Multi-swath SAR interferometry is a powerful tool for assessing sub-wavelength changes over large-scale areas. The azimuth variation of the line of sight (LOS) induces phase jumps between adjacent bursts in the interferograms which contain useful information about the motion. In this work, we present a multitask convolutional neural network that simultaneously decouples the interferometric phase due to displacements in the LOS direction from that due to displacements in the along-track direction, and predicts a proxy for the alongtrack displacement. We show results using a single pair of Sentinel-1 acquisitions over the inland region of Greenland, where glacier flows occur in the winter season within the revisit time

SAR image compression is essential for managing the large amounts of data generated by Synthetic Aperture Radar (SAR) systems, ensuring efficient storage, transmission and processing without compromising essential information. Various techniques, including wavelet-based methods and predictive coding, are commonly used to achieve an optimal balance between compression ratio and image quality. Autoencoders within a deep learning framework have been successfully applied to SAR image compression; however, the simultaneous challenge of compression and speckle reduction remains unsolved due to the lack of ground truth. This study addresses this gap by proposing a self-supervised framework for SAR speckle reduction and extending its application to the joint problem of SAR image compression. The developed network learns a representation of SAR data that not only facilitates effective speckle reduction, but also enables image compression. We compare our method with state-of-the-art despeckling and compression algorithms and show that we can perform both tasks together with excellent performance.

Building damage assessment is critical in regions facing geopolitical challenges. This paper explores the use of Synthetic Aperture Radar remote sensing data, specifically from the Sentinel-1 constellation, to improve the accuracy and operational efficiency of building damage assessment. The approach is based on the use of SAR backscatter time series for a pixel-wise change detection methodology. As a case study, we consider the state of Ukraine, which has experienced significant building damage due to the ongoing Russo-Ukrainian war. Using the presented approach, we demonstrate the feasibility of change detection from freely available SAR data with weekly temporal resolution and at a nationwide spatial scale.

The scanning synthetic aperture radar (ScanSAR) and Terrain Observation by Progressive Scans (TOPSs) burst acquisition modes are nowadays among the most widely used in synthetic aperture radar (SAR) satellite missions. Both allow for increased coverage at the expense of azimuth resolution. However, the intermittent nature of the burst acquisition results in an increased sensitivity toward burst edges to displacements in the along-track (AT) dimension. In the presence of azimuth motion in the scene, phase jumps between bursts occur. In this contribution, this increased sensitivity is considered as an opportunity to obtain information on the North-South displacement, in which current SAR systems are less sensitive due to their quasi-polar orbits. Specifically, we suggest the usage of a multitask learning (MTL) architecture trained in a supervised fashion to separate the phase contribution due to displacements in the zero-Doppler (ZD) direction from AT displacements and to further provide a first rough estimation for the along-track displacement. Through an ad hoc network architecture and loss functions, we inject information about the interferometric SAR system model into the learning process, following a machine learning approach. We apply our method to the estimation of inland glacier flow from Sentinel-1 interferometric wide (IW)-swath data. We show that we are able to estimate, with an excellent performance, AT surface displacements of a few centimeters to several tens of centimeters, providing an improvement in accuracy compared with speckle tracking, and in coverage compared with techniques that exploit the burst-overlap differential phase.

This paper focuses on the estimation of interferometric SAR parameters, a step that precedes the entire interferometric processing chain to produce derived information such as digital elevation models and ground displacement. Deep learning, especially convolutional neural networks (CNN), has revolutionized image denoising and has recently received considerable attention. However, traditional supervised approaches require labeled images for training, which are generally unavailable or inaccurate, especially in remote sensing applications. To overcome this limitation, semi- and self-supervised denoising approaches have recently been proposed. These can learn from exclusively noisy samples, which can be obtained from pairs of noisy images or from noisy values within the same image. In this paper, we build on the foundation of these self-supervised learning methods, in particular, we borrow concepts from the Noise2Void and Noise2Self approaches, which have already shown excellent performance in various image denoising tasks. We extend this method to address the challenges specific to InSAR phase and coherence estimation, where the complex-valued nature of SAR interferograms poses unique processing considerations.

Image-to-image translation between different imaging modalities in Earth observation has become a widely utilized application area of deep learning. However, most of the translation is performed on real-valued data, to some extent neglecting the opportunities of complex-valued SAR data for interferometric methods. In this work, we propose a multi-task deep learning approach for simulating complex-valued InSAR data based on splitting the overall task into multi-modal image-toimage translation sub-tasks. Instead of synthesizing complex-valued SAR data directly, magnitudes, phase values and coherence magnitudes are simulated in parallel and combined to full complex-valued information afterward. With experiments on a Sentinel-1 interferogram, conditioned by DEM and land cover data, we demonstrate the feasibility of the approach.

Object recognition in synthetic aperture radar images is a well studied topic that has gained a significant amount of attention within the last decades. Modern approaches are based on machine learning, i.e. deep learning, and often show excellent performance. What is so far missing in the literature is a study dedicated to the generalization capabilities of object recognition approaches, i.e. how well a given system can be transferred to new and previously unseen data. In this paper, the proposed recognition model is trained and tested on a unique dataset of 25 high-resolution TerraSAR-X images (X-band), acquired over four different airports in Staring Spotlight mode. We show how classification performance changes for different application scenarios which require different training and evaluation setups.

Phase unwrapping (PU) is among the most critical tasks in synthetic aperture radar (SAR) interferometry (InSAR). Due to the presence of noise, the interferogram usually presents phase inconsistencies, also called residues, which imply a nonunivocal solution. This work investigates the PU problem from a semantic segmentation perspective by exploiting convolutional neural network (CNN) models. In particular, by exploiting a popular deep-learning architecture, we introduce the interferometric coherence as an input feature and analyze the performance increase against classical methods. For the network training, we generate a variegated data set by introducing a controlled number of phase residues, and considering both synthetic and real InSAR data. Eventually, we compare the proposed method to state-of-the-art algorithms on synthetic and real InSAR data taken from the TanDEM-X mission, obtaining encouraging results.

The TanDEM-X Synthetic Aperture Radar (SAR) system allows for the recording of the bistatic interferometric coherence, which adds additional information to the common amplitude images acquired by monostatic SAR systems. More concretely, the volume decorrelation factor, which influences the interferometric coherence, has been proved to be a reliable indicator of vegetated areas and was exploited in [1] to generate the global TanDEM-X Forest/Non-Forest Map, based on a supervised clustering algorithm. In this work, we investigate ad-hoc training strategies to extent the Convolutional Neural Network (CNN) presented in [2] for mapping forests and monitoring the extend of the Amazonas using TanDEM-X. By applying the proposed method on single TanDEM-X images, we achieved a significant performance improvement with respect to the clustering approach, with an f-score increase of 0.13, using as reference a forest map of 2010 based on Landsat data. The improvement in the forest classification makes it possible to skip the weighted mosaicking of overlapping images used in the clustering approach for achieving a good final accuracy. In this way, we were able to generate three time-tagged mosaics over the Amazon rainforest, by utilizing the nominal TanDEM-X acquisitions between 2011 and 2017. In the final paper, we will present more consolidated results, including the validation and comparison of the generated mosaics, as well as change detection investigations, aimed at showing the capabilities of Deep Learning approaches for forest mapping and monitoring with bistatic TanDEM-X images.

Decorrelation phenomena are always present in synthetic aperture radar interferometry (InSAR). While this implies a certain level of signal degradation, decorrelation is also a characteristic of the type of imaged target itself and can, therefore, be seen as a source of information. In this letter, we investigate InSAR decorrelation effects at the X-band by fitting volume and temporal decorrelation trends using the unique combination of data provided by the TanDEM-X (TDX) and PAZ spaceborne missions. The innovative use of this constellation allows for the acquisition of both single- and repeat-pass data at short revisit times. The concurrent availability of simultaneous acquisitions and the fine temporal resolution makes this constellation the ideal observation scenario for the study of decorrelation phenomena. Overall, we analyze five test sites, characterized by the presence of different land cover classes, and for each of them, we provide volume and temporal decorrelation fitting parameters. The performed analysis gives a first insight on the potential of combining bistatic and repeat-pass InSAR acquisitions also in view of future spaceborne constellations, which could benefit from the TDX/PAZ experience.

Nowadays, deep learning (DL) finds application in a large number of scientific fields, among which the estimation and the enhancement of signals disrupted by the noise of different natures. In this article, we address the problem of the estimation of the interferometric parameters from synthetic aperture radar (SAR) data. In particular, we combine convolutional neural networks together with the concept of residual learning to define a novel architecture, named Φ-Net, for the joint estimation of the interferometric phase and coherence. Φ-Net is trained using synthetic data obtained by an innovative strategy based on the theoretical modeling of the physics behind the SAR acquisition principle. This strategy allows the network to generalize the estimation problem with respect to: 1) different noise levels; 2) the nature of the imaged target on the ground; and 3) the acquisition geometry. We then analyze the Φ-Net performance on an independent data set of synthesized interferometric data, as well as on real InSAR data from the TanDEM-X and Sentinel-1 missions. The proposed architecture provides better results with respect to state-of-the-art InSAR algorithms on both synthetic and real test data. Finally, we perform an application-oriented study on the retrieval of the topographic information, which shows that Φ-Net is a strong candidate for the generation of high-quality digital elevation models at a resolution close to the one of the original single-look complex data.

Decorrelation phenomena are always present in Synthetic Aperture Radar Interferometry (InSAR). While this implies a certain level of signal degradation, the decorrelation is also a characteristic of the type of imaged target itself and can therefore be seen as a source of information. Correctly accounting for the type and amount of decorrelation is crucial when using InSAR systems for land classification purposes. In this paper we aim at modeling InSAR decorrelation effects at X-band for several land cover classes. In particular we model the volume and temporal decorrelation, by exploiting TanDEM-X and PAZ joint time-series. The uniqueness of the combined use of these two missions is the availability of simultaneous bistatic as well as short revisit time repeat-pass acquisitions, making it the ideal observation scenario for the study of decorrelation phenomena. The paper shows the preliminary results of the analysis on the city of Madrid (Spain) and for two land cover classes.

In the last few years, deep learning (DL) has been successfully and massively employed in computer vision for discriminative tasks, such as image classification or object detection. This kind of problems are core to many remote sensing (RS) applications as well, though with domain-specific peculiarities. Therefore, there is a growing interest on the use of DL methods for RS tasks. Here, we consider the forest/non-forest classification problem with TanDEM-X data, and test two state-of-the-art DL models, suitably adapting them to the specific task. Our experiments confirm the great potential of DL methods for RS applications.

Nonlocal algorithms have been proven to be a very effective tool for the reconstruction of the interferometric phase field in synthetic aperture radar images. The offset-compensated nonlocal filtering is a recent concept proposed to cope with the problem of the rare patch effect which arises especially in presence of terrain slopes with a large topography variation. In this paper we assess the quality of different Digital Elevation Models (DEMs) generated by varying the phase estimation algorithm. In the specific, we aim at analyzing the performance of the Offset-Compensated InSAR-BM3D filter with respect to state-of-the-art nonlocal filters, i.e. the InSAR-BM3D and the NLSAR filters. We perform experiments on real TanDEM-X data and exploit a high-resolution LiDAR DEM over the Austrian Alps in order to analyze the filter performance in terms of DEM's residual height error and details preservation.

The EC/ESA Copernicus program provides a long-term data base that is a unique opportunity for the constant monitoring of the dynamic processes of our Planet. The observation of the forest coverage is of primary importance for the study of the carbon cycle and plays a fundamental role for the management of Earth's natural resources. In this paper we present a strategy to map forested areas by exploiting interferometric Sentinel-1 acquisitions. We observe the evolution in time of the temporal decorrelation of Sentinel-1 stacks and provide a processing and classification framework. We show results over the Amazon rainforest, in particular over the Brazilian Rondonia state, where intensive deforestation phenomena take place.

In this paper, we present a novel approach to monitor the evolution of deforested areas in the Amazon rainforest, by combining Sentinel-1 and TanDEM-X SAR data. The idea is firstly to exploit the large coverage and short revisit time provided by the constellation of Sentinel-1 satellites in order to cover the entire arch of deforestation about ones per month. The goal is here to discriminate forest/non- forest by exploiting C-band backscatter signatures in dual polarization together with the behavior of the interferometric coherence in time, in order to identify the so-called deforestation hotspots: local areas characterized by a significant amount of on-going deforestation activities. Secondly, high-resolution time series of bistatic TanDEM-X data can be acquired over these hot spots with a repeat-cycle of 11 days, and used to track fast changes at small scales, aiming at identifying specific on-going deforestation activities. In the final paper, we intent to present more consolidated results, supported by a large scale acquisition scenario.

Nonlocal filters show outstanding performance in the field of interferometric phase restoration by providing strong filtering power together with high spatial features preservation. In this work we focus on the generation of Digital Elevation Models (DEM) from a pair of interferometric SAR images. In the specific, we aim at comparing the performance of state-of-the-art InSAR filtering approaches on the basis of their noise suppression and detail preservation capabilities. We exploit a dataset of TanDEM-X SAR data relative to the volcanic area of the Kamchatka region (Russia).

Tandem-L is a proposal for a spaceborne L-band SAR mission for the systematic observation of dynamic processes on the Earth's surface with hitherto unparalleled quality and resolution. The enormous amount of data to be acquired will be used to provide information concerning dynamic processes in the biosphere, geosphere, cryosphere, and hydrosphere [1]. Within the scope of this mission, several operational phases have been established and corresponding flight configurations have been designed. Subsequently an observation concept has been developed in order to fulfill all the scientific requirements. A sophisticated planning and optimization process has been implemented. It combines the predetermined satellite resources, ground station network, and the scientific requirements to generate an acquisition timeline on individual data take level. The resulting timeline fulfills all requirements for the various scientific applications and can be used to analyze different aspects of the mission like data volumes and orbit usage.

The block-matching 3-D (BM3D) algorithm, based on the nonlocal approach, is one of the most effective methods to date for additive white Gaussian noise image denoising. Likewise, its extension to synthetic aperture radar (SAR) amplitude images, SAR-BM3D, is a state-of-the-art SAR despeckling algorithm. In this paper, we further extend BM3D to address the restoration of SAR interferometric phase images. While keeping the general structure of BM3D, its processing steps are modified to take into account the peculiarities of the SAR interferometry signal. Experiments on simulated and real-world Tandem-X SAR interferometric pairs prove the effectiveness of the proposed method.