-water class was partitioned into herbaceous and non-herbaceous classes. In Level-III classification, the herbaceous class was further divided into bog, fen, and marsh classes, while the non-herbaceous class was subsequently partitioned into urban, upland, and swamp classes. In Level-II and -III classifications, different polarimetric decomposition approaches, including Cloude-Pottier, Freeman-Durden, Yamaguchi decompositions, and Kennaugh matrix elements were extracted to aid the RF classifier. The overall accuracy and kappa coefficient were determined in each classification level for evaluating the classification results. The importance of input features was also determined using the variable importance obtained by RF. It was found that the Kennaugh matrix elements, Yamaguchi, and Freeman-Durden decompositions were the most important parameters for wetland classification in this study. Using this new hierarchical RF classification approach, an overall accuracy of up to 94% was obtained for classifying different land cover types in the study area. [ABSTRACT FROM AUTHOR]

This study proposes a new three-component method for timely detection of land cover changes using polarimetric synthetic aperture radar (PolSAR) images. The three components are object-oriented image analysis (OOIA), change vector analysis (CVA), and post-classification comparison (PCC). First, two PolSAR images acquired over the same area at different dates are segmented hierarchically to delineate land parcels (image objects). Then, parcel-based CVA is performed with the coherency matrices of the PolSAR data to detect changed parcels. Finally, PCC based on a parcel-based classification algorithm integrating polarimetric decomposition, decision tree algorithms, and support vector machines is used to determine the type of change for the changed parcels. Compared with conventional PCC based on the widely used Wishart supervised classification, the three-component method achieves much higher accuracy for land cover change detection with PolSAR images. The contribution of each component is evaluated by excluding it from the method. The integration of OOIA in the method greatly reduces the false alarms caused by speckle noise in PolSAR images as well as improves the accuracy of PolSAR image classification. CVA contributes to the method by significantly reducing the effect of the classification errors on the change detection. The use of PCC in the method not only identifies different types of land cover change but also reduces the false alarms introduced by the change in the environment. The three-component method is validated in land development detection, which is important to many developing countries that are confronting a growing problem of unauthorized construction land expansion. The results show that the three-component method is effective in detecting land developments with PolSAR images. [ABSTRACT FROM AUTHOR]

We propose an algorithm for estimating urban density from polarimetric synthetic aperture radar (SAR) images, and compare the urban density patterns of global megacities. SAR images are uniquely able to detect structural information of objects, but they are very sensitive to orientation angle. This issue has been an obstacle to applying SAR images to urban areas. Kajimoto and Susaki (2013b) proposed an algorithm to handle this issue. The effects of polarization orientation angle (POA) are removed by rotating the coherency matrix and then calculating the mean and standard deviation of scattering power by POA domain. The algorithm can estimate urban density from a single fully polarimetric SAR image but has the drawback that the generated urban density maps of multiple images are not comparable with each other because the algorithm generates a relative urban density valid only within the analyzed image. We therefore extend the method by calculating POA-domain statistics from all images of interest so that the generated maps can be compared. Estimated urban densities are assessed on two types of urban density generated from GIS data, building-to-land ratio and floor-area ratio. We demonstrate that the extended method can estimate urban density with reasonable accuracy. Finally, we generate two scattergrams of indices derived from urban density maps of global megacities. An analysis using the scattergrams indicates insightful information about the patterns of urban development. We conclude that the proposed algorithm and the analysis using the obtained results are beneficial to understanding the conditions in megacities. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, the linear discriminative Laplacian eigenmaps (LDLE) dimensionality reduction (DR) algorithm is introduced to C-band polarimetric synthetic aperture radar (PolSAR) agricultural classification. A collection of homogenous areas of the same crop class usually presents physical parameter variation, such as the biomass and soil moisture. Furthermore, the local incidence angle also impacts a lot on the same crop category when the vegetation layer is penetrable with C-band radar. We name this phenomenon as the “observed variation of the same category” (OVSC). The most common PolSAR features, e.g., the Freeman–Durden and Cloude–Pottier decompositions, show an inadequate performance with OVSC. In our research, more than 40 coherent and incoherent PolSAR decomposition models are stacked into the high-dimensionality feature cube to describe the various physical parameters. The LDLE algorithm is then performed on the observed feature cube, with the aim of simultaneously pushing the local samples of the same category closer to each other, as well as maximizing the distance between local samples of different categories in the learnt subspace. Finally, the classification result is obtained by nearest neighbor (NN) or Wishart classification in the reduced feature space. In the simulation experiment, eight crop blocks are picked to generate a test patch from the 1991 Airborne Synthetic Aperture Radar (AIRSAR) C-band fully polarimetric data from of Flevoland test site. Locality preserving projections (LPP) and principal component analysis (PCA) are then utilized to evaluate the DR results of the proposed method. The classification results show that LDLE can distinguish the influence of the physical parameters and achieve a 99% overall accuracy, which is better than LPP (97%), PCA (88%), NN (89%), and Wishart (88%). In the real data experiment, the Chinese Hailaer nationalized farm RadarSat2 PolSAR test set is used, and the classification accuracy is around 94%, which is again better than LPP (90%), PCA (88%), NN (89%), and Wishart (85%). Both experiments suggest that the LDLE algorithm is an effective way of relieving the OVSC phenomenon. [Copyright &y& Elsevier]

Abstract: In this paper, a robust radial basis function (RBF) network based classifier is proposed for polarimetric synthetic aperture radar (SAR) images. The proposed feature extraction process utilizes the covariance matrix elements, the H/α/A decomposition based features combined with the backscattering power (span), and the gray level co-occurrence matrix (GLCM) based texture features, which are projected onto a lower dimensional feature space using principal components analysis. For the classifier training, both conventional backpropagation (BP) and multidimensional particle swarm optimization (MD-PSO) based dynamic clustering are explored. By combining complete polarimetric covariance matrix and eigenvalue decomposition based pixel values with textural information (contrast, correlation, energy, and homogeneity) in the feature set, and employing automated evolutionary RBF classifier for the pattern recognition unit, the overall classification performance is shown to be significantly improved. An experimental study is performed using the fully polarimetric San Francisco Bay and Flevoland data sets acquired by the NASA/Jet Propulsion Laboratory Airborne SAR (AIRSAR) at L-band to evaluate the performance of the proposed classifier. Classification results (in terms of confusion matrix, overall accuracy and classification map) compared with the major state of the art algorithms demonstrate the effectiveness of the proposed RBF network classifier. [Copyright &y& Elsevier]

Finding appropriate class-separating metric and labeling rules is crucial in the construction of image classifiers. In this paper a Divergence-Chebyshev Neighbors Pursuit (DCNP) algorithm is proposed for rapid Polarimetric Synthetic Aperture Radar (PolSAR) image classification. First, an information-theoretic divergence is defined to measure the similarity of polarimetric features between pixels. Then a divergence-Chebyshev distance is defined to reveal the affinity of pixels in both the polarization and spatial domains. Moreover, inspired by human's learning characteristic that the knowledge is learned little by little, the DCNP algorithm is designed to progressively determine the labels of unknown pixels. Some experiments are conducted on several real PolSAR image datasets and the results show that our method can achieve accurate classification with a small number of labeled data, and outperforms its counterparts in terms of several guidelines. • A simple PolSAR image classifier based on KNNC is proposed. • A divergence-chebyshev distance is defined to reveal the affinity of pixels. • A DCNP algorithm is proposed to progressively classify pixels. [ABSTRACT FROM AUTHOR]