GIS Colloquium – Talks (Winter Term 2014/15)

On the relationship of space and content of volunteered geographic information

Stefan Hahmann – Thu, Oct 23, 2014, 4.15 pm, Hörsaal Berliner Straße 48

Within recent years, there has been a significant progress of the World Wide Web, which evolved to become the so-called Web 2.0. The most important feature of this new quality of the WWW is the participation of the users in generating content. This trend facilitates the formation of user communities which collaborate on diverse projects, where they collect and publish information. The portion of the resulting User-generated content, which is directly or indirectly geospatially referenced, is often termed more specifically volunteered geographic information (VGI). The existence of this type of information opens new opportunities for (GIScience) research. The first part of the talk addresses the question for which share of (not only geo-) information there exists a relationship between space and content of the information, such that the information is locatable in geospace. In this context, the assumption that about 80% of all information has a reference to space has been well known within the community of geographic information system users since the early 1980s. In the second part of the talk the question is investigated in how far volunteered geographic information that is produced on mobile devices is related to the locations where it is published. For this purpose, a collection of microblogging-texts produced on mobile devices has served as research corpus.

Pelagios – Linking Data about the Past through Geography

Rainer Simon, Leif Isaksen, Pau de Soto – Thu, Oct 30, 2014, 4.15 pm, Hörsaal Berliner Straße 48

Pelagios is a community initiative that aims to facilitate better linkage between online resources that document the past. Each of our member projects represents a different perspective on our shared history, whether expressed through text, image, map, or archaeological record. But as a group we believe passionately that the combination of all of our contributions is enormously more valuable than the sum of its parts. The medium through which Pelagios creates links between heterogenous data is geography. Using shared convetions to encode and publish place references contained in their data, our partners implicitly build up a network of documents connected to places, documents connected to documents, and places connected to places. Contributing to the wider ecosystem of the Graph of Humanities Data that is gathering pace in the Digital Humanities (linking data about people, places, events, canonical references, etc.), we believe that this will ultimately open up new avenues for computational and quantitative research in a variety of fields including History, Geography, Archaeology, Classics, Genealogy and Modern Languages. In its initial phases, Pelagios has had a specific focus on classical antiquity. That is, the known world at the time of Ancient Greece and the Roman Empire, around and beyond the Mediterranean and the Middle East. More recently, however, the project has also started to expand its geographical and temporal scope towards the European Middle Ages, as well as to the Early Islamic and Chinese world. Supported by funding from the Andrew W. Mellon Foundation and the Open Humanities Awards, we are currently working on the collaborative annotation of relevant Geographic Documents (texts and maps) from these eras. In this one-day workshop, we invite participants to join us in this process hands-on, using Open Source annotation tools developed by Pelagios, and to discuss how our tools and the resulting data can be used for further research, within and beyond the Pelagios project. Further information here and here.

Smartphones vs. LiDAR for agricultural monitoring: A Participatory Sensing approach to derive 3D Crop Height Models

Sabrina Marx – Thu, Nov 6, 2014, 4.15 pm, Hörsaal Berliner Straße 48

Since a growing number of people all over the world are carrying smartphones or tablets, data collection via geolocation-aware mobile applications has brought new opportunities for agricultural monitoring in low-resource settings. The integration of participatory sensing provides valuable information and could be beneficial to deepen the understanding of complex phenomena such as malnutrition and climate change in Sub-Saharan Africa. The strengths and limitations of participatory sensing for deriving 3D crop height models are pointed out in this talk. Students were instructed to collect the underlying geoinformation on plant properties (e.g. height, density) within a maize field in Heidelberg using smartphones and GPS devices. The results are benchmarked against highly-detailed information from a terrestrial laser scanner. Although this approach is limited e.g. due to location inaccuracies, the incorporation of human observation provides interesting perspectives for gathering agricultural data from smallholder farmers on a small-scale.

Algorithms for focus-and-context maps

Jan-Henrik Haunert – Tue, Nov 25, 2014, 4.30 pm, Seminarraum Berliner Straße 48

A focus-and-context map provides a user with detailed information about an area of interest while displaying context for orientation. In my talk I present algorithmic approaches to the automatic generation of focus-and-context maps from large-scale geospatial data. This includes an optimization-based distortion technique to enlarge a user-selected region in a map as well as algorithms for map generalization and map labeling that target navigation applications.

Big Earth Observation Databases: infrastructure and spatiotemporal analysis

Gilberto Camara – Thu, Nov 27, 2014, 4.15 pm, Hörsaal Berliner Straße 48

Current scientific methods for extracting information for Earth observation data lag far behind our capacity to build sophisticated satellites. These satellites produce massive amounts of data, but only a fraction of that data is effectively used for scientific research and operational applications. This presentation will address a key scientific problem: How can substantially improve the extraction of information from big Earth Observation data sets in an open and reproducible way? The presentation will focus on the innovative techniques of large-scale databases, that can hold thousands of images joined in space and time. Users are then able to develop algorithms that can seamlessly span partitions in space, time, and spectral dimensions. Array databases are specially well suited to be combined with spatiotemporal data analysis, which enable producing new types of land change information. In this work, we introduce the Fields data type as a possible way to represent large data sets. We propose a generic data type for fields that can represent different types of spatiotemporal data, such as trajectories, time series, remote sensing and, climate data. We show how to represent existing algebras for spatial data with the Fields data type. We also argue that array databases are the best support for processing big spatial data and show how to use the Fields data type with array databases.

Short bio: Dr. Gilberto Câmara is a Brazilian researcher in Geoinformatics, Spatial Databases, and Environmental Modelling at the Image Processing Division of Brazil's National Institute for Space Research (INPE). He is internationally recognized for promoting free access for geospatial data and for setting up an efficient satellite monitoring of the Brazilian Amazon rainforest. As recognition for his work, he was inducted as a Doctor honoris causa from the University of Münster (Germany) and as a Chevalier (Knight) of the Ordre National du Mérite of France. He received the Global Citizen Award of the Global Spatial Data Infrastructure Association. He is also a Fellow of the Faculty of Geoinformation Science and Earth Observation (ITC) in the Netherlands and a Senior Member of the Association for Computing Machinery (ACM).

Location Based Services (LBS) - Cases of navigation and wayfinding

Jukka Krisp – Thu, Dec 11, 2014, 4.15 pm, Hörsaal Berliner Straße 48

Personalized navigation and way-finding for outdoor and indoor environments are prominent research areas of Location-Based-Services (LBS). Current research cases include innovative concepts for car navigation and pedestrian navigation. We present the idea to provide drivers a routing suggestion, which avoids complicated crossings in urban areas. Based on the topological characteristics of the dataset, measured by the number of nodes, we classify crossings into a continuum reaching from easy to drive to difficult to drive. The user can choose to compute alternative routes. This methodology is one step in building a full inexperienced drivers routing system. A case study for indoor routing explores the situation within the Technical University Munich's main building. Currently available data on floor footprints, points of interests (POIs), indoor-landmarks, pictures and a routing graph for the building are used to investigate a routing service. A betweeness centrality measure can be used to compute a hierarchical indoor routing network. Unlike in car navigation, turn restrictions and one way roads are to some extent obsolete in an indoor environment. The betweenness measure has been used in other studies to investigate the importance of certain edges within a network.

Geo-Information Fusion: Gaining additional value for real-time Digital Earth applications

Florian Hillen – Thu, Dec 18, 2014, 4.15 pm, Hörsaal Berliner Straße 48

The Digital Earth vision by Al Gore recently has evolved to a powerful real-time toolbox for various use cases. Nowadays, almost every geo-sensor data can easily be integrated in a Digital Earth application in real-time and near real-time. This can be in-situ sensor data, smartphone sensor data or also high-resolution remote sensing imagery. However, the benefit of combining multiple data sources is only rarely exploited. Remote sensing data, for example, generally cover large areas but do not deliver information for hidden areas (e.g. under bridges, in house) or under cloud cover. In contrast to that, in-situ sensors deliver punctual information only but may provide information for areas that are invisible to remote sensors. Thus, the first idea that comes to mind is to use the advantages of the respective sensor types to eliminate the disadvantages of the other. The real-time aspect is a crucial point in this process, especially for time-critical applications like early warning systems, decision support systems for security issues or precision fertilisation for agricultural areas. To date, there is a lack of usage regarding real-time integration of fused geo-information even though the benefit is obvious.

Enhancing IR by Adding Geographical Information: Text- and Geo-based Search within a Corpus of Literary Texts

Bastian Entrup – Wed, Jan 28, 2015, 11.15 am, Hörsaal Berliner Straße 48

The eHumanitites Project GeoBib aims at building a georeferenced online bibliography of early Holocaust Literature (1933-1949). It includes Information on texts (bibliographical data) and on authors (biographical data). Furthermore, the places occurring in the texts are annotated and georeferenced. The planned website will provide means to search for persons, places, and texts. Texts, their authors and the places connected with persons and texts will be combined. This enables the user to find texts not only by defining restrictions on bibliographical information but also on the author’s biographical data. Since both texts and authors are georeferenced, spatial searches are possible: One will be able to find all texts containing reports on, e.g., Auschwitz, or all texts that take place around Berlin or all texts by authors from Heidelberg. This talk will show how the geo-data enhances the possibilities of IR in this Digital Humanities project and how georeferencing enables peoples to conduct new research in the humanities. Furthermore, the combination of text- and geo-based search queries will be explained.

Das EAGLE-Konzept – Modellentwurf zur semantischen Integration von Landbedeckungs- und Landnutzungsdaten im europäischen Kontext

Stephan Arnold – Thu, Feb 05, 2015, 4.15 pm, Hörsaal Berliner Straße 48

The current environmental challenges require the interconnection of ecological, economic and social factors at local to global scales. There is therefore a fundamental need to monitor these factors, their impact on land, their spatial distribution and changes over time in the form of land cover (LC) and land use (LU) observations. To work effectively across the required temporal and spatial scales these observations need to be modeled in a consistent and machine readable way. A broad variety of LC/LU classification systems have evolved over time in response to specific needs and available technology. Each application emphasizes different aspects of LC and LU and many have mixed LC and LU information. Incompatibility caused by variations in class definitions (semantic overlaps/gaps etc.) often hampers the exchange of data between different applications. The globalization of information on land requires harmonization, which so far was approached by spatial and thematic generalization resulting in coarsely aggregated data. Future tasks require a more differentiated and detailed description of landscape. Meanwhile, progress in the development of remote sensing and database technology has increased the methodological capabilities in the land monitoring domain and opened the way for a more sophisticated approach in land description. The EAGLE concept (developed by the Eionet Action Group on Land Monitoring in Europe) represents a data model that consistently separates LC and LU information by decomposing landscape into Land Cover Components, Land Use Attributes and further landscape Characteristics. The concept can be used (a) as a semantic translation tool, (b) for semantic ontological analysis of existing class definitions, (c) as a guideline for the design of classification systems or mapping activities. It can be considered as a conceptual proposal for a future European land monitoring framework building upon preceding achievements while integrating new possibilities of parameterized data storage and modelling.