Search

A timely and accurate spatial mapping of built-up areas (BA) is crucial in making cities and human settlements safe, resilient, and sustainable. Synthetic Aperture Radar (SAR) data are useful for BA mapping due to strong coherent backscatter from diverse human-made targets, distinct texture patterns, and sensitivity to its geometric characteristics. However, BA mapping using SAR data is still challenging due to various geometrical and physical factors that often lead to misinterpretations. Therefore, this study presents a novel method to map BA using single-date dual polarized Sentinel-1 SAR data to overcome such challenges in complex built-up scenarios. Here, we propose three normalized descriptors derived from the Stokes vector elements. These three descriptors are found to be vital in characterizing different types of BA structures. We then consider the symmetric quadratic mean of the normalized descriptors to propose a novel dual polarimetric radar built-up index (DpRBI). The proposed index provides information about the average normalized polarized power, which is useful for mapping diverse types of BA or their combination in an image. We assess the proposed methodology utilizing the Sentinel-1 image acquired over three major cities: Delhi, Milan, and Barcelona. The BA map accuracy for all three cities was found to be ≈ 85 % with a low BA omission error. We compared the mapping results with available BA/land cover products: World Settlement Footprint 2019 (WSF-19) and World cover 2020 (WC-20). The BA map obtained using the proposed index was better than WSF-19 and WC-20, particularly for BA omission errors. For example, we observed a relatively low BA omission error of 22.34 % using DpRBI compared to 31.86 % and 41.79 % for WC-20 and WSF-19, respectively, over Milan, Italy. The code is available at: https://github.com/navv37/DpRBI [ABSTRACT FROM AUTHOR]

Building height retrieval from synthetic aperture radar (SAR) imagery is of great importance for urban applications, yet highly challenging due to the complexity of SAR data. This paper addresses the issue of building height retrieval in large-scale urban areas from a single TerraSAR-X spotlight or stripmap image. Based on the radar viewing geometry, we propose that this problem be formulated as a bounding box regression problem and therefore allows for integrating height data from multiple data sources in generating ground truth on a larger scale. We introduce building footprints from geographic information system (GIS) data as complementary information and propose a bounding box regression network that exploits the location relationship between a building's footprint and its bounding box, enabling fast computation. The method is validated on four urban data sets using TerraSAR-X images in both high-resolution spotlight and stripmap modes. Experimental results show that the proposed network can reduce the computation cost significantly while keeping the height accuracy of individual buildings compared to a Faster R-CNN based method. Moreover, we investigate the impact of inaccurate GIS data on our proposed network, and this study shows that the bounding box regression network is robust against positioning errors in GIS data. The proposed method has great potential to be applied to regional or even global scales. Our code will be made publicly available at github.com/ya0-sun/bbox-SAR-building. [ABSTRACT FROM AUTHOR]

Fully Polarimetric Synthetic Aperture Radar (PolSAR) has the advantages of all-weather, day and night observation and high resolution capabilities. The collected data are usually sorted in Sinclair matrix, coherence or covariance matrices which are directly related to physical properties of natural media and backscattering mechanism. Additional information related to the nature of scattering medium can be exploited through polarimetric decomposition theorems. Accordingly, PolSAR image classification gains increasing attentions from remote sensing communities in recent years. However, the above polarimetric measurements or parameters cannot provide sufficient information for accurate PolSAR image classification in some scenarios, e.g. in complex urban areas where different scattering mediums may exhibit similar PolSAR response due to couples of unavoidable reasons. Inspired by the complementarity between spectral and spatial features bringing remarkable improvements in optical image classification, the complementary information between polarimetric and spatial features may also contribute to PolSAR image classification. Therefore, the roles of textural features such as contrast, dissimilarity, homogeneity and local range, morphological profiles (MPs) in PolSAR image classification are investigated using two advanced ensemble learning (EL) classifiers: Random Forest and Rotation Forest. Supervised Wishart classifier and support vector machines (SVMs) are used as benchmark classifiers for the evaluation and comparison purposes. Experimental results with three Radarsat-2 images in quad polarization mode indicate that classification accuracies could be significantly increased by integrating spatial and polarimetric features using ensemble learning strategies. Rotation Forest can get better accuracy than SVM and Random Forest, in the meantime, Random Forest is much faster than Rotation Forest. [ABSTRACT FROM AUTHOR]

In the last decade Differential Synthetic Aperture Radar (DInSAR) data were successfully tested in a number of case studies for the detection, mapping and monitoring of ground displacements associated with natural or anthropogenic phenomena. More recently, several national and regional projects all around the world provided rich data archives whose confident use, however, should rely on multidisciplinary experts in order to avoid misleading interpretations. To this aim, the present work first introduces a general framework for the use of DInSAR data; then, focusing on the analysis of subsidence phenomena and the related consequences to the exposed facilities, a set of original procedures is proposed. By drawing a multiscale approach the study highlights the different goals to be pursued at different scales of analysis via high/very high resolution SAR sensors and presents the results with reference to the case study of the Campania region (southern Italy) where widespread ground displacements occurred and damages of different severity were recorded. [ABSTRACT FROM AUTHOR]