L-band Polarimetric Synthetic Aperture Radar Research Articles

UAVSAR Polarimetric Calibration

Uninhabited aerial vehicle synthetic aperture radar (UAVSAR) is a reconfigurable polarimetric L-band SAR that operates in quad-polarization mode and is specifically designed to acquire airborne repeat-track SAR data for interferometric measurements. In this paper, we present details of the UAVSAR radar performance, the radiometric calibration, and the polarimetric calibration. For the radiometric calibration, we employ an array of trihedral corner reflectors, as well as distributed targets. We show that UAVSAR is a well-calibrated SAR system for polarimetric applications, with absolute radiometric calibration bias better than 1 dB, residual root-mean-square (RMS) errors of ~0.7 dB, and RMS phase errors ~5.3°. For the polarimetric calibration, we have evaluated the methods of Quegan and Ainsworth et al. for crosstalk calibration and find that the method of Quegan gives crosstalk estimates that depend on target type, whereas the method of Ainsworth et al. gives more stable crosstalk estimates. We find that both methods estimate leakage of the copolarizations into the cross-polarizations to be on the order of -30 dB.

A Three-Component Fisher-Based Feature Weighting Method for Supervised PolSAR Image Classification

This letter presents a feature weighting method for polarimetric synthetic aperture radar (PolSAR) image classification. Appropriate feature weighting is essential for obtaining accurate classifications but so far has remained an open research problem. We propose in this letter a supervised three-component feature weighting method based on the Fisher linear discriminant. Fisher linear discriminant method is used to calculate a coefficient for each feature. Then, these coefficients are modified according to a three-component scattering power decomposition model, combining both physical and statistical scattering characteristics to adapt them for the particular scattering mechanisms inherent in PolSAR data and assigned to the coherency matrix to enhance the discriminating ability of the features. Freeman decomposition and Wishart classifier are used to classify the PolSAR image. The effectiveness of the proposed method is demonstrated by experiments NASA/JPL AIRSAR L-band and CSA Radarsat-2 C-band PolSAR images of the San Francisco area.

Polarimetric synthetic aperture radar image classification using fuzzy logic in the H / α -Wishart algorithm

To solve the problem that the H/α-Wishart unsupervised classification algorithm can generate only inflexible clusters due to arbitrarily fixed zone boundaries in the clustering processing, a refined fuzzy logic based classification scheme called the H/α-Wishart fuzzy clustering algorithm is proposed in this paper. A fuzzy membership function was developed for the degree of pixels belonging to each class instead of an arbitrary boundary. To devise a unified fuzzy function, a normalized Wishart distance is proposed during the clustering step in the new algorithm. Then the degree of membership is computed to implement fuzzy clustering. After an iterative procedure, the algorithm yields a classification result. The new classification scheme is applied to two L-band polarimetric synthetic aperture radar (PolSAR) images and an X-band high-resolution PolSAR image of a field in LingShui, Hainan Province, China. Experimental results show that the classification precision of the refined algorithm is greater than that of the H/α-Wishart algorithm and that the refined algorithm performs well in differentiating shadows and water areas.

Land Cover Classification from Polarimetric SAR Data Based on Image Segmentation and Decision Trees

. The aim of this article is to present a new classification scheme for polarimetric synthetic aperture radar (POLSAR) data by integrating image segmentation and decision tree. In the first stage of the scheme, POLSAR data is segmented using the multiresolution algorithm embedded in the Definiens eCognition software into small segments consisting of homogeneous pixels. For each segment, an average coherency matrix is generated to represent the overall polarimetric characteristics. In the second stage, 5 elaborately chosen parameters, including the H, A, α parameters from Cloude and Pottier decomposition and 2 parameters associated with the depolarization effect of radar targets, form the decision attributes to classify the segments. An L-band POLSAR image over Dutch Flevoland acquired by the airborne synthetic aperture radar (AIRSAR) system is used as test data. The proposed scheme achieves 13% higher overall accuracy when compared with the widely adopted Wishart classifier, and comparable or higher accuracy when compared with other segment-based algorithms. The 2 depolarization-related parameters, which have never been used for classification before, are found to have powerful discriminability in classifying different vegetation types. This study also demonstrates that the 5 parameters we have chosen are an ideal combination of POLSAR feature parameters and might be applicable for general classification purposes.Résumé. Le but de cet article est de présenter une nouvelle méthode de classification de données provenant d’un radar polarimétrique à synthèse d’ouverture (POLSAR) en intégrant la segmentation d’image et l’arbre de décision. Dans la première étape du procédé, les données POLSAR sont segmentées en petits segments composés de pixels homogènes en utilisant l’algorithme multirésolution incorporé dans le logiciel eCognition de Definiens. Pour chaque segment, une matrice de cohérence moyenne est générée pour représenter les caractéristiques polarimétriques globales. Dans la deuxième étape, cing paramètres minutieusement choisis, y compris les paramètres H, A, α de la décomposition de Cloude-Pottier et deux paramètres associés à l’effet de dépolarisation des cibles radar, forment les attributs de décision pour classer les segments. Une image POLSAR en bande L du Flevoland néerlandais acquise par le système AIRSAR est utilisée comme données d’évaluation. La méthode proposée atteint une précision globale 13 % plus élevée lorsque comparée au classificateur Wishart largement utilisé et une précision comparable ou plus élevée en comparaison avec d’autres algorithmes basés sur des segments. Les deux paramètres liés à la dépolarisation, qui n’ont jamais été utilisés pour la classification, s’avèrent avoir une haute capacité de discrimination pour classer divers types de végétation. Cette étude démontre également que les cing paramètres que nous avons choisis forment une combinaison idéale de paramètres POLSAR qui peut être applicable à des fins générales de classification.

Detecting the Depth of a Subsurface Brine Layer in Lop Nur Lake Basin Using Polarimetric L-Band SAR

Lop Nur once was a huge lake located in northwestern China. At present, there is no surface water in Lop Nur Lake basin and on SAR images it looks like an “Ear.” The objective of this paper is to retrieve the depth of subsurface brine layer in Lop Nur by copolarized phase difference of surface scattering. Based on field investigation and analysis of sample properties, a two-layer scattering structure was proposed with detailed explanations of scattering mechanisms. The relationship between copolarized phase difference and the brine layer depth in the region of Lop Nur were studied. The copolarized phase difference of surface scattering was extracted by model-based polarimetric decomposition method. A good linear correlation between measured subsurface brine layer depth and copolarized phase difference with R2 reaching 0.82 was found. Furthermore, the subsurface brine layer depth of the entire lake area was analyzed. According to the retrieved maps, some interesting phenomena were found, and several hypotheses about the past water withdrawal process and the environmental evolution had been proposed to theoretically explain these phenomena. Based on the penetration capability of SAR the reconstruction of historical evolution process of Lop Nur will be an interesting topic for future research.

Superpixel Segmentation for Polarimetric SAR Imagery Using Local Iterative Clustering

The simple linear iterative clustering (SLIC) algorithm shows good performance in superpixel generation for optical imagery. However, SLIC can perform poorly when there is too much noise in the image. To solve this problem, we have improved the cluster center initialization step and the postprocessing step, and then introduce the SLIC superpixel segmentation algorithm to the polarimetric synthetic aperture radar (PolSAR) image processing field. Experiments using AirSAR and ESAR L-band PolSAR data show that the improved SLIC algorithm can overcome the effect of speckle noise in PolSAR imagery, and it shows a better performance in detail preservation than the original SLIC algorithm and the normalized cuts superpixel segmentation algorithm.

Unsupervised classification of polarimetric synthetic aperture radar interferometry using polarimetric interferometric similarity parameters and SPAN

In this study, a new unsupervised terrain classification method is proposed for L-band polarimetric synthetic aperture radar (SAR) interferometry (PolInSAR) datasets. It can solve the misclassification problems caused by the inherent scattering characteristics of polarimetric SAR (PolSAR). Compared with PolSAR, PolInSAR is much more sensitive to the position and distribution of orientated scatterers. In this study, the polarimetric interferometric similarity parameter (PISP) is proposed to display this sensitivity. It is defined by three optimised scattering mechanisms obtained from PolInSAR datasets. Therefore, it closely relates to the coherent scattering mechanism, and makes use of the phase and amplitude's information. Combined with the scattering power SPAN, the PISP is used to divide the pixels into initial regions. Then, the optimal coherence set parameters A 1 and A 2 are used to distinguish target types in each region based on the number of independent coherent scattering mechanisms. At last, the clusters are merged based on scattering characteristics. The proposed method not only employs optimum coherence but also uses the information of optimum scattering mechanisms. The effectiveness of the proposed method is demonstrated using German Aerospace Center's (DLR) E-SAR L-band PolInSAR datasets.

Mean-Shift-Based Speckle Filtering of Polarimetric SAR Data

The mean shift algorithm, which uses a moving window and utilizes both spatial and range information contained in an image, is widely employed in digital image filtering and segmentation. However, because of the large dynamic range of synthetic aperture radar (SAR) images, applying the conventional mean shift algorithm directly to SAR image filtering will not produce meaningful results. This paper proposes an adaptive variable asymmetric bandwidth selection approach to be used in a newly derived generalized mean shift algorithm. The proposed mean shift algorithm is very versatile and can be used for SAR and polarimetric SAR (PolSAR) image filtering directly without any preprocessing steps. Monte Carlo-simulated PolSAR data are used to demonstrate the effectiveness of the proposed algorithm in speckle filtering by comparing it with other filters. Experimental Synthetic Aperture Radar (ESAR) L-band and Radarsat-2 C-band PolSAR data are used to evaluate its ability to preserve the polarimetric information of PolSAR data. The effects of initial value estimating and multilook processing on the filtered results are discussed at the end of this paper.

Unsupervised classification of PolInSAR based on improved four-component decomposition

In this letter, a new unsupervised terrain classification method is proposed for L-band polarimetric synthetic aperture radar (SAR) interferometry (PolInSAR). It can solve the misclassification problems caused by the inherent scattering characteristics of polarimetric SAR (PolSAR). First, the polarimetric interferometric similarity parameter (PISP) is proposed from PolInSAR data sets. Then, an improved four-component decomposition method is proposed based on PISP. It can partly solve the volume component overestimation problems. Using the improved decomposition result, the image is divided into three basic categories. The optimal coherence set parameters A1 and A2 are used to define each category based on the number of independent coherent scattering mechanisms. The proposed method not only employs optimum coherences but also uses the information of optimum scattering mechanisms. The effectiveness of the proposed method is demonstrated using PolInSAR data sets of German Aerospace Center’s (DLR) experimental synthetic aperture radar (E-SAR).

Recognition of supraglacial debris in the Tianshan Mountains on polarimetric SAR images

Debris cover is widely present in the ablation areas of mountain glaciers. Such debris plays an important role in glacier mass balance, but it complicates the glacier boundary and area estimation when using remotely sensed images. Automatic recognition of the debris is much more difficult than recognizing bare ice on optical remotely sensed images because the debris has almost the same spectral characteristics of the surrounding terrain. In this paper, the scattering characteristics of the debris on the C- and L-band polarimetric SAR images are presented and analyzed to study the possibility of recognizing debris thickness and extent. The Koxkar glacier, in the central Tianshan Mountain Range, was selected as the study area. It was found that debris is distinct from bare ice in cross polarization (HV) of both the L- and C-bands. The L-band volume scattering component, acquired from target decomposition, increases with debris thickness when the debris is less than 25cm deep, but the volume scattering does not increases if the debris is thicker than 25cm. In contrast, the C-band SAR data does not appear to be sensitive to debris thickness. This difference may be caused by different radar penetration depth of the debris. On the basis of the polarimetric property analysis, two methods were developed to determine the extent of the debris. In the first method, the simple threshold to distinguish debris from non-debris areas is presented. Using this threshold, approximately 90% of the debris can be recognized. In the second method, by combining more polarimetric features and support vector machines, recognition of debris reaches an accuracy greater than 95%.

Polarimetric SAR Image Segmentation Using Statistical Region Merging

The statistical region merging (SRM) algorithm exhibits efficient performance in solving significant noise corruption and does not depend on the data distribution. These advantages make SRM suitable for the segmentation of synthetic aperture radar (SAR) images, which are characterized by speckle noise and different distributions of various data types and spatial resolutions. However, the original SRM algorithm is designed for RGB and gray images characterized by additive noise and having a range of [0, 255]. In this letter, the SRM algorithm is generalized so that it can be applied to images with larger range and multiplicative noise. The original 4-neighborhood models are also generalized into 8-neighborhood models. The effectiveness of the generalized SRM (GSRM) algorithm is demonstrated by AirSAR and ESAR L-band Polarimetric SAR (PolSAR) data. Given that the input data of the GSRM algorithm can be single- or multi-dimensional, the proposed GSRM algorithm can be used for single- and multi-polarized as well as for fully polarimetric SAR data.

Basal Area and Biomass Estimates of Loblolly Pine Stands Using L-band UAVSAR

Fully polarimetric L-band Synthetic Aperture Radar ( SAR ) backscatter was collected using NASA ’s Unmanned Aerial Vehicle ( UAV ) SAR and regressed with in situ measurements of basal area ( BA ) and above ground biomass ( AGB ) of mature loblolly pine stands in North Carolina. Results found HH polarization consistently displayed the lowest correlations where HV and VV exhibited the highest correlations in all groups for both BA and AGB . When plantation stands were analyzed separately (plantation versus natural), correlation improved signifi cantly for both BA (R 2 = 0.65, HV ) and AGB (R 2 = 0.66, VV ). Similarly, results improved when natural stands were analyzed separately resulting in the highest correlation for AGB (R 2 = 0.63, HV and VV ). Data decomposition using the Freeman 3-component model indicated that the relative low correlations were due to the saturation of the L-band backscatter across the majority of the study area.

Oil spill detection from SAR image using SVM based classification

Abstract. In this paper, the potential of fully polarimetric L-band SAR data for detecting sea oil spills is investigated using polarimetric decompositions and texture analysis based on SVM classifier. First, power and magnitude measurements of HH and VV polarization modes and, Pauli, Freeman and Krogager decompositions are computed and applied in SVM classifier. Texture analysis is used for identification using SVM method. The texture features i.e. Mean, Variance, Contrast and Dissimilarity from them are then extracted. Experiments are conducted on full polarimetric SAR data acquired from PALSAR sensor of ALOS satellite on August 25, 2006. An accuracy assessment indicated overall accuracy of 78.92% and 96.46% for the power measurement of the VV polarization and the Krogager decomposition respectively in first step. But by use of texture analysis the results are improved to 96.44% and 96.65% quality for mean of power and magnitude measurements of HH and VV polarizations and the Krogager decomposition. Results show that the Krogager polarimetric decomposition method has the satisfying result for detection of sea oil spill on the sea surface and the texture analysis presents the good results.

Evaluation of Digital Classification of Polarimetric SAR Data for Iron-Mineralized Laterites Mapping in the Amazon Region

This study evaluates the potential of C- and L-band polarimetric SAR data for the discrimination of iron-mineralized laterites in the Brazilian Amazon region. The study area is the N1 plateau located on the northern border of the Carajás Mineral Province, the most important Brazilian mineral province which has numerous mineral deposits, particularly the world’s largest iron deposits. The plateau is covered by low-density savanna-type vegetation (campus rupestres) which contrasts visibly with the dense equatorial forest. The laterites are subdivided into three units: chemical crust, iron-ore duricrust, and hematite, of which only the latter two are of economic interest. Full polarimetric data from the airborne R99B sensor of the SIVAM/CENSIPAM (L-band) system and the RADARSAT-2 satellite (C-band) were evaluated. The study focused on an assessment of distinct schemes for digital classification based on decomposition theory and hybrid approach, which incorporates statistical analysis as input data derived from the target decomposition modeling. The results indicated that the polarimetric classifications presented a poor performance, with global Kappa values below 0.20. The accuracy for the identification of units of economic interest varied from 55% to 89%, albeit with high commission error values. In addition, the results using L-band were considered superior compared to C-band, which suggest that the roughness scale for laterite discrimination in the area is nearer to L than to C-band.

Unsupervised Land Cover/Land Use Classification Using PolSAR Imagery Based on Scattering Similarity

This paper presents a new unsupervised land cover/land use classification scheme using polarimetric synthetic aperture radar (PolSAR) imagery based on polarimetric scattering similarity. Compared with the H/alpha classification scheme based on a dominant “average” scattering mechanism, the proposed scheme has such advantages as the following: (1) The major scattering mechanism represents a target scattering in the low-entropy case; (2) it also represents both the major and minor scattering mechanisms in the medium-entropy case; and (3) all the scattering mechanisms in the high-entropy case can be represented. The major and minor scattering mechanisms have been identified automatically based on the relative magnitude of multiple-scattering similarities. The canonical scattering corresponding to maximum scattering similarity is regarded as the major scattering mechanism. The result obtained using the National Aeronautics and Space Administration/Jet Propulsion Laboratory's AIRSAR L-band PolSAR imagery reveals that the proposed scheme is more effective as compared to the existing models and promises to increase the accuracy of the classification and interpretation.

Impacts of Spatial Variability on Aboveground Biomass Estimation from L-Band Radar in a Temperate Forest

Estimation of forest aboveground biomass (AGB) has become one of the main challenges of remote sensing science for global observation of carbon storage and changes in the past few decades. We examine the impact of plot size at different spatial resolutions, incidence angles, and polarizations on the forest biomass estimation using L-band polarimetric Synthetic Aperture Radar data acquired by NASA’s Unmanned Aerial Vehicle Synthetic Aperture Radar (UAVSAR) airborne system. Field inventory data from 32 1.0 ha plots (AGB < 200 Mg ha−1) in approximately even-aged forests in a temperate to boreal transitional region in the state of Maine were divided into subplots at four different spatial scales (0.0625 ha, 0.25 ha, 0.5 ha, and 1.0 ha) to quantify aboveground biomass variations. The results showed a large variability in aboveground biomass at smaller plot size (0.0625 ha). The variability decreased substantially at larger plot sizes (>0.5 ha), suggesting a stability of field-estimated biomass at scales of about 1.0 ha. UAVSAR backscatter was linked to the field estimates of aboveground biomass to develop parametric equations based on polarized returns to accurately map biomass over the entire radar image. Radar backscatter values at all three polarizations (HH, VV, HV) were positively correlated with field aboveground biomass at all four spatial scales, with the highest correlation at the 1.0 ha scale. Among polarizations, the cross-polarized HV had the highest sensitivity to field estimated aboveground biomass (R2 = 0.68). Algorithms were developed that combined three radar backscatter polarizations (HH, HV, and VV) to estimate aboveground biomass at the four spatial scales. The predicted aboveground biomass from these algorithms resulted in decreasing estimation error as the pixel size increased, with the best results at the 1 ha scale with an R2 of 0.67 (p < 0.0001), and an overall RMSE of 44 Mg·ha−1. For AGB < 150 Mg·ha−1, the error reduced to 23 Mg·ha−1 (±15%), suggesting an improved AGB prediction below the L-band sensitivity range to biomass. Results also showed larger bias in aboveground biomass estimation from radar at smaller scales that improved at larger spatial scales of 1.0 ha with underestimation of −3.62 Mg·ha−1 over the entire biomass range.

Building Parameters Extraction From Spaceborne PolInSAR Image Using a Built-Up Area Scattering Model

Polarimetric interferometric synthetic aperture radar (SAR) is more advantageous than either polarimetric or interferometric SAR alone in extracting information from vertical object structure. Generally, for spaceborne SAR, its resolution is low and several targets may be contained in the same resolution cell. Based on the characteristics of spaceborne SAR, a built-up area scattering model (BASM), considering buildings and vegetation in the same resolution cell, is proposed to extract building parameters using L-band polarimetric interferometric SAR data. For single-baseline polarimetric interferometric SAR, the scattering component related to the natural targets could be removed from polarimetric coherency matrices and polarimetric interferometric matrix. Then building parameters could be estimated using the BASM after detecting built-up areas. The BASM is validated using ALOS PALSAR L-band polarimetric interferometric SAR data and experimental results indicate that building parameters could be effectively extracted by the proposed BASM.

Assessment of Scattering Mechanism of Polarimetric SAR Signal From Mountainous Forest Areas

This study aims to understand the effects of the topographic slope variation on the vegetation backscattering characteristics of polarimetric imaging radar. Most of the previous studies in modeling microwave backscattering signatures of vegetated area have been carried out over relatively flat areas. In order to model vegetation scattering mechanisms of mountainous forests, this paper presents an improvement of the radiative transfer model that accounts for the tilted scattering surface beneath a forest canopy. An L-band polarimetric synthetic aperture radar (SAR) data set acquired by the NASA/Jet Propulsion Laboratory Airborne SAR system was used to test and to assess the vegetation scattering model for a sloping forest area. Experimental result shows the proposed model to be adequate for evaluating slope-induced changes in microwave scattering mechanisms.

OIL DETECTION IN A COASTAL MARSH WITH POLARIMETRIC SAR

Abstract. The NASA UAVSAR was deployed June 2010 to support Deep Water Horizon oil spill response activities specifically, oil detection and characterization, oil extent mapping in wetlands, coastal resource impact detection, and ecosystem recovery. The UAVSAR platform demonstrated enhanced capability to act rapidly and provide targeted mapping response. Our research focused on the effectiveness of high spatial resolution and fully polarimetric L-band Synthetic Aperture Radar (PolSAR) for mapping oil in wetlands, specifically within Barataria Bay in eastern coastal Louisiana. Barataria Bay contained a numerous site observations confirming spatially extensive shoreline oil impacts, multiple oil spill UAVSAR collections, and a near anniversary 2009 collection. PolSAR oil detection relied on decomposition and subsequent classifications of the single look complex (SLC) calibrated radar cross sections representing the complex elements of the scattering matrix. Initial analyses results found that shoreline marsh structural damage as well as oil on marsh plants and sediments without canopy structural damage were exhibited as anomalous features on post-spill SLC scenes but were not evident on the pre-spill SLC scene collected in 2009. Pre-spill and post-spill Freeman-Durden (FD) and Cloude-Pottier (CP) decompositions and the Wishart classifications seeded with the FD and CP classes (Wishart-FD) also highlighted these nearshore features as a change in dominate scatter from pre-spill to post-spill. SLC analyses also indicated penetration of oil ladened waters into interior marshes well past the immediate shorelines; however, these post-spill SLC analyses results could not be validated due to the lack of observational data and possible flooding in the pre-spill SLC scene.

Four-Component Model-Based Decomposition of Polarimetric SAR Data for Special Ground Objects

A four-component model-based decomposition for polarimetric synthetic aperture radar (SAR) images is proposed to deal with the ground objects with orientation angles around 45°. In the previous decompositions, these special targets are mixed with the vegetated areas. With the deficiency of the previous decompositions analyzed, the ambiguity between two scattering mechanisms is clarified. A rotated Fresnel dihedral reflection model is introduced in the proposed algorithm, to model the scattering characteristics of these special targets. The nonnegative eigenvalue decomposition is applied to the remainder coherency matrix to prevent negative powers of the decomposed scattering mechanisms. Another advantage of the proposed decomposition is that it makes use of all the information provided by the coherency matrix, which remains unachieved in the previous model-based decompositions. Experimental Synthetic Aperture Radar (E-SAR) L-band polarimetric SAR data acquired over Oberpfaffenhofen, Germany, are analyzed in this letter. Experimental results indicate that the special ground objects have acquired correct scattering mechanisms, which verifies the effectiveness of the proposed method.