SWOT is an innovative altimetry mission launched by NASA and CNES in December 2022. The data acquired by the HR mode of the principal instrument KaRIn enable precise estimation of elevation, area and related parameters for continental water surfaces, globally and repeatedly. In this article we briefly present the different science data products, with real-world examples and some preliminary results on performance. The final presentation will report on the global performance assessments which are due for the SWOT Science Validation Meeting in June 2024.

Water detection is a key step in the operational processing of KaRIn HR images from the SWOT mission, which provides water surface extent and elevation for continental water surfaces globally and repeatedly. A new water detection method has been developed for this sensor due to its specific characteristics. In this article we present examples of detected water masks and preliminary assessments of their accuracy. We also provide perspectives for further improvements in the water detection algorithm.

River floodplains and intertidal coastal environments are key areas for which the Surface Water and Ocean Topography (SWOT) altimetry mission is expected to provide valuable data, improving our ability to monitor these dynamic systems. SWOT provides water surface extent and elevation for continental water surfaces globally and regularly. This article presents two methods to derive the topography of such environments from SWOT data. The first one, the waterline method, is the approach currently developed to produce worldwide floodplain/intertidal DEM. The second method uses a direct approach by exploiting InSAR heights extracted directly from SWOT products. An example is shown for the Bay of Veys located in Normandy (France).

The SWOT satellite is a near-nadir Ka-band interferometric radar, capable of monitoring water bodies larger than 6 ha. Launched in December 2022, the satellite was in a calibration/validation orbit until July 2023, where it acquired measurements every day over certain regions. Canadian lakes were ice-covered at the start of the calibration period, offering the opportunity to study the Ka-band backscatter in the presence of ice and snow. The SWOT signal is also affected when the water surface is very smooth (e.g. in the absence of wind), or attenuated by heavy precipitation. These preliminary results demonstrate for the first time the impact of ice, snow, wind, and rain on the detection of water bodies by the SWOT satellite signal.

This paper presents a semi-guided method to detect lakes in Sentinel-1 SAR data. The proposed approach is an adaptation of the grab-cut framework developed in [1]. Starting from a coarse bounding box around the lake, an accurate segmentation is extracted using a Conditional Random Field formalism and a graph-cut based optimization. Then an extension of this approach to process jointly a stack of multi-temporal data is presented. A temporal regularization term is introduced to control the joint segmentation. The proposed approach is evaluated on Sentinel-1 datasets. Qualitative and quantitative results demonstrate the interest of the proposed framework and its robustness to the initial-ization polygon of the lake.

The goal of the Surface Water and Ocean Topography (SWOT) mission, a future wide-swath radar altimeter, is to measure the elevations of ocean and continental water surfaces with unprecedented precision. A simulator of SWOT high-rate (HR) interferometric SAR images has been developed to provide SWOT-like data to scientists and as a tool to develop and test HR ground processing algorithms. In this letter, we evaluate the errors of the extracted water levels in different hydrodynamic and topographic contexts in the Seine estuary based on simulated SWOT HR data. The results confirm height errors generally below 10 cm after km-scale averaging for the majority of the observations. Higher errors mostly occur in areas where the surrounding topography causes layover. This letter also reveals the interactions between tide and streamflow and their impact on the spatial variability of the water levels in the estuary.

The upcoming Surface Water and Ocean Topography (SWOT) mission, projected for launch in 2021, will provide global measurements of water elevation for ocean and continental water bodies with unprecedented resolution and accuracy. This article describes how the variations in height and extent of continental water surfaces, as measured by SWOT's principal instrument, the Ka-band Radar Interferometer (KaRIn), can be used to derive floodplain Digital Elevation Models (DEMs), in principle with submeter accuracy. The proposed method is based on InSAR processing, refined geolocation, and the so-called bath-tub ring approach. Preliminary results obtained on multitemporal simulated SWOT data are presented.

Classification of SAR images is a challenging task as the radiometric properties of a class may not be constant throughout the image. The assumption made in most classification algorithms that a class can be modeled by constant parameters is then not valid. In this paper, we propose a classification algorithm based on two Markov random fields that accounts for local and global variations of the parameters inside the image and produces a regularized classification. This algorithm is applied on airborne TropiSAR and simulated SWOT HR data. Both quantitative and visual results are provided, demonstrating the effectiveness of the proposed method.

The objective of the Surface Water and Ocean Topography (SWOT) mission is to regularly monitor the height of the earth's water surfaces. One of the challenges toward obtaining global measurements of these surfaces is to detect small water areas. In this article we introduce a method for the detection of thin water surfaces, such as rivers, in SWOT images. It combines a low-level step (segment detection) with a high-level regularization of these features. The method is then tested on a simulated SWOT image.

Surface Water and Ocean Topography (SWOT) is a future wide-swath radar altimetry mission. The main instrument is the Ka-band radar interferometer (KaRIn). It provides interferometric radar images that will be processed to obtain water elevation estimates worldwide. Due to the specific geometric and radiometric characteristics of KaRIn/SWOT, the phase-to-height conversion is one of the most critical processing steps. We propose methods to estimate water heights without using classical approaches based on spatial phase unwrapping and control points. Results obtained on the simulated SWOT data are shown.

In preparation for the Surface Water and Ocean Topography (SWOT) mission, scheduled for launch around October 2020, ONERA and CNES evaluated SAR processors intended for airborne SAR processing in configurations close to that of the eventual space platform. Already during preliminary studies from aircraft (with ONERA operated touring moto-glider carrying the BuSARd radar, and later on the JPL operated airSWOT radar), difficulties due to both extremely low incident angle and high frequency band emerged. The most significant issues are: first the strong topographic effects on the aperture, which makes the spectral support strongly varying across swath even for small separation between the reference trajectory and the effective one; second, the large distortion across-track induced by even small altitude deviation. These two effects combine in making interferometric pairs of SAR images difficult to co-generate and to motion compensate to an accurate prior altitude reference. Mitigation issues are proposed that allow to use faster frequency domain processing and simpler surface water altitude retrieval.

The principal instrument of the NASA/CNES wide-swath altimetry mission Surface Water and Ocean Topography (SWOT) is the Ka-band Radar Interferometer (KaRIn), a bistatic synthetic aperture radar (SAR) system operating on near-nadir swaths on both sides of the satellite track. There are limited reports on backscattering from natural surfaces at this short wavelength and particular observation geometry. Near-field backscattering measurements on water, as well as the first interferometric airborne SAR acquisitions at Ka-band covering the 0.6 °-3.9 ° incidence range of KaRIn, were therefore conducted. The experimental results confirm expected characteristics of near-nadir Ka-band interferometric SAR imagery, such as strong water/land radiometric contrast (typically in the order of 10 dB) and very high interferometric coherence on water.

This article summarizes the data processing of the proposed wide-swath altimetry mission Surface Water and Ocean Topography (SWOT), and more specifically the processing steps up to level 2 for the Ka-band Radar Interferometer (KaRIn), which is a near-nadir viewing bistatic SAR system. A general overview of the proposed processing chains and data products is given, and some particularly challenging processing steps are commented on.

In this paper, a new technique is presented to detect the river networks in simulated SWOT images. The proposed algorithm is based on a noise reduction step followed by a directional morphological filter. In this work, the speckle noise reduction has been achieved by using a Curvelet-based filter preserving the structures of interest. After the filtering task, a river contrast enhancement has been presented by using the Path-Opening filter. This morphological filtering has retained the curvilinear structures on the image independently of their orientation. Hence, the river detection has been possible by a simple thresholding on the Path-Opening result. The obtained results are evaluated using a visual inspection and a quantitative evaluation. The potential of the proposed algorithm has been evaluated by studying the robustness of the parameters.

This paper gives an overview of an empirical cross-calibration technique developed for the Surface Water Ocean Topography mission (SWOT). The method is here used to detect and to mitigate two spatially coherent errors in SWOT topography data: the baseline roll error whose signature is linear across track, and the baseline length error whose signature is quadratic across track. Assuming that topography data are corrupted by coherent error signatures that we can model, we extract the signatures, and we empirically use the error estimates to correct SWOT data. The cross-calibration is tackled with a two-step scheme. The first step is to get local estimates over cross-calibration zones, and the second step is to perform a global interpolation of local error estimates and to mitigate the error everywhere. Three methods are used to get local error estimates: 1) we remove a static first guess reference such as a digital elevation model, 2) we exploit overlapping diamonds between SWOT swaths, and 3) we exploit overlapping segments with traditional pulse-limited altimetry sensors. Then, the along-track propagation is performed taking the local estimates as an input, and an optimal interpolator (1-D objective analysis) constrained with a priori statistical knowledge of the problem. The rationale of this paper is to assume that SWOT's scientific requirements are met on all errors but the ones being cross-calibrated. In other words, the algorithms presented in this paper are not needed at this stage of the mission definition, and they are able to deal with higher error levels (e.g., if hardware constraints are relaxed and replaced by additional ground processing). Even in our most pessimistic theoretical scenarios of baseline roll and baseline length errors (up to 70 cm RMS of uncorrected topography error), the cross-calibration algorithm reduces coherent errors to less than 2 cm (outer edges of the swath). Residual errors are subcentimetric for very low-frequency errors (e.g., orbital revolution). Sensitivity tests highlight the benefits of using additional pulse-limited altimeters and optimal inversion schemes when the problem is more difficult to solve (e.g., wavelengths of less than 1000 km), but also to provide a geographically homogeneous correction that cannot be obtained with SWOT's sampling alone.

The future spatial SWOT mission will use a new altimetric sensor "KaRIn", which is a Ka-band interfero metric SAR system operating on near-nadir swaths on both sides of the satellite track [1]. The objective is to study the height of the earth's water surfaces, mainly oceans, but also continental water surfaces such as lakes and rivers. This article dedicated to water surface detection methods, presents a multi-scale line extraction approach that has been adapted to the specificities of the KaRIn instrument, and gives some preliminary results obtained on simulated SAR images.

The principal instrument of the wide-swath altimetry mission SWOT is KaRIn, a Ka-band interferometric SAR system operating on near-nadir swaths on both sides of the satellite track. Due to the short wavelength and particular observation geometry, there are very limited reports on the backscattering from natural surfaces. Models and simulators that cover radiometric and geometric aspects have therefore been developed. This article describes airborne acquisitions and ground measurements that are used to validate the models and simulators, and shows some preliminary results.

The principal instrument of the SWOT (Surface Water and Ocean Topography) altimetry mission is KaRIn, a Ka-band interferometric SAR system operating on near-nadir swaths on both sides of the satellite track. This article briefly describes the measurement principle, the processing steps and the specificities of the interferometric SAR data of KaRIn as compared to conventional spaceborne SAR systems.

With the ever-increasing number and diversity of Earth observation satellites, it steadily becomes more important to be able to analyze compound data sets consisting of different types of images acquired by different sensors. In this paper, we examine different ways of obtaining joint distributions of such images, and we propose a method that enables incorporation of correlations between images while keeping a good fit to the marginal distributions. The approach basically consists of two steps. First, the marginal densities are specified. Based on this specification, each marginal variable is transformed to a normal distributed variable. The joint distribution of the transformed variables is assumed to be multivariate normal. Transforming back to the original scale gives a joint distribution with dependence, where the initial marginal distributions are preserved. The parameters of the new joint distribution can be estimated. The focus is on marginal distributions that are Gamma,K, or Gaussian, although any distribution could be considered. The joint distributions produced by the transformation method can be used in supervised classification of radar and optical images. Results obtained for a set of four-look synthetic aperture radar (SAR) images, as well as a combination of SAR and optical images, are presented.

Several earth observation satellites acquire a panchromatic band with very high resolution and three or more spectral bands with lower resolution. Pan-sharpened versions of the spectral bands are in many cases available to the end users. In this study we assess the suitability of pan-sharpened images for automated analysis, as compared to resampled versions of the original spectral bands or the use of methods developed specifically for multiresolution data sets. We limit ourselves to supervised classification and concentrate on features at the highest resolution level. Tests on real and simulated satellite data indicate that pan-sharpened images are well suited for such analysis.

We assess the performance of synthetic aperture radar (SAR) compact polarimetry architectures based on mixed basis measurements, where the transmitter polarization is either circular or orientated at 45/spl deg/(/spl pi//4), and the receivers are at horizontal and vertical polarizations with respect to the radar line of sight. An original algorithm is proposed to reconstruct the full polarimetric (FP) information from this architecture. The performance assessment is twofold: it first concerns the level of information preserved in comparison with FP, both for point target analysis and crop fields classification, using L-band SIRC/XSAR images acquired over Landes forest and Jet Propulsion Laboratory AIRSAR images acquired over Flevoland. Then, it addresses the space implementation complexity, in terms of processed swath, downloading features, power budget, calibration, and ionospheric effects. The polarization uniqueness in transmission of this mixed basis mode, hereafter referred to as the /spl pi//4 mode, maintains the standard lower pulse repetition frequency operation and hence maximizes the coverage of the sensor. Because of the mismatch between transmitter and receiver basis, the power budget is deteriorated by a factor of 3 dB, but it can partly be compensated.

Several earth observation satellites acquire image bands with different spatial resolutions, e.g., a panchromatic band with high resolution and spectral bands with lower resolution. Likewise, we often face the problem of different resolutions when performing joint analysis of images acquired by different satellites. This work presents models and methods for classification of multiresolution images. The approach is based on the concept of a reference resolution, corresponding to the highest resolution in the dataset. Prior knowledge about the spatial characteristics of the classes is specified through a Markov random field model at the reference resolution. Data at coarser scales are modeled as mixed pixels by relating the observations to the classes at the reference resolution. A Bayesian framework for classification based on this multiscale model is proposed. The classification is realized by an iterative conditional modes (ICM) algorithm. The parameter estimation can be based both on a training set and on pixels with unknown class. A computationally efficient scheme based on a combination of the ICM and the expectation-maximization algorithm is proposed. Results obtained on simulated and real satellite images are presented.

An important question with respect to the performance of the interferometric cartwheel and similar bistatic SAR systems, is whether the ambiguities combine coherently and introduce artifacts in the resulting interferograms. We here review the theoretical behavior of range and azimuth ambiguities and extend the description to the case of non-zero Doppler processing. Simulation results confirm that ambiguities cannot significantly degrade the performance of an L-band interferometric cartwheel

In this article we describe a Bayesian model for integration of multiscale image data. The approach is based on the concept of a reference resolution. Data at this and lower resolutions are connected to the reference resolution through a fully specified statistical model. Algorithms for parameter estimation and classification based on the multiscale model are proposed, and results and comparisons with singlescale classification are presented.

Abstract — For a given ground cover class, there is no straightforward way of expressing the joint distribution of a set of correlated radar images represented in amplitude or intensity. In this article we propose a general transformation method that permits incorporation of inter-image covariance while keeping a good fit to the marginal distributions. The approach is here studied for Gamma marginals, and the results of tests on a multitemporal series of ERS-1 multi-look images are presented. I.

Due to the enormous quantity of radar images acquired by satellites and through shuttle missions, there is an evident need for efficient automatic analysis tools. This paper describes unsupervised classification of radar images in the framework of hidden Markov models and generalized mixture estimation. Hidden Markov chain models, applied to a Hilbert-Peano scan of the image, constitute a fast and robust alternative to hidden Markov random field models for spatial regularization of image analysis problems, even though the latter provide a finer and more intuitive modeling of spatial relationships. We here compare the two approaches and show that they can be combined in a way that conserves their respective advantages. We also describe how the distribution families and parameters of classes with constant or textured radar reflectivity can be determined through generalized mixture estimation. Sample results obtained on real and simulated radar images are presented.

The principal goal of the FOREMMS project is to develop and demonstrate an advanced forest environmental monitoring and management system prototype that uses both remote sensing data, ground measurements and ancillary data. This article presents the FOREMMS prototype on a conceptual level, including a system overview and a high-level description of the data collection and information extraction.

We study two related topics in decision fusion for multisensor image classification. The first topic is the use of a weighted logarithmic opinion pool compared to the statistical product combination rule. The performance is compared on three data sets. The second topic is related to different criteria for parameter estimation for a statistical fusion model. We propose an alternative criterion for estimation of the mean vector and the covariance matrix of a Gaussian model based on minimizing the number of misclassified training samples and compare the performance of this to the traditional maximum likelihood approach.

The authors compare estimators of the mean radar reflectivity on images with different spectral properties. By working on complex data rather than detected images, they can take the speckle correlation into account and thus obtain more accurate estimates. A robust and computationally efficient approximation of the optimal estimator is proposed.

The authors develop optimal criteria for detection and localization of step edges in single look complex (SLC) synthetic aperture radar (SAR) images. By working on complex data rather than intensity images, they can easily take the speckle autocorrelation into account, obtain more accurate estimates of local mean reflectivities, and thus achieve better edge detection and edge localization than with operators known from the literature. Algorithms for the two-dimensional (2D) implementation of the methods are proposed, and some segmentation results are shown.

In a paper on morphological image segmentation, Najman and Schmitt (1996) introduce the powerful concept of edge dynamics. In this communication, we show that the method that they propose to compute the edge dynamics gives erroneous results for certain spatial configurations, and we propose a new algorithm which always yields correct edge dynamics. The reply presents in detail the algorithm of the watershed, which have been sketched in the original article and criticized in the comment. First, the formal definition of the flooding list, the key data structure of the algorithm, is given. Then, the construction of this flooding list and of the watershed are described and proved.

Edge detection is a fundamental issue in image analysis. Due to the presence of speckle, which can be modeled as a strong, multiplicative noise, edge detection in synthetic aperture radar (SAR) images is extremely difficult, and edge detectors developed for optical images are inefficient. Several robust operators have been developed for the detection of isolated step edges in speckled images. The authors propose a new step-edge detector for SAR images, which is optimal in the minimum mean square error (MSSE) sense under a stochastic multiedge model. It computes a normalized ratio of exponentially weighted averages (ROEWA) on opposite sides of the central pixel. This is done in the horizontal and vertical direction, and the magnitude of the two components yields an edge strength map. Thresholding of the edge strength map by a modified version of the watershed algorithm and region merging to eliminate false edges complete an efficient segmentation scheme. Experimental results obtained from simulated SAR images as well as ERS-1 data are presented.

Edge detection is a fundamental issue in image analysis. Due to the presence of speckle, which can be modelled as a strong multiplicative noise, edge detection in synthetic aperture radar (SAR) images is very difficult and methods developed for optical images are inefficient. We here propose a new edge detector for SAR images which is optimum in the MMSE sense for a stochastic multiedge model. It computes a normalized ratio of exponentially weighted averages (ROEWA) on opposite sides of the central pixel. This is done in the horizontal and vertical direction, and the module of the two components yields an edge strength map. Thresholding of the edge strength map and postprocessing to eliminate false edges are briefly discussed. We present results on simulated SAR images and ERS1 data.

Segmentation plays a key role in many image processing applications, including SAR image analysis. Given a high quality segmentation, important tasks such as adaptive noise filtering and classification can be greatly simplified. However, the presence of a strong, multiplicative noise known as speckle makes edge detection in SAR images very difficult. A new, hybrid segmentation technique has permitted us to obtain segmentations of relatively good quality prior to filtering. This has led us to develop a region-based version of the LMMSE speckle filter. The segmentations may further be used for the classification of raw or filtered images. In this paper we give an overview of the entire scheme, which segments, filters and classifies SAR images. The study is limited to agricultural scenes composed of distinct parcels of relatively homogeneous reflectivity.