Andreas Felbier

Urban Footprint Processor—Fully Automated Processing Chain Generating Settlement Masks From Global Data of the TanDEM-X Mission

The German TerraSAR-X add-on for Digital Elevation Measurement (TanDEM-X) mission (TDM) collects two global data sets of very high resolution (VHR) synthetic aperture radar (SAR) images between 2011 and 2013. Such imagery provides a unique information source for the identification of built-up areas in a so far unique spatial detail. This letter presents the novel implementation of a fully automated processing system for the delineation of human settlements worldwide based on the SAR data acquired in the context of the TDM. The proposed Urban Footprint Processor (UFP) includes three main processing stages dedicated to: i) the extraction of texture information suitable for highlighting regions characterized by highly structured and heterogeneous built-up areas; ii) the generation of a binary settlement layer (built-up, non-built-up) based on an unsupervised classification scheme accounting for both the original backscattering amplitude and the extracted texture; and iii) a final post-editing and mosaicking phase aimed at providing the final Urban Footprint (UF) product for arbitrary geographical regions. Experimental results assess the high potential of the TDM data and the proposed UFP to provide highly accurate geo-data for an improved global mapping of human settlements.

TanDEM-X mission—new perspectives for the inventory and monitoring of global settlement patterns

Abstract. TerraSAR-X add-on for digital elevation measurement (TanDEM-X) is a German Earth observation mission collecting a total of two global coverages of very high resolution (VHR) synthetic aperture radar (SAR) X-band data with a spatial resolution of around three meters in the years 2011 and 2012. With these, the TanDEM-X mission (TDM) will provide a unique data set which is complementary to existing global coverages based on medium (MR) or high resolution (HR) optical imagery. The capabilities of the TDM in terms of supporting the analysis and monitoring of global human settlement patterns are explored and demonstrated. The basic methodology for a fully-operational detection and delineation of built-up areas from VHR SAR data is presented along with a description of the resulting geo-information product—the urban footprint (UF) mask—and the operational processing environment for the UF production. Moreover, potential follow-on analyses based on the intermediate products generated in the context of the UF analysis are introduced and discussed. The results of the study indicate the high potential of the TDM with respect to an analysis of urbanization patterns, peri-urbanization, spatio-temporal dynamics of settlement development as well as population estimation, vulnerability assessment and modeling of global change.

Pattern-Based Accuracy Assessment of an Urban Footprint Classification Using TerraSAR-X Data

Assessing the accuracy of land-cover classifications is a major challenge in remote sensing. This is mostly due to the absence of geometrically and thematically highly resolved, reliable, area wide, and up-to-date reference data. This study focuses on a multifaceted accuracy assessment of an urban footprint classification derived from a single-polarized TerraSAR-X image in stripmap mode for the city of Padang in Indonesia. For this purpose, a pixel-based approach was used to identify the urbanized and nonurbanized areas. As reference, a geometrically and thematically highly resolved, accurate, and detailed 3-D city model is available. Based on this data, the classification result is assessed by basic methodologies-square measures and error matrix. Beyond that, the accuracy of the urban footprint classification is analyzed in dependence of the physical structure of the complex urban landscape-defined by built-up density and building volumes. Results reveal that the accuracy of classification results varies in dependence of the structural characteristics of the particular urban environment. Furthermore, the study shows what is thematically mapped by an urban footprint classification.

Identification and characterization of urban structures using VHR SAR data

The global process of urbanization is associated with various ecological, social and economic changes in both the built-up area and the adjacent natural or cultivated landscape. To manage the effects and impacts of this development, effective urban and regional planning requires accurate and up to date information on the urban dynamics. This paper introduces a methodology to automatically detect human settlements and then further characterize the identified built-up areas in terms of the building density based on VHR SAR data. The SAR imagery is acquired by the German satellite system TerraSAR-X. Regarding the delineation of the built-up area in the region of Munich we achieved an overall accuracy of 94 % and a Kappa of 0.86. The estimation of building density showed a coefficient of determination (r2) of up to 0.74. The mean absolute error of the modeled building densities was 5 %.

Normalization of TanDEM-X DSM Data in Urban Environments With Morphological Filters

The TanDEM-X mission (TDM) is a spaceborne radar interferometer which delivers a global digital surface model (DSM) with an unprecedented spatial resolution. This allows resolving objects above ground such as buildings. Extracting and characterizing those objects in an automated manner represents a challenging problem but opens simultaneously a broad range of large-area applications. In this paper, we discuss and evaluate the suitability of morphological filters (MFs) for the derivation of normalized DSMs from the TDM in complex urban environments and introduce a novel region-growing-based progressive MF procedure. This approach is jointly proposed and can be combined with a postclassification processing scheme to specifically allow for a viable reconstruction of urban morphology in a challenging terrain. The filter approach comprises a multistep procedure using concepts of morphological image filtering, region growing, and interpolation techniques. Therefore, it extends the idea of progressive MFs. The latter aim to identify nonground pixels in the DSM by gradually increasing the size of a structuring element and applying iteratively an elevation difference threshold. After the identification of initial nonground pixels, here, potential nonground pixels are identified within each iteration, and their similarity with respect to neighboring nonground pixels is assessed. Pixels are finally labeled as nonground if a constraint is fulfilled. The postclassification processing scheme adapts techniques of object-based image analyses to further refine regions of classified nonground pixels. Digital terrain models are subsequently generated by interpolating between identified ground pixels. Experimental results are obtained for settlement areas that cover large parts of the cities of Izmir (Turkey) and Wuppertal (Germany). They confirm the capability of the proposed approaches for a reduction of omission errors compared to basic MF-based methods when classifying ground pixels, which is favorable in a mountainous terrain with steep slopes.

A novel method for building height estmation using TanDEM-X data

The main goal of the German TanDEM-X mission (TDM) is the provision of a global digital elevation model (DEM) at the unique spatial resolution of ~12m, which, compared to other freely available DEMs [e.g., STRM-DEM (90m), ASTER-GDEM (30m)], holds the potential to discriminate objects above the ground as, e.g., buildings. Accordingly, in this paper we present a novel unsupervised method for automatically estimating local height variations in built-up areas by means of the TanDEM-X DEM. In the first step, we identify points lying on the ground surface by means of a dedicated algorithm. In particular, the basic idea is that pixels associated with potential infrastructural elements can be identified and then excluded by analyzing the relative change in elevation with respect to their neighbors (indeed they locally exhibit greater height values with respect to ground pixels). Next, the remaining samples are used to generate a digital terrain model (DTM) of the investigated region by employing the Natural Neighbors (NN) interpolation algorithm. The local building height is finally estimated by subtracting the DTM from the original DEM. Preliminary experimental results obtained for the area of Dongying (China) which includes the Yellow River Delta (YRD) assess the effectiveness of the proposed approach and its potential to provide a reliable indication of the overall distribution of building height at large scale.

Pixel-based classification algorithm for mapping urban footprints from radar data: a case study for RADARSAT-2

The process of rapid urbanization is attended by various adverse effects on the society, the ecology, and the economy. Effective urban and regional planning is the key to minimizing these effects and the impact on the people who live in cities. For this reason, detailed geospatial information on urban dynamics is needed, which can help to analyze and understand the process of urbanization. Different studies have shown that high-resolution radar imagery is an excellent basis for classifying, monitoring, and analyzing the outline and spatio-temporal development of urban agglomerations. Specifically, the analysis of texture information based on local speckle characteristics has shown its ability to generate large area urban maps. In this study a pixel-based, fully automatic classification approach developed for TerraSAR-X StripMap data was transferred and applied to five RADARSAT-2 images acquired in ultra-fine mode. The algorithm was applied to data from the cities of Mannheim and Ludwigshafen and their rural surroundings in Germany. The classification approach was validated for its multisensoral transferability and robustness. A relative comparison to a TerraSAR-X classification and an absolute comparison to a reference dataset show promising results and allow for the conclusion that the methodology can be fully applied to RADARSAT-2 data.

The global urban footprint — Processing status and cross comparison to existing human settlement products

The main goal of the TanDEM-X mission (TDM) is the generation of a global digital elevation model (DEM). The global SAR dataset, which is made available in the context of the TDM, is also used to create a global human settlement layer, the Global Urban Footprint (GUF). This paper presents a first large area cross comparison between the Global Urban Footprint and existing human settlement products, which shows promising results with an achieved confidence of 95.86% Overall, 71.15% Producers and 85.22% Users accuracy.

Unsupervised high-resolution global monitoring of urban settlements

The TanDEM-X (TerraSAR-X add-on for Digital Elevation Measurement, TDM) German mission collects two global data sets of very high resolution (VHR) synthetic aperture radar (SAR) imagery between 2011 and 2013. With a spatial resolution of ~3 m, these images provide a unique data source for the delineation of built-up areas in a spatial detail that could not be addressed so far by currently existing global human settlements layers (GHSL). In this paper we present a novel unsupervised procedure that automatically detects built-up areas from TDM data at the unprecedented spatial resolution of 0.4 arcsec (~12 m). The proposed technique is being used to generate a world-wide inventory of human settlements referred to as Global Urban Footprint (GUF), which is also intended to be publicly provided at a spatial resolution of ~3.0 arcsec (i.e., ~50-75 m).

Monitoring of global urbanization-time series analyses for mega cities based on optical and SAR data

Mega cities and their development are a synonym for the steady and dynamic trend of global urbanization. The number of these extreme metropolises exploded from three mega cities in 1975 (Mexico City, New York, Tokyo) to actually 27 and it is expected that more than 100 new ones will emerge within the next three decades. However, particularly in developing countries the spatiotemporal characteristics of mega cities development are still not well known. Therefore, this study introduces the concept and results of a global monitoring of the spatiotemporal mega cities development based on Earth Observation (EO) data. We applied straightforward, semi-automated object-oriented and pixel-based image classification algorithms to high resolution (HR) optical (Landsat) and very high resolution (VHR) SAR (TerraSAR-X) imagery covering a time span of 40 years in an interval of 10 years. By mapping the extent of the urban area for each single decade, we could subsequently apply post-classification change detection methods to visualize the dimension, patterns and dynamics of urban sprawl for the current 27 mega cities. The results of this study demonstrate the capabilities and benefits of satellite-based EO to support the collection of data on global urbanization trends and patterns. The analyses also showed the high potential of VHR SAR data for a more detailed characterization of settlement patterns and urban morphology.