Martin Klotz

Object-Based Morphological Profiles for Classification of Remote Sensing Imagery

Morphological operators (MOs) and their enhancements such as morphological profiles (MPs) are subject to a lively scientific contemplation since they are found to be beneficial for, for example, classification of very high spatial resolution panchromatic, multi-, and hyperspectral imagery. They account for spatial structures with differing magnitudes and, thus, provide a comprehensive multilevel description of an image. In this paper, we introduce the concept of object-based MPs (OMPs) to also encode shape-related, topological, and hierarchical properties of image objects in an exhaustive way. Thereby, we seek to benefit from the so-called object-based image analysis framework by partitioning the original image into objects with a segmentation algorithm on multiple scales. The obtained spatial entities (i.e., objects) are used to aggregate multiple sequences obtained with MOs according to statistical measures of central tendency. This strategy is followed to simultaneously preserve and characterize shape properties of objects and enable both the topological and hierarchical decompositions of an image with respect to the progressive application of MOs. Subsequently, supervised classification models are learned by considering this additionally encoded information. Experimental results are obtained with a random forest classifier with heuristically tuned hyperparameters and a wrapper-based feature selection scheme. We evaluated the results for two test sites of panchromatic WorldView-II imagery, which was acquired over an urban environment. In this setting, the proposed OMPs allow for significant improvements with respect to classification accuracy compared to standard MPs (i.e., obtained by paired sequences of erosion, dilation, opening, closing, opening by top-hat, and closing by top-hat operations).

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.

Spatio-temporal cross-city comparison using multi-sensoral remote sensing for Mexican cities

Our planet is more and more transforming into an urban world, in which the dynamics of urbanization have overcome the ability to govern cities. In a situation of uncoordinated urban growth, regional and urban planning lack technologies and methodologies to measure, monitor and analyze the spatio-temporal pattern of dynamic urban sprawl. This paper focuses on methods using remote sensing data to analyze, quantify and compare spatial urbanization processes. Urban sprawl is detected at the level of urban footprints using a post-classification change detection approach based on multi-sensoral Landsat and TerraSAR-X data. Spatio-temporal analysis combines absolute parameters (e.g. areal growth), location-based zonal statistics and gradient analysis (e. g. urban core versus the urban fringes) as well as spatial metrics (e.g. Largest Patch Index) to quantitatively characterise the spatial pattern of city developments. The study aims to detect spatial analogies as well as differences for the four largest Mexican urban agglomerations.

Digital deserts on the ground and from space

Slums are among the most visible manifestation of urban poverty. In this vein, earth observation (EO) has been widely accepted as a tool to approximate associated socioeconomic disparities at the city level. In this work, we explore the potential of a novel data source - location-based social networks - in conjunction with EO-based slum maps. Applying meaningful location quotients for spatial clustering of digital hot and cold spots in an experimental setting, we find that such data can add generalized spatial knowledge to space-based methods via the designation of less digitally-oriented population groups. Conversely, slums derived from remote sensing show substantial quantitative correspondence with clustering results, and thus, even enable to reflect underlying intra-urban socioeconomic characteristics.

Urbane Strukturen der Macht

Manhattens Geschaftsviertel, Downtown und Midtown, gelten bis heute als Inbegriff und Urbild von Central Business Districs (CBDs). Gerade die interne Struktur von Stadten hat sich im Zuge dynamischer Urbanisierungsprozesse im 20. und 21. Jahrhundert hin zu einer immer komplexeren Geographie funktionsraumlicher Zentren entwickelt. Business Districts bilden dabei Polarisationskerne der urbanen Entwicklung, die bei Weitem nicht mehr nur in zentralen Lagen zu finden sind. Bislang sind deren Anordnung und Lage auf der Skala groser Millionen- und Megacities aber nur schwer zu erfassen. Der Beitrag zeigt exemplarisch, welches Potential die physische Perspektive der Erdbeobachtung zur Kartierung und Untersuchung dieser funktionsraumlichen Strukturen auf verschiedenen raumlichen Skalen besitzt. Dabei ermoglicht eine kleinraumige Untersuchung die Charakterisierung des physischen Erscheinungsbildes dieser Zentren (Morphologie) sowie eine grosflachige Kartierung die Analyse ihrer Verteilung im Raum (Geographie). In den dargelegten Beispielen wird dabei deutlich, dass Business Districts keineswegs einen homogenen physischen Strukturtyp darstellen und dass die raumliche Verteilung von BDs in der kurzlich dynamisch gewachsen Megacity Istanbul polyzentralere Raummuster messbar sind als in den Megacities Paris und London.

Delimiting central business districts — A physical approach using remote sensing

Central Business Districts (CBDs) are an apparent structural type of large cities. Although the conceptual definition of CBDs is mostly functional, this urban structure type has characteristic physical features. The paper presents a conceptual framework to map these CBDs using physical parameters. We identify physical parameters from the published literature and statistically designate the CBD from a 3-D city model by the example of the test site La Defense, Paris, France. From it, we develop a method to detect CBDs from a combination of large-area Cartosat-1 high resolution digital surface models for the entire spatial extent of mega city Paris. Accuracy assessment shows that CBDs are detected with an accuracy of 83.3% and are spatially delineated with an overall accuracy of 83.7%.

Validation of the DLR Global Urban Footprint in rural areas: A case study for Burkina Faso

Earth observation from space has often been used for the mapping of settlements on a global scale. The recently developed Global Urban Footprint (GUF - 2011/2013) aims at a global settlement classification with an improved geometric resolution based on data from the TerraSAR-X/TanDEM-X missions. Due to manifold challenges the settlement classifications are subject to distinct accuracy variations across global landscapes. This study presents a systematic accuracy assessment of the GUF product for the case study of Burkina Faso, an African country characterized by mainly rural, small-scale and fragmented settlement structures. As a first step, we conduct a relative, non site-specific comparison of available national and global human settlement layers (HSL) on national scale by the quantification of total settlement areas and derived density measures. In a second step, we aim at an absolute accuracy assessment on local scale using appropriate reference data. Therefore we use absolute accuracy measures based on the error matrix as well as pattern-based evaluation techniques with regard to physical settlement variations such as settlement size. Results clearly show the enhanced mapping capabilities of new high resolution global settlement products such as the GUF, especially for rural areas.

Der Werkzeugkasten der urbanen Fernerkundung – Daten und Produkte

Stadte sind im Laufe des „Anthropozans“ zum primaren Lebensraum des Menschen geworden. Viele stadtgeographische Fragestellungen, die sich mit den sozialen, okologischen und okonomischen Aspekten unseres Zusammenlebens in Stadten befassen, haben einen raumlichen Bezug. Sie bedingen daher quantitatives Wissen − Wissen, das uber die Fernerkundung als Geoinformation verfugbar wird. Der Beitrag gibt einen systematischen Uberblick uber die Vielfalt lokalr bis globaler fernerkundlicher Daten und Produkte im urbanen Kontext sowie deren technischen Eigenschaften und Funktionalitaten. Dieser Uberblick uber den gegenwartigen Werkezugkasten zur raumbezogenen Stadtforschung wird erganzt durch einen Ausblick auf neue Satellitenmissionen in naher Zukunft, die eine noch breitere Palette qualitativ hochwertiger Werkzeuge bereit stellen werden.