Speckle suppression in synthetic aperture radar (SAR) images is a key processing step which continues to be a research topic. A wide variety of methods, using either spatially-based approaches or transform-based strategies, have been developed and have shown to provide outstanding results. However, recent advances in deep learning techniques and their application to SAR image despeckling have been demonstrated to offer state-of-the-art results. Unfortunately, they have been mostly applied to single-polarimetric images. The extension of a deep learning-based approach for speckle removal to polarimetric SAR (PolSAR) images is complicated because of the complex nature of the covariance matrices measured for every image pixel, the properties of which must be preserved during filtering. In this work, we propose a complete framework to remove speckle in PolSAR images using a convolutional neural network (CNN). The methodology includes a reversible transformation of the original complex covariance matrix to obtain a set of real-valued intensity bands which are fed to the neural network. In addition, the proposed method includes a change detection strategy to avoid the neural network to learn erroneous features in areas affected by temporal changes, so that the network only learns the underlying speckle component present in the data. The method is implemented and tested with dual-polarimetric images acquired by Sentinel-1. The different experiments carried out in this work show that the proposed methodology provides good results in both speckle reduction and resolution preservation, similar to or even better than the ones obtained with state-of-the-art methods. Also importantly, it is shown that the neural network is not generating artifacts or introducing bias in the filtered images, making them suitable for further polarimetric processing and exploitation.

The demonstrated efficacy of interferometric synthetic aperture radar (InSAR) techniques has spurred the development of innovative SAR satellite missions like BIOMASS and NiSAR, poised for extensive application in forest monitoring. Nevertheless, prevailing methodologies for retrieving forest variables, including forest height and above-ground biomass, encounter substantial limitations. Traditionally, successful forest mapping necessitates a non-zero spatial perpendicular baseline, full polarimetry, and a relatively small (close-to-zero) temporal baseline. This study presents a novel approach for extracting forest biophysical variables by modeling the temporal decorrelation of repeat-pass InSAR coherence. We explore a hypothesis regarding the potential relationship between the temporal decorrelation of InSAR coherence and forest variables, such as tree height and aboveground biomass. This hypothesis is tested across diverse test sites in Finland, Canada, and Germany. Our findings suggest a viable means of extracting forestry information by quantifying the temporal decorrelation of C-Band InSAR coherence. We establish a clear connection between the temporal decay rate and crucial forest variables, such as forest above-ground biomass and tree height

Acquiring high-resolution and accurate forest height is essential for estimating terrestrial carbon storage and detecting changes. Since 2010, TanDEM-X has obtained an unprecedented global interferometric SAR dataset and has been widely used to estimate forest height. However, its accuracy is limited due to insufficient observational information. Since September 2018, NASA’s Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) has acquired global discrete surface elevation and forest canopy height, providing an excellent opportunity to improve the performance of TanDEM-X forest height estimation. Based on our previous research, this paper proposes some methods to estimate highresolution and large-scale forest height using these two missions combined with optical remote sensing data, and estimates forest height in Spain. As validated against LiDAR data, the RMSE of the estimated optimal average forest height ranges from 1.83 m to 3.70 m, at the resolution of onehectare forest stands.

In this paper, we explore the utilisation of the complex coherence between HH and VV channels as a tool for crop monitoring. The analysis includes the examination of time series of polarimetric data collected from both airborne and space-borne platforms at L and C band. Results suggest that the L band exhibits a higher sensitivity to crop changes than the C-band in both amplitude and phase of the coherence between the co-polar channels. Moreover, incidence angle is identified as a factor influencing the values of the complex coherence, and hence its sensitivity to crop features.

Data obtained during a ground-based SAR experiment and an associated field campaign have been exploited to study the rate and sources of decorrelation in an agricultural test site in the conditions of observation of a geosynchronous SAR. It was found that the scene is less affected by temporal decorrelation when the primary image is acquired during night time or early morning. Additionally, a periodic oscillation on a sub-daily scale was observed when creating coherence time series with increasing temporal baseline. Two factors which strongly contribute to these oscillations are the daily cycles of soil moisture and evapotranspiration.

TanDEM-X InSAR data has exhibited commendable performance in forest height inversion, with the semi-empirical SINC model (SeEm-SINC) proving its robustness in flat regions. However, the relationship between InSAR coherence and forest height become unreliable in forest scenes of mountainous terrain. Significant overestimation or underestimation of forest height occurs when using coherence only in areas with positive or negative slope, and the bias is strongly correlated with the range slope. Building upon the SeEm-SINC model, we investigate the relationship between slope and the bias involved in forest height inversion. Additionally, we present a forest height inversion approach utilizing TanDEM-X InSAR ascending and descending data. Experimental results demonstrate that this method effectively mitigates estimation deviations caused by slope in forested mountainous areas.

In this work we explore the potential of time series of high-resolution radar imagery from Capella Space to enable crop monitoring in areas characterised by very small agricultural fields. Thanks to the spatial resolution provided by the Capella Space SAR images, adjacent fields are well separated in the analysis and, consequently, can be properly monitored and classified. The high-resolution feature complements the all-weather and sun independent operation of radar, hence being an excellent asset for this application.

The usage of Sentinel-1 Single Look Complex (SLC) and its derived interferometric products was limited due to the complexity of the workflows and computational resources required to access, download, process, store, and exploit them. We propose SAR2Cube as a framework based on open source software capable to efficiently store Sentinel-1 Single Look Complex (SLC) images into data cubes and produce on-the-fly Interferometric Synthetic Aperture Radar (InSAR) products, also providing a benchmark with the official European Space Agency (ESA) SNAP processing toolbox.

Forest height is an indispensable parameter for natural resource investigations. Spaceborne Interferometric SAR (InSAR) has the sensitivity to measure forest height, especially TanDEM-X, which provides high-quality interferometric coherence without the effects of atmospheric delay and temporal decorrelation. In this paper, we seriously considered the limited penetrability of TanDEM-X InSAR, and proposed a two-step machine learning (ML) method to estimate large-scale forest height by combining TanDEM-X InSAR data, ICESat-2 data, and Landsat 8 data. The forest scattering phase center (SPC) height of InSAR is estimated by the first ML, and on this basis, the relationship between the SPC height and forest height is established by the second ML. We validated the effectiveness of proposed method in a Spanish Mediterranean climate forest region with airborne LiDAR data. As validated against LiDAR data, the accuracy of the estimated SPC height ranges from 2.07 m to 2.49 m, and the root mean square error (RMSE) of the average forest height ranges from 1.79 m to 2.93 m, at the resolution of four-hectare forest stands.

Monitoring rice growth from the space has aroused the interests of researchers from all over the world. Previous researches about rice phenology focused on dividing paddy rice into different phenological intervals, without knowing the specific date when a certain phenological stage starts. In this paper, different characteristics of the start of three important phenological stages are analyzed. A set of C-band SAR images acquired by SENTINEL-1 has been used to retrieve the start of three key phenological stages (leaf development, stem elongation and inflorescence emergence) of paddy rice. A dynamic time warping method was applied for the alignment of SAR time series curves, which are constructed using backscattering coefficients of different polarimetric channels (VH, VV and VH/VV ratio). Results show that all three kinds of time series curves show great potential in detecting start of leaf development of rice and VH/VV curves perform better in recognizing start of inflorescence emergence.

Time series of repeat-pass interferometric coherence measured from Sentinel-1 images (C-band) and TerraSAR-X/TanDEM-X and PAZ (X-band) have been evaluated for crop monitoring and crop classification. Temporal decorrelation at X-band has been studied in detail thanks to the availability of multiple temporal baselines (4 days, 7 days, 11 days, 15 days, etc.). Different responses are found for different crop classes and also depending on the growth condition of the plants. Based on such a sensitivity, crop classification using series of C- and X-band coherence as input features has been also evaluated, showing the contribution of interferometry in this domain.

The TanDEM-X InSAR have shown great potential for estimation of forest height with high-resolution over large-scale. The feasibility of retrieving forest height is verified in subset area of different forest scenarios, indicating that InSAR coherence can be a reliable variable for inverting forest height without using interferometric phase. Therefore, for the Spanish forest with relatively short canopy height over Mediterranean climate, TanDEM-X InSAR data was used in this paper to map a large-scale forest height. The experimental results show that the proposed mapping method of forest height in large-scale has the advantages of promising accuracy and simplicity.

This study proposes, for the first time, an approach to remove thermal noise from the wave coherency matrix,$\mathrm {C_{2}}$, estimated from single-look complex dual-polarization Interferometric Wide Swath mode Sentinel-1 synthetic aperture radar data. The approach is straightforward; it exploits the ThermalNoiseRemoval module, provided by the European Space Agency (ESA) in its Sentinel Application Platform (SNAP) software, to remove thermal noise from the channel intensities. Then, noise correction on the complex data is applied, in order to estimate the noise-free$\mathrm {C_{2}}$matrix. As a further novelty, the proposed approach can be implemented in SNAP, through the use of a processing graph that is here provided. The method is applied on a dense time series of Sentinel-1 data, collected on an agricultural area located near Seville, Spain. The impact of thermal noise on the estimation of the eigendecomposition parameters of$\mathrm {C_{2}}$, i.e., entropy ($H_{2}$), average alpha angle ($\overline {\alpha _{2}}$), and anisotropy ($A_{2}$), is assessed for different land-cover types, namely river, rice, forest, and urban areas. Monte Carlo simulations are implemented to assess the performance of the proposed approach in estimating$H_{2}$,$\overline {\alpha _{2}}$, and$A_{2}$. Results show that the proposed noise removal method improves the estimation of these parameters for the considered land-cover classes.

The contribution of Polarimetric SAR Interferometry (PolInSAR) observables to crop-type classification is investigated in this letter. The focus is set on characteristic parameters of the Coherence Region (CoRe), i.e. the representation in the polar plot of the PolInSAR data. For this purpose, time series of dual-pol HH-VV single-pass TanDEM-X bistatic data acquired over an agricultural area in Spain are exploited. In the experiment, up to 13 different crop types are evaluated. Crop classification is performed by means of the well-known Random Forest algorithm. The retrieved accuracy metrics highlight the potential of the evaluated PolInSAR descriptors for this application. Some PolInSAR features have proven to be enough representative of the scene, such as the Trace Coherence, which yields a classification accuracy of 75% and 87% at pixel and field level, respectively, on its own. Using all the PolInSAR parameters jointly as input features, classification reaches around 90% and 94% accuracy at pixel and field level, respectively. However, there are some PolInSAR feature subsets, e.g. the coherence measured at the Pauli channels or the foci of the ellipse which represents the CoRe, which yield accuracy levels very close to these maxima. These results demonstrate the suitability of the PolInSAR parameters for crop-type classification. Results are further improved when both polarimetric and PolInSAR features are combined, reaching 94% and 96% accuracy at pixel and field level, respectively.

In this study, we advance a new family of model-based decompositions adapted for dual-pol synthetic aperture radar data. These are formulated using the Stokes vector formalism, coupled to mappings from full quad-pol decomposition theory. A generalized model-based decomposition is developed, which allows separation of an arbitrary Stokes vector into partially polarized and polarized wave components. We employ the widely used random dipole cloud as a volume model but, in general, nondipole options can be used. The cross-polarized phase$\delta $, and the$\alpha $angle, which is a function of the ratio between wave components, measure the transformation of polarization state on reflection. We apply the decomposition to dual-pol data provided by Sentinel-1 (S1) covering different scenarios, such as agricultural, forest, urban, and glacial land-ice. We show that the polarized term of received polarization state is not usually the same as the transmitted one, and can therefore be used for key applications, e.g., classification and geo-physical parameter estimation. We show that, for vegetated terrain, depolarization is not the only influencing factor to S1 backscattered intensities and, in the case of vertical crops (e.g., rice), this allows the crop orientation effects to be decoupled from volume scattering in the canopy. We demonstrate that coherent dual-pol systems show strong phase signatures over glaciers, where the polarized contribution significantly affects the backscattered state, resulting in elliptical polarization on receive. This is a key result for S1, for which dual-pol phase analysis coupled to dense time series offer great opportunities for land-ice monitoring.

This article investigates, for the first time, the inclusion of the operator Trace Coherence (TrCoh) in polarimetric and interferometric synthetic aperture radar (SAR) methodologies for the estimation of biophysical parameters of vegetation. A modified inversion algorithm based on the well-known Random Volume over Ground (RVoG) model, which employs the TrCoh, is described and evaluated. In this regard, a different set of coherence extrema is used as input for the retrieval stage. In addition, the proposed methodology improves the inversion algorithm by employing analytical solutions rather than approximations. Validation is carried out exploiting single-pass HH-VV bistatic TanDEM-X data, together with reference data acquired over a paddy rice area in Spain. The added value of the TrCoh and the convenience of the use of analytical solutions are assessed by comparing with the conventional polarimetric SAR interferometry (PolInSAR) algorithm. Results demonstrate that the modified proposed methodology is computationally more effective than current methods on this dataset. For the same scene, the steps required for inversion are computed in 6 min with the conventional method, while it only takes 6 s with the proposed approach. Moreover, vegetation height estimates exhibit a higher accuracy with the proposed method in all fields under evaluation. The root-mean-squared error reached with the modified method improves by 7 cm with respect to the conventional algorithm.

Polarimetric persistent scatterer interferometry (PolPSI) takes advantage of polarimetric optimization algorithms that enhance interferograms’ phase quality by adequately combining the available polarization channels (e.g., HH, VV, HV, and VH) into an improved one. Amplitude dispersion ($D_{A}$) is one of the commonly used phase quality metrics for this optimization. The resolution of the images is supposed to have an impact on the performance of$D_{A}$-based PolPSI in terms of both pixel density and quality. In this research, this impact is investigated. Specifically, 30 quad-pol RADARSAT-2 images over Barcelona with a resolution around 5 m in both range and azimuth are employed to generate additional data sets with degraded resolutions, ranging from 7.5 to 20 m. The results confirm that, in all cases, the ability of$D_{A}$to select high-quality pixels, i.e., persistent scatterers, decreases when the spatial resolution worsens because the loss of resolution increases the number of scatterers present in a resolution cell. In addition, it would be expected that the performance of the polarimetric optimization of$D_{A}$would tend to decrease when the spatial resolution worsens. However, for all employed resolutions, the polarimetric optimization improves the density and quality of PSs with respect to that of any single polarimetric channel. Moreover, this improvement is more noticeable, in relative terms, as the image resolution degrades.

Polarimetric persistent scatterer interferometry (PolPSI) takes advantage of polarimetric optimization algorithms that enhance interferograms' phase quality by adequately combining the available polarization channels (e.g., HH, VV, HV and VH) into an improved one. Amplitude dispersion ($D_{A}$) is one of the commonly used phase quality metrics for this optimization. The resolution of the images is supposed to have an impact on the performance of$D_{A}$based PolPSI in terms of both pixels density and quality. In this research, this impact is investigated. The results confirm that in all cases the ability of the$D_{A}$to select high-quality pixels, i.e. Persistent Scatterers, decreases when the spatial resolution worsens because the loss of resolution increases the number of scatterers present in a resolution cell. However, for all employed resolutions (ranging from around 5 m to 20 m) the polarimetric optimization improves the density and quality of PSs with respect to that of any single polarimetric channel. Moreover, this improvement is more noticeable, in relative terms, as the image resolution degrades.

The first results of a project aimed at estimating forest height over the entire Spain by means of TanDEM-X data are shown and discussed in this work. Four test sites representative of the Spanish forests are introduced, as well as the data used for validation of results. Results obtained over one of the test sites (in Teruel province) are presented here. Among the challenges found in the project, the influence of slope in mountain areas and the relatively short height of the forest canopy make this work distinctive to previous projects employing TanDEM-X data for estimation of heights in tropical, boreal and temperate regions.

In this study, we propose a new vegetation index (DpRVI) for dual polarimetric synthetic aperture radar (SAR) data. The evaluation of this new index is performed with a particular attention towards the preparation of the NASA-ISRO SAR (NISAR) L-band system science objective. The proposed vegetation index is derived for two dual-pol (HH-HV and VV-VH) modes obtained through a simulation from L-band full-pol UAVSAR data. Time-series simulated NISAR data are obtained from the UAVSAR data acquired during the SMAPVEX12 campaign over the CAL/VAL test site in Winnipeg (Canada), to assess the proposed vegetation index. The temporal trend of DpRVI follows the growth stages of canola with a promising correlation of DpRVI with several biophysical variables. Correlation analysis indicates that DpRVI derived for VV-VH mode correlates better with the canola biophysical parameters than the HH-HV mode.

This paper compares single-pol SAR interferometric and polarimetric SAR interferometric strategies for the estimation of agricultural crop parameters using bistatic TanDEM-X data. An experiment is conducted with data acquired over rice fields in Spain during the Science Phase of this mission. Results show that PolInSAR-based approach and a method based on the coherence amplitude provide then most accurate height estimates. Results derived from single-pol approaches depend on the varying signal penetration of HH and VV channels at different growth phases. Inversion schemes based on the interferometric phase and the complex coherence are also analyzed and provide good results in some cases.

This work presents a novel methodology to extract and analyse multi-temporal polarimetric SAR (PolSAR) information from a stack of co-registered images. The method is based on the analysis of PolSAR changes between every image with respect to the rest of images in the stack. The changes are organized in a matrix form to encode the polarimetric evolution of a target. The change matrix is then used to visually understand a target evolution and its SAR response based on the evolution of scattering mechanisms due to the target physical evolution. Additionally, we design and test an image classification algorithm in a supervised learning fashion by using typical change matrices as training data. The methodology is tested exploiting C-band quad-pol RADARSAT-2 data with special interest on agricultural fields such as rice in Seville, South-West of Spain and in the Indian Head in Canada as part of the Agrisar 2009 campaign.

A set of six spotlight TerraSAR-X images acquired at HH and VV polarizations in 2009 over an agricultural site in Germany are employed to evaluate the potential contribution of polarimetric features derived from this copolar mode to crop-type mapping. Results show that the inclusion of the correlation between copolar channels in the set of input features of the classifier consistently improves the classification performance with respect to the use of only backscattering coefficients. An increase around 8%-10% in overall accuracy, depending on the experiment setup, is achieved. Both user and producer accuracies are improved for all crop types, being the most noticeable contribution for barley, oat, and sugar beet. Different sets of input features, as well as classification and evaluation strategies, are tested in order to assess the robustness of this contribution.

This article evaluates the effect of the double-bounce (DB) decorrelation term that appears in single-pass bistatic acquisitions, as in the TanDEM-X system, on the inversion of scene parameters by means of polarimetric SAR interferometry (PolInSAR). The retrieval of all scene parameters involved in the Random Volume over Ground (RVoG) model (i.e., ground topography, vegetation height, extinction, and ground-to-volume ratios) is affected by this term when the radar response from the ground is dominated by the DB. The estimation error in all these parameters is analyzed by means of simulations over a wide range of system configurations and scene variables for both agricultural crops and forest scenarios. Simulations demonstrate that the inclusion of the DB term, which complicates the inversion algorithm, is necessary for the angles of incidence shallower than 30° to achieve an estimation error below 10% in vegetation height and to avoid a significant underestimation in the ground-to-volume ratios. At steep incidences, this decorrelation term does not affect the estimation of vegetation height and ground-to-volume ratios. Regarding the extinction, this parameter is intrinsically not well estimated, since most retrieved values are close to the initial guesses employed for the optimization algorithm, regardless of the use or not of the DB decorrelation term. Finally, these findings are compared with the experimental results from the TanDEM-X data acquired over the rice fields in Spain for the available system parameters (baseline and incidence angle) of the acquired data set.

This work deals with the classification of wheat phenology by regressing the synthetic aperture radar (SAR) backscatter coefficients (VV, VH) to vegetation water content (VWC) and plant area index (PAI) through a representation learning network. The representation network architecture consists of a pair (VV, VH) of two regression layers (VWC, PAI) which finally converge to a classification (crop phenology) layer. The study was conducted with the Sentinel-1 C-band SAR data acquired during the SMAPVEX16 campaign in Manitoba, Canada. Using this framework, the wheat phenology was classified to an accuracy of 86.67%. However, in comparison, the classification accuracy reduced by ~ 20% while using only the backscatter coefficients of (VV, VH) polarization channels. The results obtained from this study justifies the potential of using a representation learning scheme for crop phenology classification with SAR data.

The role of global sensitivity analysis (GSA) is to quantify and rank the most influential features for biophysical variable estimation. In this letter, an approximation model, called high-dimensional model representation (HDMR), is utilized to develop a regression method in conjunction with a GSA in the context of determining key input drivers in the estimation of crop biophysical variables from polarimetric synthetic aperture radar data. A multitemporal Radarsat-2 data set is used for the retrieval of three biophysical variables of barley: leaf area index, normalized difference vegetation index, and Biologische Bundesanstalt, Bundessortenamt und CHemische Industrie stage. The HDMR technique is first adopted to estimate a regression model with all available polarimetric features for each biophysical parameter, and sensitivity indices of each feature are then derived to explain the original space with a smaller number of features in which a final regression model is established. To evaluate the applicability of this methodology, root-mean square and coefficient of determination were performed under different amounts of samples. Results highlight that HDMR can be used effectively in biophysical variable estimation for not only reducing computational cost but also for providing a robust regression.

Tracking crop's biophysical parameters using temporal Pol-SAR (Polarimetric Synthetic Aperture Radar Data) data is an active research topic in precision agriculture due to the sensitivity of PolSAR acquisition to canopy's physical and geometrical structure. Reconstruction of polarimetric features from collection of SAR data is computationally expensive, and more important, the inter-features correlations cause decreased performance in regression based biophysical parameter estimation. With the scope of operational crop monitoring, this study provides key variables to drive Leaf Area Index (LAI) from polarimetric data based on global sensitivity analysis (GSA) addressing the ranking of the most influential features. We applied variance-based GSA for temporal fully-polarimetric RadarSAt-2 images acquired through the cultivation period of two crops; canola and barley. Among 20 polarimetric features, anisotropy and correlation magnitude between co-polar channels were found to be the most influential polarimetric features for canola and barley, respectively.

Differential SAR Interferometry (DInSAR) is a powerful remote sensing technique employed to monitor surface displacements, such as ground subsidence or strong deformations caused by geological activity. The quality of the interferometric phase between two combined SAR images is essential for the estimation of the surface deformation. Multi-pIe decorrelation factors may degrade the quality of the measurements and, then, the development of filtering methods for noise suppression is mandatory. In this work, we propose a new strategy to improve noise reduction while preserving the original phase structure. The new method consists in an iterative filter in which noise reduction is achieved progressively. The original phase is filtered with adaptive kernels based on Chebyshev interpolation functions. The filter is especially useful for DInSAR geophysical applications, such as earthquakes or volcanic eruptions monitoring. The performance of the proposed method has been tested with both simulated data and recently acquired Sentinel-1 SAR data which mapped the August 2016 Central Italy earthquake.

In bistatic acquisitions the presence of a double-bounce contribution at the ground affects the interferometric coherence with a decorrelation factor which is usually overlooked in studies employing polarimetric SAR interferometry. The standard acquisition mode of TanDEM-X is bistatic, so the influence of this contribution in the estimation of scene parameters (ground topography and vegetation height) is studied here. The analysis is carried out both with simulations and real data acquired over rice fields during the science phase of TanDEM-X. Results show that the error in height and topography is small when incidence angle is below 30 degrees, but may become noticeable for shallower incidences.

SAR polarimetry (PolSAR) can play an important role in monitoring agricultural fields both in terms of improving detection of specific plants conditions and providing physical information regarding the change. Such information can be used to help retrieving the phenological stage and eventually identifying stress conditions. In this work, a new change detection based on PolSAR data is first tested over time series of images acquired over agricultural fields. The methodology is based on the use of the normalised difference between covariance matrices acquired at two different instants. A diagonalisation of such matrix allows identifying the scattering mechanisms that suffer the largest change. The methodology is tested exploiting C-band quad-polarimetric RADARSAT-2 data over rice fields in Sevilla, South-West of Spain.

InSAR coherence is a promising parameter for land-cover classification and mapping. The ESA SEOM SInCohMap project is devised to test and analyze multi-temporal InSAR coherence potentialities exploiting dense multitemporal data from the Sentinel-1 constellation. In the framework of the project, this paper shows the first classification results using machine learning algorithms over a two-year period of InSAR coherence data. The evaluation is performed on the test site of Doñana (Seville, Southwestern Spain), mainly an agricultural area where different land covers can be identified. Classification results exploiting InSAR coherence shows accuracies around 80 % for this site.

The interferometric coherence is a measure of the correlation between two SAR images and constitutes a commonly used estimator of the phase quality. Its estimation requires a spatial average within a 2-D window, usually named as multilook. The multilook processing allows reducing noise at the expenses of a resolution loss. In this letter, we analyze the influence of the multilook size while applying a polarimetric optimization of the coherence. The same optimization algorithm has been carried out with different multilook sizes and also with the nonlocal SAR filter filter, which has the advantage of preserving the original resolution of the interferogram. Our experiments have been carried out with a single pair of quad-polarimetric RADARSAT-2 images mapping the Mount Etna's volcanic eruption of May 2008. Results obtained with this particular data set show that the coherence is increased notably with respect to conventional channels when small multilook sizes are employed, especially over low-vegetated areas. Conversely, very decorrelated areas benefit from larger multilook sizes but do not exhibit an additional improvement with the polarimetric optimization.

This letter proposes a modified general polarimetric model-based decomposition method which includes a simplified Neumann volume scattering model (SNVSM). This is useful to avoid a known limitation in one of the state-of-the-art general model-based decomposition methods (i.e., Chen's method), which considers only four possible discrete volume scattering models. Two types of SNVSM, assuming horizontal or vertical dipoles, are derived from the Neumann volume scattering model. The resulting volume coherency matrix exhibits a continuous range of volume scattering models. In addition, this volume model covers both random and nonrandom volume cases, which are distinguished by a randomness parameter. Monte Carlo simulations are used to test this approach. The proposed method with SNVSM overall improves the final accuracy of estimated parameters in comparison with the original approach and shows consistency with another existing generalized volume scattering model (GVSM). In addition, results from two fully polarimetric C- and L-band AIRSAR images over San Francisco region show that the proposed method produces reasonably physical results and outperforms the traditional Y4R method. Finally, the differences obtained between SNVSM and GVSM in two building areas show the potential advantage of SNVSM in identifying more types of volume scenes than that of GVSM.

Phase quality is a key element in the analysis of the deformation of the Earth's surface carried out with differential synthetic aperture radar interferometry. Various decorrelation sources may degrade the surface deformation estimates, and thus, phase filters are needed for this kind of application. The well-known Goldstein filter is the most widely used due to its simple implementation and computational efficiency. In the past years, improved filters have been proposed, which are based on this filter but introduce variations in the data processing. The effectiveness of these filters mostly depends on the size of the filtering window, the weight of the smoothed spectrum, and the kernel used to filter the spectrum. In this paper, we evaluate the performance of four of these filters and present a new method that outperforms all of them. The proposed filter is based on an iterative method in which the original phase is denoised progressively with adaptive filtering windows of different sizes. The effectiveness of the filter is controlled by the interferometric coherence, a direct indicator of the phase quality. Moreover, we introduce some modifications regarding the processing of the power spectrum. Specifically, we propose to smooth the original phase using a new filter which is based on a Chebyshev interpolation scheme. The performance of the new filter has been tested on both simulated and real interferograms, acquired by RADARSAT-2 and the Uninhabited Aerial Vehicle Synthetic Aperture Radar, which mapped two different geological events that caused surface deformation.

While polarimetric persistent scatterer InSAR (PSI) is an effective technique for increasing the number and quality of selected PS pixels, existing methods are suboptimal; a polarimetric channel combination is selected for each pixel based either on amplitude, which works well only for high-amplitude scatterers such as man-made structures, or on the assumption that pixels in a surrounding window all have the same scattering mechanism. In this paper, we present a new polarimetric PSI method in which we use a phase-based criterion to select the optimal channel for each pixel, which can work well even in nonurban environments. This algorithm is based on polarimetric optimization of temporal coherence, as defined in the Stanford Method for PS (StaMPS), to identify the scatterers with stable phase characteristics. We form all possible copolar and cross-polar interferograms from the available polarimetric channels and find the optimum coefficients for each pixel using defined search spaces to optimize the temporal coherence. We apply our algorithm, PolStaMPS, to an area in the Tehran basin that is covered primarily by vegetation. Our results confirm that the algorithm substantially improves on StaMPS performance, increasing the number of PS pixels by 48%, 80%, and 82% with respect to HH+VV, VV, and HH channels, respectively, and increasing the signal-to-noise ratio of selected pixels.

Space-borne interferometric SAR has advanced significantly in the last decades, with many successful Earth monitoring applications. The key point of this success lies in the fact that interferometric SAR supplies unprecedented phase and amplitude information characterizing target's physical parameters. This paper presents the role of interferometric SAR, specifically considering the bistatic mission TanDEM-X, for shallow lake water level estimation by using sediment storage.

Polarimetric SAR interferometry has been recently applied with TanDEM-X data to the retrieval of vegetation height in rice fields, which constitutes the first demonstration of this technique applied to agricultural crops with satellite data [1]. In this work we extend that study by including an analysis of the effect of incidence angle and baseline on the retrieval of height. The study is based on the exploitation of 6 time series of TanDEM-X acquisitions during its science phase: 3 over Sevilla (Spain), with incidence angles of 22, 30 and 39 degrees, and 3 over Ipsala (Turkey), with incidence angles of 30, 36 and 44 degrees.

This paper presents a study of the sensibility of the incoherent (electromagnetic backscattering model) and coherent (DInSAR and PolInSAR inversion) crop height estimation methods of SAR imaging. The methods were compared for paddy-rice crop height monitoring with a TanDEM-X dataset. For this, rice-cultivated agricultural fields located in Northern Turkey were selected. Intensive ground data collection during the cultivation period in 2015 was carried out. The accuracy analysis showed that the requirement of external (vegetation-free) DEM in DInSAR-based crop height estimation decreases its performance compared to the PolInSAR and backscattering inversion methods.

A new methodology to estimate the growth stages of agricultural crops using the time series of polarimetric synthetic aperture radar (PolSAR) images is proposed. The methodology is based on the complex Wishart classifier and both phenological intervals and training areas are identified measuring the distances among polarimetric covariance matrices obtained from the time series of PolSAR data. Consequently, the computation of PolSAR features, which is the main step of state-of-the-art methods, is no longer needed, and the proposed approach can be applied in the same way to any crop type. Experiments undertaken on a dense time series of fully polarimetric C-band RADARSAT-2 images, collected at incidence angles ranging from 23° to 39°, in ascending/descending orbit passes, demonstrate that the proposed methodology can be successfully applied to retrieve the phenological stages of four different crop types. In addition, the effect of combining beams corresponding to different sensor's configurations has been evaluated, showing that it affects the retrieval accuracies. Validation with ground data shows the following: overall accuracy is between 54% and 86%; producer's accuracy (PA) and user's accuracy (UA) range between 21% and 100% and between 33% and 100%, respectively.

The Normalised Difference Vegetation Index (NDVI) is well suited to provide the phenological state of crops, but time-series of NDVI data gathered by satellites are usually too noisy to be used directly. Common approaches apply signal processing steps based on fitting analytical functions and then a few stages can be estimated by defining thresholds along the fitted curves. Dynamic tools constitute an alternative way to exploit these NDVI data with the advantages of providing real-time estimates which are defined in a continuous range of values. In this work, we propose a particle filter method to retrieve phenological states of rice crops using time-series of NDVI satellite observations. This methodology achieves a high determination factor for the complete phenological range. The end of season dates are estimated and compared with different fitting-based methods, providing much better results.

In this study application results are presented derived from multi-parametric SAR observations covering five different thematic domains: forest, agriculture, ocean, urban and cryosphere. In total 21 application products have been selected and described. Their application on different data sets, space- and airborne sensors, was demonstrated and can independently be reproduced by any scientist. The results and algorithms are available soon through Springer.

Dual-pol HHVV images acquired by TerraSAR-X/TanDEM-X during 4 years, with 3 different incidence angles, over rice fields in Spain are analysed and employed to retrieve the phe-nological state of rice crops in a scale of 3 to 5 possible intervals. A decision tree approach is employed for the retrieval algorithm. Results show that the joint processing of all angles provides a good performance, only 2-4% worse than a separate processing. The role of polarimetry is demonstrated by obtaining a total accuracy of 88.6% when coherent dual-pol data are considered, instead of 75.6% for incoherent dual-pol.

C-band RADARSAT-2 quad-Poland X-band COSMO-SkyMed incoherent dual-polarization PingPong SAR data are used to monitor corn fields phenology. The polarimetric analysis carried out at C-band shows a sensitivity of polarimetric observables to corn phenology that allows to interpret the different polarimetric signatures occurring along the cultivation cycle. Unfortunately, at the X-band, a physical interpretation is not properly provided, due to the unavailability of radar acquisitions for the most of the cultivation cycle. This is due to defense requirements which caused conflict acquisitions with the planned COSMO-SkyMed SAR data.

In this work a novel methodology for the estimation of land subsidence is proposed. PSI methods assume an almost linear behavior which may not fit well in particular scenarios, for instance those induced by piezometric level changes. In order to overcome some limitations of the classical methods, in this study we analyze the ability of a dynamic approach in such a context. In particular, the approach is based on the combination of a hydrological model and InSAR PSI data to describe the subsidence as a process of a dynamic system. A simple subsidence model related to piezometric level changes is employed to derive the methodology. The estimation process is conducted by a Kalman Filter. The evaluation over simulated and real data shows the capabilities of this approach.

The coherent nature of the acquisition by TerraSAR-X of both copolar channels (HH and VV) enables the generation of many different polarimetric observables with physical interpretation, as have recently been used for monitoring rice fields. In this letter, the influence of incidence angle upon these polarimetric observables is analyzed by comparing three stacks of images that were acquired simultaneously at different incidence angles (22°, 30°, and 40°) during a whole cultivation campaign. We show that the response of observables related to dominance (entropy, ratios of components) and type of scattering mechanisms (alpha angles) is not greatly influenced by incidence angle at some stages: early and advanced vegetative phases, and maturation. Moreover, the acquisition geometry drives the sensitivity to the presence of the initial stems and tillers, being detected earlier at shallower angles. This analysis is a necessary step before studying potential methodologies for combining different orbits and beams for reducing the time between acquisitions for monitoring purposes.

In this paper, a novel approach for exploiting multitemporal remote sensing data focused on real-time monitoring of agricultural crops is presented. The methodology is defined in a dynamical system context using state-space techniques, which enables the possibility of merging past temporal information with an update for each new acquisition. The dynamic system context allows us to exploit classical tools in this domain to perform the estimation of relevant variables. A general methodology is proposed, and a particular instance is defined in this study based on polarimetric radar data to track the phenological stages of a set of crops. A model generation from empirical data through principal component analysis is presented, and an extended Kalman filter is adapted to perform phenological stage estimation. Results employing quad-pol Radarsat-2 data over three different cereals are analyzed. The potential of this methodology to retrieve vegetation variables in real time is shown.

In this letter, a new approach for crop phenology estimation with remote sensing is presented. The proposed methodology is aimed to exploit tools from a dynamical system context. From a temporal sequence of images, a geometrical model is derived, which allows us to translate this temporal domain into the estimation problem. The evolution model in state space is obtained through dimensional reduction by a principal component analysis, defining the state variables, of the observations. Then, estimation is achieved by combining the generated model with actual samples in an optimal way using a Kalman filter. As a proof of concept, an example with results obtained with this approach over rice fields by exploiting stacks of TerraSAR-X dual polarization images is shown.

A set of ten RADARSAT-2 images acquired in fully polarimetric mode over a test site with rice fields in Seville, Spain, has been analyzed to extract the main features of the C-band radar backscatter as a function of rice phenology. After observing the evolutions versus phenology of different polarimetric observables and explaining their behavior in terms of scattering mechanisms present in the scene, a simple retrieval approach has been proposed. This algorithm is based on three polarimetric observables and provides estimates from a set of four relevant intervals of phenological stages. The validation against ground data, carried out at parcel level for a set of six stands and up to nine dates per stand, provides a 96% rate of coincidence. Moreover, an equivalent compact-pol retrieval algorithm has been also proposed and validated, providing the same performance at parcel level. In all cases, the inversion is carried out by exploiting a single satellite acquisition, without any other auxiliary information.

Persistent scatterer (PS) interferometry (PSI) techniques are designed to measure ground deformations using satellite synthetic aperture radar (SAR) data. They rely on the identification of pixels not severely affected by spatial or temporal decorrelation, which, in general, correspond to pointlike PSs commonly found in urban areas. However, in urban areas, we can find not only PSs but also distributed scatterers (DSs) whose phase information may be exploited for PSI applications. Estimation of DS parameters requires speckle filtering to be applied to the complex SAR data, but conventional speckle filtering approaches tend to mask PS information due to spatial averaging. In the context of single-polarization PSI, adaptive speckle filtering strategies based on the exploitation of amplitude temporal statistics have been proposed, which seek to avoid spatial filtering on nonhomogeneous areas. Given the growing interest on polarimetric PSI techniques, i.e., those using polarimetric diversity to increase performance over conventional single-polarization PSI, in this paper, we propose an adaptive spatial filter driven by polarimetric temporal statistics, rather than single-polarization amplitudes. The proposed approach is able to filter DS while preserving PS information. In addition, a new methodology for the joint processing of PS and DS in the context of PSI is introduced. The technique has been tested for two different urban data sets: 41 dual-polarization TerraSAR-X images of Murcia (Spain) and 31 full-polarization Radarsat-2 images of Barcelona (Spain). Results show an important improvement in terms of number of pixels with valid deformation information, hence denser area coverage.

In previous works, a general framework to exploit polarimetric diversity to optimize the results of persistent scatterers interferometry (PSI) was presented, but tested only with dual-pol data. In this paper, the performance of these algorithms is assessed using fully polarimetric data, acquired by the Radarsat-2 satellite over the urban area of Barcelona, Spain. In addition, two new highly efficient polarimetric optimization methods, mean intensity polarimetric optimization and joint diagonalization-based polarimetric optimization, are introduced and evaluated. Given the variety of dual-pol configurations provided by current polarimetric satellites, such as TerraSAR-X and Radarsat-2, and the upcoming launch of Sentinel-1, ALOS-2, and Radarsat Constellation Mission, a study has been also carried out to determine the best performing dual-pol configurations for polarimetric PSI. Subsidence maps of the area of study are computed for single-pol, dual-pol, and full-pol data, which show the increase in pixel density with valid deformation results as more polarimetric information is made available. In particular, for full-pol data we get an increase of up to 2.5 times more pixels for coherence-based PSI techniques (degraded resolution), and over four times more for amplitude-based approaches (full resolution), in comparison with single-pol data. Both higher density and quality of pixels yield better results in terms of coverage and accuracy.

In modern society, the anthropogenic influences on ecosystems are central points to understand the evolution of our planet. A polarimetric synthetic aperture radar may have a significant contribution in tackling problems concerning land use change, since such data are available with any-weather conditions. Additionally, the discrimination capability can be enhanced by the polarimetric analysis. Recently, an algorithm able to identify targets scattering an electromagnetic wave with any degree of polarization has been developed, which makes use of a vector rearrangement of the elements of the coherency matrix. In the present work, this target detector is modified to perform change detection between two polarimetric acquisitions, for land use monitoring purposes. Regarding the selection of the detector parameters, a physical rationale is followed, developing a new parameterization of the algebraic space where the detector is defined. As it will be illustrated in the following, this space is 6-D complex with restrictions due to the physical feasibility of the vectors. Specifically, a link between the detector parameters and the angle differences of the eigenvector model is obtained. Moreover, a dual polarimetric version of the change detector is developed, in case quad-polarimetric data are not available. With the purpose of testing the methodology, a variety of data sets were exploited: quad-polarimetric airborne data at L-band (E-SAR), quad-polarimetric satellite data at C-band (Radarsat-2), and dual-polarimetric satellite data at X-band (TerraSAR-X). The algorithm results show agreement with the available information about land changes. Moreover, a comparison with a known change detector based on the maximum likelihood ratio is presented, providing improvements in some conditions. The two methodologies differ in the analysis of the total amplitude of the backscattering, where the proposed algorithm does not take this into consideration.

A complete polarimetric study on C-band RADARSAT-2 data over vineyards in Spain has been performed in order to explore the sensitivity to growth stages. This analysis suggests that scattering response of vines can be approximated as a two-component scenario comprising a strong return from soil and a weak depolarized return due to grapes and leaves. A Water Cloud Model parameterization of the cross-polar intensity is proposed as a function of the development of grapes and LAI changes over time.

The use of polarimetric interferometry to monitor vegetation covers, assuming simple models like the RVoG, is influenced by the acquisition mode of the interferometer: single-tx or alternate-tx. Theoretical expressions obtained for the single-tx mode have not been validated with real data yet. In this work we address this validation by employing a new laboratory SAR system and by analyzing TanDEM-X data.

An example of the potential of hybrid-polarity architecture parameters for agricultural monitoring is presented in this letter. Time series of such descriptors have been generated from quad-polarization radar airborne measurements at the L-band acquired during the AgriSAR 2006 campaign. An assessment on the sensitivity of these observables for monitoring the cultivation of winter wheat, maize, and winter rape crops has been addressed. All possible correlations among these polarimetric descriptors and crop coverage, phenological stage, leaf area index, wet biomass, and crop height have been analyzed for the whole growth season. The results of this study reveal that hybrid-polarity systems provide sensitivity in many cases to retrieve both quantitative information and qualitative information from agricultural crops.

A set of three quad-pol images acquired at the L-band in interferometric repeat-pass mode by the German Aerospace Center (DLR) with the Experimental SAR (E-SAR) system, in parallel with the AgriSAR2006 campaign, has been used to provide, for the first time with airborne data, a demonstration of the retrieval of vegetation height from agricultural crops by means of polarimetric SAR interferometry (PolInSAR)-based techniques. Despite the low frequency of the data, hence providing a weak response from the vegetation volume in contrast to the ground, accurate estimates of vegetation height at field level have been obtained over winter rape and maize fields. The same procedure does not yield valid estimates for wheat, barley, and sugar beet fields due to a mismatch with the physical model employed in the inversion and to the specific crop condition at the date of acquisition. These results show the value of the information provided by both interferometry and polarimetry for some agriculture monitoring practices.

This letter is aimed at presenting results confirming the contribution of polarimetry to improve the performance of persistent-scatterer interferometry. The improvement is provided by the identification of more pixels with good phase quality, under criteria commonly employed in this context, after a search in the available polarimetric space. The ground deformation results obtained with a series of 41 dual-polarization images acquired by TerraSAR-X over the metropolitan area of Murcia, Spain, have been used to illustrate this approach.

The feasibility of retrieving the phenological stage of rice fields at a particular date by employing coherent copolar dual-pol X-band radar images acquired by the TerraSAR-X sensor has been investigated in this paper. A set of polarimetric observables that can be derived from this data type has been studied by using a time series of images gathered during the whole cultivation period of rice. Among the analyzed parameters, besides backscattering coefficients and ratios, we have observed clear signatures in the correlation (in magnitude and phase) between channels in both the linear and Pauli bases, as well as in parameters provided by target decomposition techniques, like entropy and alpha from the eigenvector decomposition. A new model-based decomposition providing estimates of a random volume component plus a polarized contribution has been proposed and employed in interpreting the radar response of rice. By exploiting the signatures of these observables in terms of the phenology of rice, a simple approach to estimate the phenological stage from a single pass has been devised. This approach has been tested with the available data acquired over a site in Spain, where rice is cultivated, ensuring ground is flooded for the whole cultivation cycle, and sowing is carried out by randomly spreading the seeds on the flooded ground. Results are in good agreement with the available ground measurements despite some limitations that exist due to the reduced swath coverage of the dual-pol HHVV mode and the high noise floor of the TerraSAR-X system.

This work will present a study about the capabilities of Polln SAR for agriculture applications with TanDEM-X data, since this sensor will acquire for the first time single-pass Polln SAR data from space. The use of single-pass acquisitions is mandatory for such application due to the rapid growth of crops. In addition, X-band provides a good trade-off between the backscattering levels from ground and vegetation volume, as already demonstrated in polarimetric studies with TerraSAR-X imagery. Unfortunately, at the time of compiling this paper there are not TanDEM-X data available yet, since the first acquisitions are planned for June 2011. Therefore, the first results will be presented at the conference.

This work presents a study about the evolution of the radar response of rice fields at X-band as a function of their phenology, by using three time series of coherent HHVV dual-pol X-band radar images acquired by the TerraSAR-X sensor at different incidence angles during a whole cultivation season. After analyzing a wide set of polarimetric observables, with particular sensitivities at different moments along the campaign, an approach to retrieve the phenological stage with a single acquisition is proposed and tested.

Several POLInSAR methodologies are based on the concept of polarimetric and interferometric coherence [1]. Many of them assume the scattering mechanisms observed at the two interferometric acquisitions are the same (i.e. equi-scattering mechanism, ESM). However, if the scatterers change, different approaches must be employed. In order to select the appropriate set of algorithms, a test of target stability should be performed as pre-processing. The authors already devel oped a polarimetric change detector [2], based on perturbation analysis. In the present paper, the change detector and the parameterization introduced in [2] are modified in order to perform the ESM test on dual polarimetric TanDEM-X data. Unfortunately, TanDEM-X data were not available at the mo ment of compiling this paper, so the algorithm was tested on TerraSAR-X HHIVV images.

In this work we review the PSI polarimetric optimization method firstly introduced in [1] and apply it to a larger stack of images, in order to confirm its suitability for the obtainment of accurate deformation maps. The approach achieves a higher density of points with valid deformation information compared to single-pol data, thus confirming the expected improvement. For a more visual insight of the optimization procedure, we also analyze the polarimetric behavior of the scene, identifying the scattering mechanisms that the algorithm selects as optimum or dominant. Results of the analysis agree with the expected for urban areas.

Complex interferometric coherence expressions for the homogeneous (random or oriented) volume-over-ground model are originally derived in a simplified way by neglecting one of the two possible contributions of the ground response: direct return from the ground surface or double-bounce interaction with stems or trunks. The influence of these depends on both system and scene parameters, and in many cases, none of them should be ignored a priori. Therefore, a more general formulation accounting for the simultaneous presence of both ground contributions should be developed. This letter derives this formulation and analyzes the combined effect of both ground returns on the complex coherences. In addition, the implications for a subsequent inversion procedure based on this model are also discussed.

This paper presents an analysis of the performance of TerraSAR-X for subsidence monitoring in urban areas. The city of Murcia has been selected as a test-site due to its high deformation rate and the set of extensometers deployed along the city that provide validation data. The obtained results have been compared with those ob-tained from ERS/ENVISAT data belonging to the same period and validated with the in-situ measurements.

This letter presents a general method for increasing the number of pixel candidates, those selected for processing in advanced differential SAR interferometry, by means of the exploitation of the polarimetric information provided by new satellite sensors. The algorithm is formulated for two different criteria of selection: the average coherence over the stack of interferograms and the amplitude dispersion index of the stack of images. Experimental results obtained with dual-pol images of TerraSAR-X over an urban area have demonstrated the expected improvement. The number of pixel candidates for an arbitrary threshold is 60% higher than that for single-pol data when using the average coherence and three times higher when using the dispersion index. The approach has also been compared to a selection based on a set of conventional channels (the copolar linear channels and the first two Pauli ones), showing a slight improvement for coherence selection and an important one for amplitude dispersion selection.

In this paper, the Freeman-Durden polarimetric decomposition concept is adapted to polarimetric SAR interferometry (PolInSAR) data. The covariance matrix obtained from PolInSAR observations is decomposed into the three scattering mechanisms matrices proposed by Freeman and Durden for polarimetric SAR (PolSAR) data. The objective is to describe each interferometric cross correlation as the sum of the contributions corresponding to direct, double-bounce, and random volume scattering processes. This procedure enables the retrieval not only of the magnitude associated with each mechanism but also of their location along the vertical dimension of the scene. One of the most important features of this algorithm is the potential to isolate more accurately the direct and volume contributions which usually cannot be correctly separated by means of PolSAR measurements. In addition, it is also possible to distinguish between direct scattering responses originated either at ground or produced by upper layers of vegetation. The proposed algorithm has been tested with simulated data from PolSARProSim software, laboratory data from maize and rice samples, and airborne data from a test site with different scenarios.

3-D high-resolution radar images have been produced for a mature rice crop sample by using wide-band fully polari-metric data collected at the EMSL, JRC-Ispra (Italy). These images have been compared with the 1D vertical density profiles produced by the Polarization Coherence Tomography technique on the same rice sample. In order to obtain a 1D vertical reflectivity profile from the 3D SAR images, several slices at certain ground-range positions have been selected and then all the horizontal cross-range contributions have been summed up. As expected, very similar qualitative results are obtained when comparing these profiles with the PCT ones. The ground exhibits the highest response for all three polarization channels up to C-band, and also for the HH channel at X-band. In addition, at X-band the VV channel decreases significantly at ground level but the HV channel remains high. Finally, an important contribution from the upper layers is observed at C-band and more notably at X-band for all three polarization channels.

The objective of this work is to investigate the coherent co-polarized behavior of rice plants during the growing stages and to explore their information content for rice monitoring at high frequencies recently available through new SAR satellite missions. Time series of dual-pol TerraSAR-X images have been acquired during the whole cultivation period over a rice site in Spain. Among different observations, the backscattering coefficients at HH and VV channels and the HH/VV ratio have confirmed to show a temporal variation that has a significant correlation with the development of the plants during the vegetative and reproductive phenological phases. In addition, the information content of the HHVV complex coherence and a dual polarimetric target decomposition is investigated and discussed. All the information layers investigated are contributing to the discrimination of rice fields from other crop types. Finally, interferograms computed with pairs of successive images (11 days separation) have been preliminary tested.

This work presents an improved InSAR phase unwrapping (PU) method based on a combination of a particle filter, a region growing technique and a matrix pencil based local slope estimator. The better performance of this new solution when compared against some representative traditional methods and the previous particle filter PU approaches is justified and illustrated with results obtained from synthetic and real data.

This letter presents a new phase unwrapping algorithm for synthetic aperture radar interferometry which combines a particle filter, a matrix-pencil (MP) local slope estimator, and an optimized region-growing technique. The advantages of the new method rely on the following contributions: The MP estimator provides better resolution to the local slope estimation, the particle filter enables simultaneous unwrapping and filtering withouta prioristatistics constraints, and the implemented region-growing technique adds diversity of unwrapping paths and ensures an optimum solution. Results introduced in this letter illustrate the main aspects of the new approach.

This paper presents a new phase-unwrapping (PU) algorithm for SAR interferometry that makes use of a particle filter (PF) to perform simultaneously noise filtering and PU. The formulation of this technique provides independence from noise statistics and is not constrained by the nonlinearity of the problem. In addition, an enhanced variant of this method combining a PF with artificial-intelligence search strategies and an omnidirectional local phase estimator, based on the mode of the power spectral density, is also presented. Results obtained with synthetic and real data show a significant improvement with respect to other conventional unwrapping algorithms in some situations.

This work presents a phase unwrapping (PU) algorithm for SAR interferometry based on a particle filter (PF). This PU algorithm performs simultaneously noise filtering and phase unwrapping. The formulation of this technique provides independence from noise statistics and is not constrained by the non linearity of the problem. Results show a significant improvement with respect to conventional PU algorithms in some situations.

In this work we have applied the polarization coherence tomography (PCT) to indoor data acquired on a rice sample. Different baselines, frequency bands and polarimetric channels have been used in order to analyze the potential of this technique for retrieving information about the vertical structure of such a specific crop and, additionally, these observations have been also used for studying the validity of the RVoG and OVoG models. Results obtained show that the PCT profiles exhibit no dependence on polarization, which is in agreement with the clustering effect observed on PolInSAR observables, i.e. coherences for different polarimetric channels are located very close one to another. In the dual-baseline case, the best profile reconstruction is obtained for the highest baseline ratio available (0.25deg and 1deg baselines) when the condition number of the inversion problem is improved by filtering out the fifth coefficient of the Legendre-Fourier expansion. Differences on the reconstructed profiles between linear and Pauli basis are negligible. The application of the triple-and four-baseline approaches without filtering out the last terms of the expansion leads to unstable results, whereas the filtered cases reveal that the contribution of a fourth baseline is null.

This letter presents a phase-unwrapping (PU) algorithm for synthetic aperture radar interferometry based on a grid-based filter. The proposed PU algorithm, which is based on state-space techniques, simultaneously performs noise filtering and PU. The formulation of this technique provides independence from noise statistics and is not constrained by the nonlinearity of the problem. Results obtained with synthetic data show a significant improvement with respect to other conventional PU algorithms in some situations.

The formulation of the complex interferometric coherence for the oriented volume over ground model (OVoG) has been recently proposed for the case of dominant double-bounce mechanism from the ground when the interferometer is operated in single-transmit (bistatic) mode. This paper analyzes the two contributions to the total coherence function: the volume and the double-bounce terms. The study is performed by observing the model predictions of the coherence loci for several agricultural crop scenarios and with system parameters corresponding to the TanDEM-X mission. For low vegetation depths (up to 1 m) neither of the two contributions is negligible when the ground-to- volume ratio is low. However, when the backscattering returns from ground and vegetation are similar, i.e. ground-to-volume ratios around 0 dB, the volume decorrelation terms does not affect the coherence so much. As predicted by theory, higher vegetation layers yield an increase of the influence of the volume term and more differences between polarimetric channels.

This work presents a phase unwrapping (PU) algorithm for SAR interferometry based on an approximate grid-based filter (GbF). This PU algorithm, which makes use of state space techniques, performs simultaneously noise filtering and phase unwrapping. The formulation of this technique provides independence from noise statistics and is not constrained by the non-linearity of the problem. Results obtained with synthetic and real data show a significant improvement with respect to conventional PU algorithms in some situations.

An alternative interpolation technique for SAR image coregistration in interferometric processing is formulated and described. The proposed algorithm is based on the 1- D Farrow interpolator and, when combined with an adequate implementation, it involves a smaller computational burden than the conventional method and yields high accuracy. Basically, this technique enables to carry out the coregistration without resorting to a set of functional weights, and it only requires arithmetic operations and 2-D FFTs. This paper includes several results and comparisons that confirm its validity: the proposed technique is combined with two different polynomial interpolation procedures (Taylor and Chebyshev); bounds on its interpolation error in the 1-D and 2-D cases are derived; and, additionally, it is tested numerically with a synthetic image.

An advanced DInSAR technique called Coherent Pixels Technique (CPT) has been used to measure the subsidence existing in Orihuela city (Spain) during the period 1993–2001 due to ground water level fall. The estimated subsidence, with values lower than 7 cm, is highly influenced by soil geotechnical conditions like the deformable soil thickness. In addition, the wells location is an important subsidence factor because they are directly related to a decrease of the piezometric level.

The formulation of the polarimetric interferometric coherences derived by simple homogeneous-volume-over-ground models is revisited for the case of ground returns dominated by the double-bounce terms. The differences between single-transmit (single-tx) and alternate-transmit modes of the interferometer are analyzed by inspecting the positions of the coherences on the complex plane. An additional volume-decorrelation term appears when the interferometer is operated in a single-tx mode, which adds complexity in the potential inversion of the model parameters. This effect is even more noticeable for the cross-polar channels in the oriented case, since the phase of the coherence with an infinite ground-to-volume ratio does not correspond to the topographic phase, as observed for the copolar channels

An efficient polynomial interpolation method is proposed to reduce the computational burden of the coregistration of synthetic aperture radar images in interferometric processing. The method can be viewed as an application of the Farrow interpolator technique and requires a series of 2-D fast Fourier transforms (FFTs). Mainly, it has two advantages relative to the usual coregistration procedure. First, it does not require to compute or store in memory any kernel sample. Second, it involves less floating-point operations than the conventional coregistration, with a clear advantage if the image needs to be oversampled. The efficiency of the method is based on the fact that the complexity of one 2-D FFT is much smaller than the complexity of one spatial convolution. These advantages are demonstrated by both analyzing the interpolation procedure itself and by defining an efficient code implementation.

Application of polarimetric synthetic aperture radar interferometry to the retrieval of geophysical parameters from vegetated scenes is based on simple direct models of such scenes. The first part of this paper presents an analysis of the correspondence between these simple models, namely, the random volume over ground and the oriented volume over ground (OVoG), and experimental data from samples of two agricultural crops (maize and rice) acquired in controlled conditions. Although an overall agreement between model and data is clear, some discrepancies have been found as a consequence of two assumptions in the model formulation: vertical homogeneity of the vegetation volume and absence of multiple scattering effects inside the volume. This paper presents the shape and location of the visible region of the experimental coherences on the complex plane and compares it with the feasible region predicted by the model. This comparison has also pointed out the low sensitivity of the direct model to extinction coefficients. In the second part, two different strategies for a complete inversion (i.e., estimation of all model parameters) of the OVoG model are proposed and compared, using the same data set. The first one is based on a combination of geometrical and numerical approaches (genetic algorithms) and the second one on a dual-baseline configuration. In all cases, ground topography is accurately estimated, with a maximum error of 10 cm. Vegetation height estimates are accurate up to 30 cm, with some bands and baseline configurations providing errors below 15 cm. However, results obtained for the extinction coefficients are not stable with frequency and exhibit high variability.

The hydrological quality of an aquifer system is evaluated by means of two parameters: its capabilities to transmit water (transmissivity, T) and to store water (storage coefficient, S). In this work, a method based on temporal data of the surface subsidence is employed to calculate storage coefficients (S) and transmissivities (T) of the Vega Media of the Segura river aquifer-system. Subsidence data are obtained by means of differential SAR interferometry. The retrieved values of S for all available wells vary from 3.2 times 10-5to 1.9 times 10-3m/m. For the only well where water flow is available, a T value of 0.302 m2/day is estimated. First results show a reasonable agreement between data calculated with this technique and other acquired by means of in situ measurements.

A study of the wide-band polarimetric backscatter of maize plants, measured in laboratory conditions, is presented. The backscatter slant-range profiles in both linear (H-V) and Pauli basis manifest a higher extinction coefficient in the vertical channel due to the dominant vertical orientation of the structure of corn plants. The difference between the horizontal and vertical range profiles as the wave penetrates into the volume is employed to retrieve the differential extinction coefficient. In addition, the polarimetric target decomposition, as proposed by Cloude and Pottier, is used to obtain range profiles of alpha, entropy, and anisotropy. All these results reveal important features to be accounted for in the retrieval of the biophysical parameters of agricultural crops by means of polarimetric synthetic aperture radar interferometry.

Polarimetric SAR Interferometry has been recently applied to the retrieval of parameters from agricultural crops. The estimation procedure is based on the assumption of the so- called Oriented Volume over Ground model for the scene. Unfor- tunately, this model leads to a nonlinear formulation that must be inverted. In addition, the inverse problem is indeterminate, since there are more unknowns than input data. In this paper, two strategies used to solve this problem are described. The first one is based on a combination of a geometrical approach (a line fit on the complex plane) and a subsequent numerical method implemented with a genetic algorithm. The second technique extends the observation space by using data from two different baselines. The discussion about these methods is illustrated with indoor data from corn and rice crops.

An existing two-layer model for forest height estimation is adapted for agricultural crops in order to develop a retrieval algorithm based on polarimetric synthetic aperture radar interferometry. This new inversion scheme is specifically tailored for vertically oriented agricultural crops, with extinction coefficients dependent on the wave polarization. Physical parameters of the vegetation scene are estimated from the location of the measured coherences in the complex plane. The proposed inversion scheme is validated experimentally with indoor wide-band polarimetric measurements on samples of corn and rice fields. Results show that the estimates of the thickness of the vegetation layer and the ground topography are reasonably accurate for a wide range of frequencies and baselines. Moreover, some interesting results are also obtained when using only dual-polarized data, which brings up new applications for present and future spaceborne missions.

Polarimetric SAR interferometry (Pol-InSAR) has been applied successfully to retrieve biophysical parameters from forest areas. This technique employs a scene model composed of a random volume over the ground. However, that model is based on the assumption that the wave propagation through the vegetation canopy is independent of polarization, which is not valid for agricultural crops because the medium shows a strong anisotropic behavior. In this work we have studied the implications of the polarization dependence of the extinction coefficients, and a redefinition of the line fit inversion model is proposed and validated with indoor measurements. As we will show, provided that agricultural vegetation is short, the influence of the different extinction coefficient is minor. Consequently, it is demonstrated that an accurate inversion of the vegetation height and the underlying topography is still possible. However, it is also shown that the model is not sensitive to changes in the extinction coefficient, so this parameter is not well estimated. Implications about the necessity of fully or partially polarimetric data have been also analyzed

An exhaustive analysis of the electromagnetic response from a maize sample, measured in laboratory conditions, is presented. Fully polarimetric wide-band radar data have been collected in order to compute backscatter profiles (i.e. 1D images) as a function of height. First results consist on backscatter images expressed in Pauli and linear basis. These plots demonstrate that the extinction coefficient is higher for vertical polarization than for horizontal polarization due to the predominant vertical orientations of the components of the corn plants (mainly the stems). In addition, the crosspolar backscattered power decreases importantly as the frequency increases, and it is always lower than the copolar returns, again as a result of the oriented morphology, which produces a very low depolarization of the signal. On the other hand, the second kind of results is based on the polarimetric target decomposition proposed by Cloude and Pottier. Height profiles of alpha, entropy and anisotropy have been calculated. These parameters have been carefully analyzed in order to extract information to be used in the application of parameters inversion algorithms

A physics-based electromagnetic model specially tailored for the computation of the radar backscatter from rice crops is being developed. The model produces estimates of the coherent polarimetric backscatter response as a function of the sensor parameters and the physical description of the scene. This model is currently under validation against experimental data gathered in the European Microwave Signature Laboratory (EMSL). First results, obtained by a direct comparison of simulated and measured backscattering coefficients (with HH, HV and VV polarizations), show a reasonable agreement. The dependence on incidence angle, frequency and morphological characteristics of the rice plants is currently being assessed.

In this study, the potential use of the non-uniform FFT (NUFFT) in SAR imaging is analyzed. The main objective has been the improvement of the computational efficiency and image accuracy of seismic migration SAR processing. Different NUFFT methods have been implemented and tested in order to choose an adequate technique for the imaging problem. Our approach consists in substituting both the Stolt interpolation and the final range inverse FFT, in the /spl omega/-k algorithm, by a single NUFFT. Numerical simulations illustrate the performance of the new method and the influence of the selection of NUFFT parameters in the precision and computation time of the SAR imaging algorithm.

In May 1999 the SMOS Earth Explorer Opportunity mission was selected by ESA to provide global soil moisture and sea salinity maps using the MIRAS L-band aperture synthesis interferometric radiometer. At this frequency there is a high dynamic range of the soil moisture signature and surface roughness effects are small, as well as vegetation effects. However, they cannot be neglected at large observation angles or if the vegetation cover is thick and/or dense. In this paper we have carried out a series of simulations to investigate the effect of the soil moisture content and the vegetation density, the influence of the vegetation structure (number, sizes and orientation of branches and leaves) and the influence of each component (trunk, branches, and leaves) on the four Stokes elements J, Th, U and V.

A recently developed technique, polarimetric radar interferometry, is applied to tackle the problem of the detection of buried objects embedded in surface clutter. An experiment with a fully polarimetric radar in an anechoic chamber has been carried out using different frequency bands and baselines. The processed results show the ability of this technique to detect buried plastic mines and to measure their depth. This technique enables the detection of plastic mines even if their backscatter response is much lower than that of the surface clutter.

A coherence optimization method, which makes use of polarimetry to enhance the quality of SAR interferograms, has been experimentally tested under laboratory conditions in an anechoic chamber. By carefully selecting the polarization in both images, the resulting interferogram exhibits an improved coherence above the standard HH or VV channel. This higher coherence produces a lower phase variance, thus estimating the underlying topography more accurately. The potential improvement that this technique provides in the generation of digital elevation models (DEM) of non-vegetated natural surfaces has been observed for the first time on some artificial surfaces created with gravel. An experiment on a true outdoor DEM has not been accomplished yet, but the first laboratory results show that the height error for an almost planar surface can be drastically reduced within a wide range of baselines by using the optimization algorithm. This algorithm leads to three possible interferograms associated with statistically independent scattering mechanisms. The phase difference between those interferograms has been employed for extracting the height of vegetation samples. This retrieval technique has been tested on three different samples: maize, rice, and young fir trees. The inverted heights are compared with ground truth for different frequency bands. The estimates are quite variable with frequency, but their complete physical justification is still in progress. Finally, an alternative simplified scheme for the optimization is proposed. The new approach (called polarization subspace method) yields suboptimum results but is more intuitive and has been used for illustrating the working principle of the original optimization algorithm.

The authors show how the entropy-alpha target decomposition scheme may be used for parametric inversion studies on random particle cloud models for vegetation layers. The decomposition is detailed first and then applied to a two-parameter model for backscatter from a random cloud of small anisotropic particles. The two main parameters used are the mean particle shape and the mean orientation angle of the cloud. An inversion algorithm is presented and applied to broad-band polarimetric radar data from the European Microwave Signature Laboratory (EMSL), Joint Research Center, Ispra, Italy. The results have been obtained from measurements of a fir tree and a ficus tree. They show a wavelength scale dependence of the shape and distribution of scatterers, which reflects the complex volume scattering nature of such problems. Moreover, the values and trends from these two trees as a function of the frequency are different, as expected from their physical structures. Consequently, this algorithm has the potential to be useful in the construction of classification schemes for vegetation.

An efficient and complete electromagnetic scattering model for multiple tree trunks above the ground is presented. The individual trunks are modeled by finite-length stratified cylinders, with an external corrugated bark layer. The ground is modeled by a dielectric half space underneath the cylinders, with a slightly rough interface. The spatial distribution of the cylinders is random over an elliptic illuminated area, and their lengths follow a Gaussian distribution. Moreover, cylinders are randomly oriented with a variable degree of noncolinearity, and the ground can have an arbitrary orientation. The scattering results provided in this paper consider the summation of all scattered fields from the trunk-ground sets as a response to the incident field. Results have been computed for the mean square error of the bistatic scattering pattern, by means of a Monte Carlo simulation. The computed patterns are highly dependent on the geometry of the problem (lengths and radii of the trunks, tilting, surface roughness, incidence, and observation angles). The variability of the involved random parameters produces softening effects and an increased cross-polarized signal. Moreover, results are very sensitive to the dielectric characteristics of the trunks. As a result, the method can be useful in the interpretation of remote-sensing measurements from vegetation and in the development of classification schemes.