Theses

Bachelor

ifgicopterUAV / UAS (Drohnen) Remote Sensing/GIS: Vegetationsspezifische Geodatenanalyse/Workflows

Thema: Im Rahmen der gemeinsamen IFGIcopter und ILÖK UAV Initiative werden  kontinuierlich vegetationsspezifische Fernerkundungsdaten unterschiedlichster UAV-Sensoren (Drohnen) aufgenommen und ausgewertet. Besondere Schwerpunkte sind die Erfassung und Analyse von Vegetationsmustern, Vitalitätsparametern und inversiver Arten mittels multispektraler UAS Daten. In diesem Kontext spielen die Datenverarbeitung und Visualisierung (auch 3D) mittels verschiedenster geoinformatischer Werkzeuge (GIS, kommerzielle Software, Web-Tools und Eigenprogrammierungen etc.) eine große Rolle. Wer Interesse an einer interdisziplinären Fragestellung in diesem Bereich hat, wende sich an die beiden Ansprechpartner [2017].

Ansprechpartner: Torsten Prinz / Jan Lehmann

Contact: Torsten Prinz

ifgicopterCrowd Managment, Geodaten Fusion und Dronen-Technologie: Möglichkeiten und Perspektiven

Thema: Im Rahmen von aktuellen IFGIcopter-Aktivitäten bietet das Spannungsfeld Drohnen/UAV und Geodatenfusion' interessante Schnittstellen hinsichtlich geoinformatischer Datenanalyse und Werkzeugentwicklung. Fernerkundungsdaten unterschiedlichster UAV-Sensoren können z.B. unterschiedlichste zeitabhängige Momentaufnahmen von 'Geo-Objektverlagerungen' erfassen und mittels Fusion mit Sekundärdaten zu Modellen einer dynamischen Entwicklung führen. In diesem Kontext spielen die Datenverarbeitung und Visualisierung (auch 3D) mittels verschiedenster geoinformatischer Werkzeuge (GIS, kommerzielle Software, Web-Tools und Eigenprogrammierungen etc.) eine große Rolle. Wer Interesse an einer interdisziplinären Fragestellung in diesem Bereich hat, wende sich an die beiden Ansprechpartner [2017].

Ansprechpartner: Torsten Prinz / Florian Hillen

Contact: Torsten Prinz

STMLEntwicklung einer Python Schnittstelle zur raumzeitlichen Analyse von Erdbeobachtungsdaten mit gdalcubes

gdalcubes ist eine am ifgi entwickelte C++ Bibliothek zur effizienten Erstellung und Verarbeitung von raumzeitlichen Data-Cubes aus größeren Satellitenbildsammlungen. Zur einfacheren Nutzbarkeit durch Datenwissenschaftler existiert zur Zeit lediglich eine Schnitstelle als R Paket. Ziel dieser Arbeit ist es, eine prototypische Schnittstelle zwischen gdalcubes und Python zu entwickeln. Dabei sollen verschiedene existierende Ansätze zur Verküpfung von Python und C++ (zum Beispiel pybind11, Cython, SWIG, Boost.Python) evaluiert werden. Schwerpunkte der Arbeit können nach Interesse besprochen werden.

Contact: Marius Appel

SITCOMExploring forms of interactions in the Immersive Video Environment (IVE)

 

IVE is a panoramic video footage that is displayed on large screens in a cave like environment that creates a sense of physical presence and enables people to better interact and intervene the image of their surroundings. However, the forms of interactions are very limited - yet needed - when users need to create objects on the screen as video overlays and interact with them (e.g., adding and scaling a tree, modifying a building façade etc.).

 

This study aims to explore the forms of interactions on IVE for effective creation and interaction of the overlays. The student will work with the IVE system in the Sitcom Lab at Ifgi. The study will cover the following steps: creating a video footage and overlays for IVE, exploration of the tools (e.g., smart phone, HTC Vive controllers, touch pad), design of the forms (e.g., adding, removing, scaling, rotating, placing the overlays) of interaction and the UI component.

 

Contact: Simge Özdal Oktay

SIIRich Data Interface for Copernicus Data

Copernicus is a European Earth observation programme with the goal “to provide accurate, timely and easily accessible information to improve the management of the environment, understand and mitigate the effects of climate change and en-sure civil security” (ESA 2015). Within Copernicus, a series of new satellite mis-sions will be executed to gather new satellite data. The first mission acquiring ra-dar data, called Sentinel-1, and the second mission gathering multi-spectral high-resolution data have already been launched.
In order to derive higher-level information products, the data needs to be pro-cessed, e.g. by using segmentation algorithms. In former times, the workflow in-cluded downloading the data, pre-processing it by selecting the spatial and/or temporal extent, and then running the actual segmentation process. The idea of rich data interfaces is that these processing facillities for deriving new information products are directly provided with the data in order to avoid downloading large raw datasets.
The aim of this bachelor thesis is to investigate how the OGC Web Processing Service (WPS) standard (Mueller & Pross, 2015), which defines a web service in-terface for geoprocessing functionality, can be utilized for such a rich data inter-face for the Copernicus. The expected outcomes are
(i) architecture patterns for coupling WPSs with copernicus data interfaces (ii) a prototypical implementation (iii) an evaluation of the concepts developed.
In order to successfully work on the thesis, good knowledge in Web technologies and Java programming is required as well as interest in remote sensing and EO-analysis. The work will be done in cooperation with the 52°North Open Source ini-tiative.
Literature:
Mueller, Matthias, and B. Pross. “OGC WPS 2.0 Interface Standard”. OpenGIS Implementation Standard, Version 2.0, OGC 14-065. (2014).
ESA. “Overview/Copernicus”. Available online at http://www.esa.int/Our_Activities/Observing_the_Earth/Copernicus/Overview3. (2015) Accessed 29/05/2015.
Contact:
Prof. Dr. Albert Remke (a.remke@52north.org) Dr. Christoph Stasch (c.stasch@52north.org)

Contact: Prof. Dr. Albert Remke

ifgicopterRe-Design des Geodatenportals StudMap14

Das ZDM/IVV sucht ab sofort in Zusammenarbeit mit dem IFGI eine/n BSc-Kandidaten/in aus der Geoinformatikzwecks innovativen "Re-Designs" des Geodatenportals StudMap14 (http://gdione4all.uni-muenster.de/joomla/index.php/studmap14

Kenntnisse/Einarbeitung in die GeoServer-Umgebung und Interesse an modernen GDI-Lösungen sind Voraussetzungen für dieses BSc-Projekt (ggf. ist eine Finanzierung mit 5 SHK-Stunden für 6 Monate möglich).

Bei Interesse direkt mit Dr. Torsten Prinz in Kontakt treten!

Contact: Torsten Prinz

SILEntwicklung eines Statistikportals zur Visualisierung und Auswertung von Umweltdaten aus dem Citizen Science Kontext

Die openSenseMap ist eine Plattform für Umweltsensordaten von Messstationen jeglicher Art. Zur Zeit werden nur Rohdaten von senseBoxen gespeichert und die Daten können sich nur pro senseBox angezeigt werden lassen. Zudem gibt es die Möglichkeit sich die gesammelten Daten für einen Zeitpunkt interpoliert darstellen zu lassen.

Ziel der Arbeit ist es, für die openSenseMap ein Portal zu entwicklen, in dem der Benutzer die Möglichkeit hat mehrere senseBoxen und Sensoren mit statistischen Methoden zu vergleichen und externe Datenquellen, wie zB. vom DWD, einzubinden.

Contact: Thomas Bartoschek

SILGeovisualisierung von offenen Umweltdaten im Web

Die openSenseMap bietet live Daten zu verschiedensten Umweltphänomenen, Jedoch ist es zur Zeit schwierig diese Daten erkunden. Ziel dieser Bachelor Arbeit wäre es neue Möglichkeiten zu schaffen, die Daten interaktiv darzustellen. Interessant wären zum Beispiel live Interpolationen über Feinstaubwerte oder die Temperaturentwicklung in Innenstädten im Hochsommer. Um diese Daten einem möglichst grossem Publikum zur Verfügung zu stellen, soll in dieser Bachelorarbeit untersucht werden, welche Möglichkeiten hier neuste Webtechnologien bieten. Verschiedene Visualisierungen sollen generiert werden und mit einer Nutzerstudie evaluiert werden.  

 

 

Contact: Thomas Bartoschek

SILQualitätssicherung von crowd-sourced Sensordaten

In der Arbeit soll untersucht werden, inwieweit Qualitätssicherung von crowd-sourced Sensordaten in einem Sensornetzwerk automatisierbar ist. Dies ist ein neues und hoch relevantes Forschungsfeld: große Datenmengen erlauben die Anwendung statistischer oder machine learning-Verfahren. Traditionelle Verfahren sind häufig nicht nutzbar, da die Daten in Echtzeit vorliegen müssen. Zudem stellen crowd-sourced Daten eine spezielle Herausforderungen dar, da nicht davon ausgegangen werden kann, dass alle Daten mit korrekten bzw. konsistenten Messverfahren erhoben wurden. Schließlich haben low-cost-Sensoren selbst Messfehler, die von professionellen Sensoren stark abweichen. oder Messstationen sind von Citizen Scientists schlecht montiert. Das Ziel ist, die Einflussfaktoren auf die Datenqualität und die Messgenauigkeit der Sensoren zu erforschen, Verfahren zur automatisierten Identifikation fehlerhafter Daten und möglicher Fehlerquellen zu entwickeln sowie automatisiert Entscheidungen über Möglichkeit zur Korrektur der Daten (bspw. über Nachkalibrierung der Sensoren) oder Ausschluss bestimmter Daten zu treffen.

Contact: Thomas Bartoschek

SILMeasuring Data Quality of Air Quality Measurements in Citizen Science

Im Kontext Citizen Science lassen sich mittelfristig räumlich hochaufgelöste Sensornetzwerke aufbauen um gesellschaftlich relevante Fragestellungen zu untersuchen. Die openSenseMap ist eine öffentlich zugängliche Plattform für das Veröffentlichen, Teilen und Visualsieren von solchen Umweltmessdaten als Open Data. Um eine flächendeckende Auflösung zu erreichen, werden die verwendeten Messstationen durch Freiwillige zusammengebaut, programmiert und installiert. Luftqualitätsmessungen nehmen hierbei eine gesonderte Rolle ein, da der Zugang zu teils komplizierter Messtechnik durch modulare Open-Source Bausätze, wie denen der senseBox, stark vereinfacht und einer breiten Masse zugänglich gemacht werden kann. In den meisten Fällen sind Freiwillige bei einer solchen Kampagne nicht wissenschaftlich ausbildet. Dadurch kann bei dieser Methodevnicht garantiert werden dass Messwerte von unterschiedlichen Stationen im Netzwerk direkt vergleichbar sind. Eine Frage die im Rahmen dieser Ausschreibung ausarbeitet werden soll ist ob trotz öffentlicher Beteiligung im Rahmen einer Messkampange in den Bürgerwissenschaften verwertbare Informationen und representative Analysen gegeben werden können. Als Anwendungsfall soll die Aussagekraft und Datenqualität von Gasmessungen (Stickoxid und Ozon) in Berlin auf der openSenseMap untersucht werden. 

Contact: Thomas Bartoschek

SILEntwicklung und Evaluierung von dynamischen Lerntutorials mit Gamification-Ansatz

Die graphische Programmieroberfläche Blockly (blockly.sensebox.de) für senseBox bietet die Möglichkeit auch ohne Programmiererfahrungen schnell und einfach in die Welt der Mikrokontroller-Programmierung einzusteigen. Im nächsten Schritt sollen interaktive Tutorials entwickelt werden, die direkt in der Blockly Oberfläche absolviert werden können. Ein Reward System basierend auf "Open Badges" (https://openbadges.org/) soll integriert werden. Ein Beispiel wie eine solche Integration aussehen könnte findet sich hier: https://studio.code.org/hoc/1
 
 

Contact: Thomas Bartoschek

SILLearning Analytics für das Programmierenlernen mit Blockly

Learning Analytics ist eine Methode zur Erfassung, Messung, Analyse und Visualisierung von Daten über Lernende und ihren Kontext, die es ermöglicht den Lernfortschritt zu verstehen und Lernpfade zu optimieren, um dem Lernenden aber auch Lehrenden Feedback zum Lernprozess zu geben. Learning Analytics lässt sich besonders gut in digitale Lernplattformen einbinden. Blockly für die senseBox ermöglicht es Anfängern in die Mikrocontrollerprogrammierung einzusteigen.Trotzdem können Fehler im "Code" entstehen, wodurch das Programm nicht kompilierbar ist. 

Ziel der Arbeit ist die Entwicklung einer Learning Analytics Komponente, die dem Nutzer Feedback zum selbst programmierten Code gibt, Hinweise bei Fehlern und Lernfortschritte trackt.

 

Contact: Thomas Bartoschek

SILSpatial Learning Analytics

Learning Analytics is a method to collect, measure, analyze and visualize data about learners and their context. It enables the understaning of the learning process and allows an adaption of learning pathes based on the collected data. It also gives feedback to the learner and teacher about the learning process.

OriGami is a GeoGame fostering spatial literacy: The player has to solve several wayfinding tasks to various locations and answer questions at these locations. At the current state, it is a single-player game.
 
In this thesis OriGami (as a WebApp) needs to be transferred to a native app (based on previous work in the Ionic framework) and extended with spatial learning analytics where the movement data of the player is taken into account. 

Contact: Thomas Bartoschek

SILUser Centered Design for educational WebGIS

Mit WebGIS NRW (webgis.nrw) existiert ein prototypisches WebGIS für den Bildungskontext, das auf modernen open Source Technoligien basiert (MapBox GL). Ziel der Arbeit ist die Weiterentwicklung des WebGIS nach User Centered Design Prinzipien und eine Evaluation der Usability.

Contact: Thomas Bartoschek

SILExplorative Analyse von Open Source Hardware Sensorik

Im Rahmen dieser Bachelorarbeit sollen neue Sensorkomponenten für Umweltphänomene (z.B. Wind, Wasser, Radioaktivität o.ä.) für die senseBox identifiziert und in das senseBox Ökosystem aus Open Source Hardware, openSenseMap Geodateninfrastruktur, Blockly-Programmierumgebung integriert und evaluiert werden. 

Contact: Thomas Bartoschek

SILGeneralisation in Sketch Maps

When people draw sketch maps, they generalise information compared to the ground-truth information they perceived in the world. For example:

- many buildings belonging to university campus are drawn as a single polygon labelled "campus",

- a complex roundabout is drawn as a few crossing lines,

- 10 side streets from the main road are drawn as a few random lines, only to indicate that "some side streets" are there. 

This is a challenge for analysing sketch maps because this information is not wrong, yet a computer system for automated analysis would interpret it as such.

In the paper linked below we presented a classification of generalisation types in sketch maps, and many other examples. We also also have a working software prototype for analysing generalisation in sketch maps.

This thesis topic can be concerned with the software implementation of different algorithms for detecting generalised spatial information (e.g., in a comparison between a sketch map and OpenStreetMap data), and/or with theoretical studies of how people generalise information.

Examples of potential research questions:

- What affects the level of specific generalisations, e.g., when people perform stronger vs weaker forms of "amalgamation" generalisation?

- What are common ways to schematise complex spatial structures (e.g., roundabout), and how people do it spontaneously in drawings?

- How to detect and algoirthmically differentiate the >lack of some object< (e.g., a missing streets in an area) from their generalisation?

Manivannan, C., Krukar, J., & Schwering, A. (2022). Spatial generalization in sketch maps: A systematic classification. Journal of Environmental Psychology, 101851. https://doi.org/10.1016/j.jenvp.2022.101851

 

Contact: Jakub Krukar

SILSketch Maps as a tool for learning new environments

For decades, sketchmaps have been used as a tool for measuring spatial knowledge - i.e., for estimating how well participants know and understand some areas. However, evidence from psychological memory studies demonstrates, that drawing something can also be a good strategy to memorise a set of object. For instance, if you need to memorise the setting of a room, drawing the room as you see it is a better memorisation strategy than repeating the names of the objects verbally or in your head. This thesis will test whether drawing a sketch map is a good memorisation strategy for spatial environments and how this approach can be implemented in a gamified app. The problem is relevant for situations in which people must learn new spatial environments, e.g. to become taxi/delivery drivers, or when they move to a new city.
The thesis can be completed with focus on one of two aspects:
**Computational focus:** You will design a teaching app that (a) records the user's trajectory together with a list of landmarks that were visible along the route, and (b) after a delay, asks users to draw the area that they have travelled. Here the key problem may be to select routes and landmarks that the user should be asked to draw (based on the recorded trajectories).
**Evaluation focus:** You will design and conduct an experiment to evaluate the following research question: does drawing a sketchmap help people memorise the environment better, compared to alternative strategies? This does not require creating an app, and can be conducted as an in-situ experiment, or inside our Virtual Reality lab.

Contact: Jakub Krukar

SILSupporting Land Rights Management through Digital Mobile Maps

Despite substantial progress in digital technologies, and the increasing adoption of digital technologies for the collection of land-related data (see e.g. ArcGIS Field Maps, Qfield, or MAST), information related to land (e.g. citizen-to-city-council-communication, purchase transaction, inheritance transactions) is still managed manually. This is a missed opportunity and this thesis will use a Research through Design (RtD) approach to explore possible futures of digital land rights management. RtD is a “type of research practice focused on improving the world by making new things that disrupt, complicate or transform the current state of the world” (Zimmerman and Forlizzi 2014). Tasks include:  

 

Requirements specification: derive requirements for applications for map-based land rights data management through scenario formulation (for an example of scenario-based study, see Degbelo and Somaskantharajan, 2020)

 

Implementation: develop a mobile app (frontend + backend) that implements some of the requirements. The mobile app may be a simple responsive web app or use a framework for mobile app development

Evaluation: validation of design decisions of the mobile app through interviews with domain experts and/or additional users. 

 

References

 

Degbelo, A. and Somaskantharajan, S. (2020) ‘Speech-based interaction for map editing on mobile devices: a scenario-based study’, in Alt, F., Schneegass, S., and Hornecker, E. (eds) Mensch und Computer 2020. Magdeburg, Germany: ACM, pp. 343–347. doi: 10.1145/3404983.3409996.

 

Zimmerman, J. and Forlizzi, J. (2014) ‘Research through design in HCI’, in Olson, J. S. and Kellogg, W. A. (eds) Ways of Knowing in HCI. New York, New York, USA: Springer New York, pp. 167–189. doi: 10.1007/978-1-4939-0378-8_8.

 

Links

ArcGIS Field Maps: https://www.esri.com/de-de/arcgis/products/arcgis-field-maps/overview 

Qfield: https://qfield.org/ 

MAST: https://www.land-links.org/tool-resource/mast-technology-2/

Contact: Auriol Degbelo

SILDisambiguation questions for spatial search sessions

Current search engines return results in almost a fraction of a second to any query submitted by users. This often comes with the drawback of users navigating through a number of irrelevant results. This thesis will explore the impact of adding query disambiguation steps of overall user satisfaction during search, and investigate the extent to which users are willing to trade response time with quality of results. Tasks envisioned include:

 

1. Data preparation

 

2. Development of a prototypical search engine, which (a) asks users which place they mean when homonyms are available (e.g. Berlin, USA vs Berlin, Germany), (b) offer disambiguation functionalities before the display of results using e.g. the Geonames API 

 

3. Development of a prototypical search engine, which simulates the functionalities of the current state of the art offered by search engines


The evaluation will involve A/B testing, and gathering of participants’ feedback during a user study.

 

Contact: Auriol Degbelo

SITCOMApps zur Förderung nachhaltigen Verhaltens

Eine Vielzahl verfügbarer Apps soll ihre Nutzer*innen dabei unterstützen, sich auf nachhaltigere Art zu verhalten.  Dies kann zum Beispiel das Mobilitätsverhalten betreffen (Rad statt Auto), den Energiekonsum (Pulli statt Heizung) oder auch Aktivitäten (Mitwirkung bei Urban Gardening).

Ziel dieser Bachelorarbeit ist es, eine repräsentative Auswahl dieser Apps systematisch zu analysieren, um einen Überblick über die verschiedenen Arten von Apps, der Hauptthemen und -strategien zu gewinnen. Von Interesse sind insbesondere auch welche Techniken aus dem Bereich "Nudging" und "Persuasive Technology" zum Einsatz kommen.

Das Thema kann auf unterschiedliche Weise bearbeitet werden, je nach Wunsch der bearbeitenden Person, z.B. als reine Analyse basierend auf detaillierter Betrachtung der Apps, als Nutzerstudie oder einer Kombination aus beidem. Denkbar ist auch der Entwurf einer App als Mockup/Prototyp, die bestehende Schwächen von verfügbaren Apps überwindet. 

Contact: Christian Kray

SILSummary of Graphical Annotations produced during Spatial Analysis

Annotations of visualizations is a useful task during exploratory data analysis. As illustrated in [1,2], users may want to record trends, clusters, distributions or outliers found on components of a geovisualization during a data analysis session. A key challenge here is that of summarizing the annotations produced by various users during their exploration activities. This thesis will look into computational and/or graphical means of producing these summaries. Tasks include: 

Prototyping: design of a web-based or mobile prototype to record annotations of web maps graphically.

Data collection: here graphical annotation of web-maps will be generated. These annotations could be synthetic or real, depending on the scope of the thesis. 

Summarization: this design, implementation and evaluation of summary approaches. 

 

References

[1] Heer, J., Viégas, F. B. and Wattenberg, M. (2009) ‘Voyagers and voyeurs: supporting asynchronous collaborative visualization’, Communications of the ACM, 52(1), pp. 87–97. doi: 10.1145/1435417.1435439.

[2] Lai, P.-C. and Degbelo, A. (2021) ‘A comparative study of typing and speech for map metadata creation’, in Partsinevelos, P., Kyriakidis, P., and Kavouras, M. (eds) Proceedings of the 24th AGILE Conference on Geographic Information Science (AGILE 2021), pp. 1–12. doi: 10.5194/agile-giss-2-7-2021.

Contact: Auriol Degbelo


Master

ifgicopterUAV / UAS (Drohnen) Remote Sensing/GIS: Vegetationsspezifische Geodatenanalyse/Workflows

Thema: Im Rahmen der gemeinsamen IFGIcopter und ILÖK UAV Initiative werden  kontinuierlich vegetationsspezifische Fernerkundungsdaten unterschiedlichster UAV-Sensoren (Drohnen) aufgenommen und ausgewertet. Besondere Schwerpunkte sind die Erfassung und Analyse von Vegetationsmustern, Vitalitätsparametern und inversiver Arten mittels multispektraler UAS Daten. In diesem Kontext spielen die Datenverarbeitung und Visualisierung (auch 3D) mittels verschiedenster geoinformatischer Werkzeuge (GIS, kommerzielle Software, Web-Tools und Eigenprogrammierungen etc.) eine große Rolle. Wer Interesse an einer interdisziplinären Fragestellung in diesem Bereich hat, wende sich an die beiden Ansprechpartner [2017].

Ansprechpartner: Torsten Prinz / Jan Lehmann

Contact: Torsten Prinz

ifgicopterCrowd Managment, Geodaten Fusion und Dronen-Technologie: Möglichkeiten und Perspektiven

Thema: Im Rahmen von aktuellen IFGIcopter-Aktivitäten bietet das Spannungsfeld Drohnen/UAV und Geodatenfusion' interessante Schnittstellen hinsichtlich geoinformatischer Datenanalyse und Werkzeugentwicklung. Fernerkundungsdaten unterschiedlichster UAV-Sensoren können z.B. unterschiedlichste zeitabhängige Momentaufnahmen von 'Geo-Objektverlagerungen' erfassen und mittels Fusion mit Sekundärdaten zu Modellen einer dynamischen Entwicklung führen. In diesem Kontext spielen die Datenverarbeitung und Visualisierung (auch 3D) mittels verschiedenster geoinformatischer Werkzeuge (GIS, kommerzielle Software, Web-Tools und Eigenprogrammierungen etc.) eine große Rolle. Wer Interesse an einer interdisziplinären Fragestellung in diesem Bereich hat, wende sich an die beiden Ansprechpartner [2017].

Ansprechpartner: Torsten Prinz / Florian Hillen

Contact: Torsten Prinz

STMLMachine Learning for the Automated Classification of Biotope Types in Remote Sensing Images

The mapping of biotope types plays an essential role in the assessment and monitoring of ecosystems. Since an extensive collection of the required parameters in the field is not feasible, remote sensing (mainly high-resolution RGB and NIR data) is commonly used. Usually the data is analyzed manually through visual interpretation by trained analysts. The aim of this project is to develop and evaluate methods for an automated segmentation (delineation) and classification of relevant biotope types at the spatial scale of established classification schemes.

Guiding questions include:

  • What are the identifiable features (e.g. spatial, spectral, temporal patterns) that are commonly used in the visual interpretation of remote sensing images to detect and classify biotope types, and can these features be integrated in models for an automatic classification?
  • How can deep learning and “classical” machine learning support the automation of classifying biotope types in remote sensing images?
  • What are the main features that are learned by well-trained machine learning models for identifying biotope types as compared to the features used in the visual interpretation by human analysts?
  • Which biotope types can or cannot be reliably classified, what are the main difficulties and which additional data might be helpful in increasing classification accuracy?

The data basis consists of high-resolution remote sensing images (10 cm, color infrared and RGB) as well as freely available medium resolution satellite data (especially Sentinel). A comprehensive reference data set of manually digitized biotope types is available for model training and validation. The practical work is expected to be conducted reproducibly in R.  

The research will be conducted through two master theses (geoinformatics and landscape ecology) in close collaboration. The specific tasks and research questions of both theses will be discussed between the students and supervisors.

Supervisors: Christian Knoth (ifgi), Hanna Meyer and Jan Lehman (ILÖK),   Reinhard Silvers (Hansa Luftbild)

Contact: Christian Knoth

SITCOMAnalysing and mapping emotions for noise quality

Sustainability is a broad concept with various interconnecting aspects. As one of them citizens’ feelings and perceptions of environmenment is usually neglected in practice due to the insufficient tools and expertise. However, the improvements in the citizen science and the possibilities provided by the digital visualisation techniques allow a better understanding of people’s emotions towards existing circumstances. This knowledge leads to more accurate assessment of sustainability and better decisions at the local scale.

The thesis aims addressing the local emotional indicators for noise quality in Münster through analysing and spatio-temporal modelling of dynamic emotions.  To achieve this aim, the study suggests to replicate selected locations in Münster at different time frames by using the Immersive Video Environment (IVE) and work with the citizens to collect information about their emotions. The student is free to choose the software and language for the analysis and spatio-temporal modelling. Basic knowledge of working with Unity and one of the programming languages will be an asset for this study.

Suggested readings:

Kals, E., Maes, J. (2002). Sustainable Development and Emotions. In: Schmuck, P., Schultz, W.P. (eds) Psychology of Sustainable Development. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-0995-0_6Kals, E., Maes, J. (2002). Sustainable Development and Emotions. In: Schmuck, P., Schultz, W.P. (eds) Psychology of Sustainable Development. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-0995-0_6

Murphy, Enda and Eoin A. King. “Mapping for sustainability: environmental noise and the city.” (2013).

Contact: Simge Özdal Oktay

SITCOMExploring forms of interactions in the Immersive Video Environment (IVE)

 

IVE is a panoramic video footage that is displayed on large screens in a cave like environment that creates a sense of physical presence and enables people to better interact and intervene the image of their surroundings. However, the forms of interactions are very limited - yet needed - when users need to create objects on the screen as video overlays and interact with them (e.g., adding and scaling a tree, modifying a building façade etc.).

 

This study aims to explore the forms of interactions on IVE for effective creation and interaction of the overlays. The student will work with the IVE system in the Sitcom Lab at Ifgi. The study will cover the following steps: creating a video footage and overlays for IVE, exploration of the tools (e.g., smart phone, HTC Vive controllers, touch pad), design of the forms (e.g., adding, removing, scaling, rotating, placing the overlays) of interaction and the UI component.

 

Contact: Simge Özdal Oktay

SITCOMDesigning a mobility dashboard for Zero Carbon Münster City

 

Urban dashboards are decision support platforms that allow its users to create and occasionally share knowledge about their towns and cities. These dashboards can be broad in the content or specific to better understand the status and the history of an urban issue such as sustainable mobility.

 

In 2022 a cycling dashboard has been developed as a prototype based on the sustainability indicators of cycling infrastructure defined by the German sustainable building council (DGNB) and open data from OSM. This thesis aiming to improve the content of the dashboard including air quality and energy consumption data, as well as expanding the mobility data. The data will be collected based on the related DGNB indicators. The open data platforms need to be realized and evaluated in order to improve the existing algorithm. At least intermediate knowledge of programming is an asset.

 

References:

 

https://static.dgnb.de/fileadmin/dgnb-system/de/quartiere/kriterien/DGNB-Kriterienkatalog-Quartiere-Kommentierungsversion-2020.pdf

 

https://www.dgnb-system.de/de/quartiere/kriterien/index.php

Contact: Simge Özdal Oktay

SILUsing VR to create self-adjusting buildings based on spatio-temporal data of their occupants

Buildings of the future will have to be much more flexible than they are now. One envisioned possibility is that building interiors will change their shapes depending on the current context of use, personal preference of their users, or tasks that the occupants have to perform within them at the given moment. While this may sound like a distant vision of the future, Virtual Reality equipment already allows us to study such scenarios today.


In this thesis, you will design a Virtual Reality building that participants will explore in Head-Mounted Displays. The VR system will monitor spatio-temporal data of the building user, and create the remaining (yet unvisited) parts of the building in response to this data, before the user gets there.

The specific context of this thesis can be adjusted based on your interests. One possibility would be to detect navigational confusion based on the occupant's walking trajectory, and - in response - provide a navigationally simplified space in the next room that the occupant is going to visit. Another possibility is to detect loss of attention in a virtual museum gallery, and - in response - provide the user with a more exciting space in the next room. The application should be evaluated in a simple user study.

Contact: Jakub Krukar and Chris Kray

GeoSimInvasive plant species modelling with cellular automata

Ecosystem services can be altered dramatically when the ecosystem is invaded by invasive plant species. Such species often facilitate their own invasion through a change of the local ecosystem conditions around them that is beneficial for their spread. This is called a self-reinforcing feedback effect. To avoid undesirable ecosystem shifts, management strategies aimed at stopping the invasion have to be developed in an early stage. To this purpose, one needs: 1) information on the current distribution of the invasive species, and 2) projections of the expected spread of the species under different future conditions.
 
Dr. André Groẞe-Stoltenberg has done his PhD on the detection and impact of the invasive species Acacia longifolia (Andrews) in a Mediterranean dune system in Portugal. A. longifolia is an N2-fixing woody plant, which increases the nitrogen level in its surroundings in the originally nutrient-poor dune system. Dr. André Groẞe-Stoltenberg has developed methods to distinguish A. longifolia from other species, using field spectroscopy and remote sensing techniques. Furthermore, he has applied spatial statistics to identify the factors that are of importance for the change in the local ecosystem conditions. Although this work has delivered many valuable insights in the spatial distribution of A. longifolia and the factors that might influence the spread, it did not yet deliver a simulation model that can be used to make future projections of the invasion under different conditions.
 
A cellular automaton (CA) is a model for simulating discrete changes over space and time.  A CA consists of: a grid of cells, a neighborhood definition, a finite set of discrete states, a finite set of transition rules, an initial state, and discrete time. The unique property of a CA is that the state of a cell at time t is a function of only the states of the cell itself and its neighbors at time t-1. Because of this property, cellular automata are suitable for modelling systems in which discrete entities (such as plants) spread by means of neighborhood effects (such as seed dispersal and self-reinforcing feedbacks).
 
The aim of this thesis is to develop a CA to project the past and future spread of A. longifolia in the described dune ecosystem in Portugal. You can build upon existing literature on cellular automata for vegetation modelling in arid Mediterranean ecosystems (e.g. by Sonia Kéfi and co-authors) and you can use the maps classified by Dr. André Groẞe-Stoltenberg. 
 
This topic is only for students who have followed Geosimulation Modelling or a similar course.
 

Contact: Judith Verstegen

SILDeveloping a Collaborative GeoGame

OriGami is a GeoGame fostering spatial literacy: The player has to solve several wayfinding tasks to various locations and answer questions at these locations. At the current state, it is a single-player game.

As part of this thesis, you would have to extend the concept of OriGami for a multi-player version, where players can compete or work together to solve tasks. After the conceptual development, you should implement a user management + implement the collaborative games and evaluate your game.

More information on OriGami can be found on our project website http://enable-project.eu/origami/.

Requirements:

You should have some experience in android programming and interest in location-based games.

Contact: Thomas Bartoschek, Angela Schwering

SIIEvaluating WPS Interoperability

The Open Geospatial Consortium (OGC) has published the Web Processing Ser-vices (WPS) specification, which defines a standard interface to geoprocessing re-sources.
A typical WPS use case is to provide functional views on very large data sets such as filter algorithms or aggregations. I.e. in those cases it is not necessary to download large data volumes and to do the processing locally but to invoke the processing capabilities of the data store remotely and to download just the re-sults, which are typically of smaller size.
Aim is to achieve plug-and-play interoperability between Client software and re-mote processing engines, as to ease access to arbitrary geocomputing capabili-ties.
The standard is intentionally kept generic to cover a variety of geoprocessing functionality ranging from simple geometric operators, such as buffering, to com-plex environmental models. But this generic nature also imposes challenges to achieving full interoperability between client and server software implementing the WPS standard.
A comprehensive study that evaluates the interoperabilty between different WPS software components is currently missing and is the core aim of this thesis. Therefore, a survey of related literature and projects on WPS should be done, in-cluding a review of the software available. Afterwards, concepts and methods for testing the pragmatic interoperability between geoprocessing web services and clients need to be developed. In a third step, the test environment has to be set up and tests have to be executed. Finally, the test results have to be evaluated.
The student will cooperate with 52°North staff, who is also active in the WPS standardization activities at OGC. Results of the thesis will be presented at an in-ternational conference and should help to improve the interoperability of geopro-cessing services. Interested students should have at least basic knowledge of Web technologies and OGC standards (SII lecture) as well as basic knowledge about testing methods (e.g. as taught in software engineering lectures).
Contact:
Prof. Dr. Albert Remke (a.remke@52north.org) Dr. Christoph Stasch (c.stasch@52north.org)

Contact: Prof. Dr. Albert Remke

SILQualitätssicherung von crowd-sourced Sensordaten

In der Arbeit soll untersucht werden, inwieweit Qualitätssicherung von crowd-sourced Sensordaten in einem Sensornetzwerk automatisierbar ist. Dies ist ein neues und hoch relevantes Forschungsfeld: große Datenmengen erlauben die Anwendung statistischer oder machine learning-Verfahren. Traditionelle Verfahren sind häufig nicht nutzbar, da die Daten in Echtzeit vorliegen müssen. Zudem stellen crowd-sourced Daten eine spezielle Herausforderungen dar, da nicht davon ausgegangen werden kann, dass alle Daten mit korrekten bzw. konsistenten Messverfahren erhoben wurden. Schließlich haben low-cost-Sensoren selbst Messfehler, die von professionellen Sensoren stark abweichen. oder Messstationen sind von Citizen Scientists schlecht montiert. Das Ziel ist, die Einflussfaktoren auf die Datenqualität und die Messgenauigkeit der Sensoren zu erforschen, Verfahren zur automatisierten Identifikation fehlerhafter Daten und möglicher Fehlerquellen zu entwickeln sowie automatisiert Entscheidungen über Möglichkeit zur Korrektur der Daten (bspw. über Nachkalibrierung der Sensoren) oder Ausschluss bestimmter Daten zu treffen.

Contact: Thomas Bartoschek

SILMeasuring Data Quality of Air Quality Measurements in Citizen Science

Im Kontext Citizen Science lassen sich mittelfristig räumlich hochaufgelöste Sensornetzwerke aufbauen um gesellschaftlich relevante Fragestellungen zu untersuchen. Die openSenseMap ist eine öffentlich zugängliche Plattform für das Veröffentlichen, Teilen und Visualsieren von solchen Umweltmessdaten als Open Data. Um eine flächendeckende Auflösung zu erreichen, werden die verwendeten Messstationen durch Freiwillige zusammengebaut, programmiert und installiert. Luftqualitätsmessungen nehmen hierbei eine gesonderte Rolle ein, da der Zugang zu teils komplizierter Messtechnik durch modulare Open-Source Bausätze, wie denen der senseBox, stark vereinfacht und einer breiten Masse zugänglich gemacht werden kann. In den meisten Fällen sind Freiwillige bei einer solchen Kampagne nicht wissenschaftlich ausbildet. Dadurch kann bei dieser Methodevnicht garantiert werden dass Messwerte von unterschiedlichen Stationen im Netzwerk direkt vergleichbar sind. Eine Frage die im Rahmen dieser Ausschreibung ausarbeitet werden soll ist ob trotz öffentlicher Beteiligung im Rahmen einer Messkampange in den Bürgerwissenschaften verwertbare Informationen und representative Analysen gegeben werden können. Als Anwendungsfall soll die Aussagekraft und Datenqualität von Gasmessungen (Stickoxid und Ozon) in Berlin auf der openSenseMap untersucht werden. 

Contact: Thomas Bartoschek

SILLearning Analytics für das Programmierenlernen mit Blockly

Learning Analytics ist eine Methode zur Erfassung, Messung, Analyse und Visualisierung von Daten über Lernende und ihren Kontext, die es ermöglicht den Lernfortschritt zu verstehen und Lernpfade zu optimieren, um dem Lernenden aber auch Lehrenden Feedback zum Lernprozess zu geben. Learning Analytics lässt sich besonders gut in digitale Lernplattformen einbinden. Blockly für die senseBox ermöglicht es Anfängern in die Mikrocontrollerprogrammierung einzusteigen.Trotzdem können Fehler im "Code" entstehen, wodurch das Programm nicht kompilierbar ist. 

Ziel der Arbeit ist die Entwicklung einer Learning Analytics Komponente, die dem Nutzer Feedback zum selbst programmierten Code gibt, Hinweise bei Fehlern und Lernfortschritte trackt.

 

Contact: Thomas Bartoschek

SILSpatial Learning Analytics

Learning Analytics is a method to collect, measure, analyze and visualize data about learners and their context. It enables the understaning of the learning process and allows an adaption of learning pathes based on the collected data. It also gives feedback to the learner and teacher about the learning process.

OriGami is a GeoGame fostering spatial literacy: The player has to solve several wayfinding tasks to various locations and answer questions at these locations. At the current state, it is a single-player game.
 
In this thesis OriGami (as a WebApp) needs to be transferred to a native app (based on previous work in the Ionic framework) and extended with spatial learning analytics where the movement data of the player is taken into account. 

Contact: Thomas Bartoschek

SILLearning Analytics for educational WebGIS

Learning Analytics is a method to collect, measure, analyze and visualize data about learners and their context. It enables the understanding of the learning process and allows an adaption of learning paths based on the collected data. It also gives feedback to the learner and teacher about the learning process.

WebGIS NRW is a prototypical WebGIS for educational purposes (webgis.nrw) based on MapBox GL.

In this thesis you should investigate the possibilities of learning analytics for the work with webGIS, i.e. cartography, map design, data analysis. A prototypical implementation of learning analytics components for webgis.nrw is part of the work.

Contact: Thomas Bartoschek

SILTesting the new Taxonomy of Human Wayfinding Tasks

In their seminal paper Wiener et al. (2009) defined the taxonomy of human wayfinding tasks. The taxonomy is based on the type of knowledge possessed by the navigator. However, it did not differentiate between any subcategories of the "Path Following" task. In other words, according to the taxonomy, there is no difference between (a) knowing your route without knowing anything about the wider surrounding enviornment, and (b) knowing your route AND knowing about the wider surrounding enviornment.

Schwering et al. (2017) argued that there are substantial differences between such two tasks and that they deserve to be distinguished in an updated taxonomy.

The goal of this thesis will be to test the hypothesis that following the same route, with the same knowledge about the route, is a cognitively different task depending on whether the navigator has, or does not have, survey knowledge about the broader envionment.

 

Wiener, J. M., Büchner, S. J., & Hölscher, C. (2009). Taxonomy of human wayfinding tasks: A knowledge-based approach. Spatial Cognition & Computation9(2), 152–165.

Schwering, A., Krukar, J., Li, R., Anacta, V. J., & Fuest, S. (2017). Wayfinding Through Orientation. Spatial Cognition & Computation17(4), 273–303. doi:10.1080/13875868.2017.1322597

 

Contact: Jakub Krukar

SILGeneralisation in Sketch Maps

When people draw sketch maps, they generalise information compared to the ground-truth information they perceived in the world. For example:

- many buildings belonging to university campus are drawn as a single polygon labelled "campus",

- a complex roundabout is drawn as a few crossing lines,

- 10 side streets from the main road are drawn as a few random lines, only to indicate that "some side streets" are there. 

This is a challenge for analysing sketch maps because this information is not wrong, yet a computer system for automated analysis would interpret it as such.

In the paper linked below we presented a classification of generalisation types in sketch maps, and many other examples. We also also have a working software prototype for analysing generalisation in sketch maps.

This thesis topic can be concerned with the software implementation of different algorithms for detecting generalised spatial information (e.g., in a comparison between a sketch map and OpenStreetMap data), and/or with theoretical studies of how people generalise information.

Examples of potential research questions:

- What affects the level of specific generalisations, e.g., when people perform stronger vs weaker forms of "amalgamation" generalisation?

- What are common ways to schematise complex spatial structures (e.g., roundabout), and how people do it spontaneously in drawings?

- How to detect and algoirthmically differentiate the >lack of some object< (e.g., a missing streets in an area) from their generalisation?

Manivannan, C., Krukar, J., & Schwering, A. (2022). Spatial generalization in sketch maps: A systematic classification. Journal of Environmental Psychology, 101851. https://doi.org/10.1016/j.jenvp.2022.101851

 

Contact: Jakub Krukar

SILAn analysis Tool for Metaphors used during Map Interaction Tasks

Metaphors reflect how people think about user interfaces, and people naturally produce spatial metaphors when talking about user interfaces (see e.g. Matlock et al, 2014). Though previous work has acknowledged their values in HCI, there is still a lack of techniques to formally describe metaphors and tools to facilitate their analysis. This thesis will use existing taxonomies of image schemas (e.g. Mandler and Canovas, 2014) to annotate people’s actions during map interaction and build a tool that help make sense of the metaphors they use. Tasks include:

Data preparation: users will be given map interaction tasks, and asked to describe their actions during the task completion (e.g. think aloud) or after completion (e.g. interview)

Encoding:  encode the actions described by users as a sequence of image schemas (e.g. AB-CD-EF-GH-EF-CD)

Analysis tool implementation: the tool will use text mining techniques and visualization techniques to help answer questions related to metaphors of map interaction. Example questions:

  • Which metaphors are used more often during the completion of map interaction tasks? [descriptive analysis]
  • Which metaphors occur more often together [co-occurrence analysis]?
  • How similar are metaphors used by two given users for a given task? [string similarity analysis]
  • What are regularities of metaphor use across users during a study? [sequence analysis]
  • Can a sequence of metaphors be used to predict a given interaction task? [predictive analysis]

 

References

 

Hurtienne, J. and Israel, J.H., 2007, February. Image schemas and their metaphorical extensions: intuitive patterns for tangible interaction. In Proceedings of the 1st international conference on Tangible and embedded interaction (pp. 127-134).

Matlock, T., Castro, S. C., Fleming, M., Gann, T. M. and Maglio, P. P. (2014) ‘Spatial metaphors of web use’, Spatial Cognition & Computation, 14(4), pp. 306–320. doi: 10.1080/13875868.2014.945587.

Mandler, J.M. and Cánovas, C.P., 2014. On defining image schemas. Language and Cognition, 6(4), pp.510-532.

 

Contact: Auriol Degbelo

SILSketch Maps as a tool for learning new environments

For decades, sketchmaps have been used as a tool for measuring spatial knowledge - i.e., for estimating how well participants know and understand some areas. However, evidence from psychological memory studies demonstrates, that drawing something can also be a good strategy to memorise a set of object. For instance, if you need to memorise the setting of a room, drawing the room as you see it is a better memorisation strategy than repeating the names of the objects verbally or in your head. This thesis will test whether drawing a sketch map is a good memorisation strategy for spatial environments and how this approach can be implemented in a gamified app. The problem is relevant for situations in which people must learn new spatial environments, e.g. to become taxi/delivery drivers, or when they move to a new city.
The thesis can be completed with focus on one of two aspects:
**Computational focus:** You will design a teaching app that (a) records the user's trajectory together with a list of landmarks that were visible along the route, and (b) after a delay, asks users to draw the area that they have travelled. Here the key problem may be to select routes and landmarks that the user should be asked to draw (based on the recorded trajectories).
**Evaluation focus:** You will design and conduct an experiment to evaluate the following research question: does drawing a sketchmap help people memorise the environment better, compared to alternative strategies? This does not require creating an app, and can be conducted as an in-situ experiment, or inside our Virtual Reality lab.

Contact: Jakub Krukar

GeoSimLand use optimization, raster vs. vector representation

What is the optimal land use, if we want to minimize deforestation and simultaneously maximize agricultural return? Finding that land use allocation is in most locations impossible, because the objectives are conflicting. On the other hand, it is possible to find compromise solutions that satisfy both objectives to some extent. Current techniques to solve that land use allocation problem belong to the cathegory Artificial Intelligence and are called Multi-Objective optimizations, for example Multi-Objective Genetic Algorithms. Spatial input data is used for the optimization, and the question arises, which representation shall be used: Fields (raster data) or objects (vector data). For computational efficiency, the chosen spatial input data for the optimization is usually of raster type instead of vector type. Nevertheless, raster representation can yield disadvantages compared to vector representation: Information regarding the topology is lost and data inaccuracies can be generated during the conversion process.

The aim of this thesis is to investigate the impact of using different data representations in land use allocation optimization. Depending on the student's interest and background, possible specialisations can be to

1) Analyse the optimization results generated with raster data and vector data on different spatial scales with real data.

2) Implement GPU-accelerated computations of the objective functions and compare the algorithm performance per representation type. 

Contact: Moritz Hildemann, Judith Verstegen

GeoSimGenetic programming to optimize 3D trajectories

Finding the optimal trajectory in an 3D space is an ongoing research topic with applications such as optimizing an underwater route for a submarine robot or a flight route for drones. The problem becomes challenging as soon as the 3D space has barriers like danger zones or protected spaces. Those barriers can be modelled as features in GIS. A research gap to be closed is to combine the trajectory optimization techniques with GIS-modelled 3D barriers. Namely, the produced 3D-routes from the optimization techniques need a validation process to ensure that no barriers are crossed. Since many validations are necessary, one requirement is a fast computation.

The aim of this thesis is to solve the trajectory optimization problem with the artificial inteligence technique called "Genetic Programming" (GP). The produced trajectories are to be converted into geographical lines, which are tested for any interference with GIS-modelled 3D barriers. 

Contact: Moritz Hildemann, Judith Verstegen

SILSupporting Land Rights Management through Digital Mobile Maps

Despite substantial progress in digital technologies, and the increasing adoption of digital technologies for the collection of land-related data (see e.g. ArcGIS Field Maps, Qfield, or MAST), information related to land (e.g. citizen-to-city-council-communication, purchase transaction, inheritance transactions) is still managed manually. This is a missed opportunity and this thesis will use a Research through Design (RtD) approach to explore possible futures of digital land rights management. RtD is a “type of research practice focused on improving the world by making new things that disrupt, complicate or transform the current state of the world” (Zimmerman and Forlizzi 2014). Tasks include:  

 

Requirements specification: derive requirements for applications for map-based land rights data management through scenario formulation (for an example of scenario-based study, see Degbelo and Somaskantharajan, 2020)

 

Implementation: develop a mobile app (frontend + backend) that implements some of the requirements. The mobile app may be a simple responsive web app or use a framework for mobile app development

Evaluation: validation of design decisions of the mobile app through interviews with domain experts and/or additional users. 

 

References

 

Degbelo, A. and Somaskantharajan, S. (2020) ‘Speech-based interaction for map editing on mobile devices: a scenario-based study’, in Alt, F., Schneegass, S., and Hornecker, E. (eds) Mensch und Computer 2020. Magdeburg, Germany: ACM, pp. 343–347. doi: 10.1145/3404983.3409996.

 

Zimmerman, J. and Forlizzi, J. (2014) ‘Research through design in HCI’, in Olson, J. S. and Kellogg, W. A. (eds) Ways of Knowing in HCI. New York, New York, USA: Springer New York, pp. 167–189. doi: 10.1007/978-1-4939-0378-8_8.

 

Links

ArcGIS Field Maps: https://www.esri.com/de-de/arcgis/products/arcgis-field-maps/overview 

Qfield: https://qfield.org/ 

MAST: https://www.land-links.org/tool-resource/mast-technology-2/

Contact: Auriol Degbelo

SILA computational approach for the comparison of evaluative city images

Evaluative images of cities (i.e. composite maps capturing citizens’ perceptions of areas in the cities) are a rich source of qualitative data that can be used by urban planners to understand how people relate to neighborhoods of their cities. They were proposed by Nasar (1990), who suggested deriving them through interviews. Recently Barros et al (2022) proposed a web-mapping approach to create city images computationally, with promising results. This thesis will build upon this and explore techniques to compare two evaluative city images computationally. Tasks include:

Data collection: collect data about how people perceive areas of a city (e.g. Münster) concerning the five dimensions originally suggested by Nasar (i.e. naturalness, upkeep, openness, historical significance, and openness).

Development of a comparison module for evaluative city images: at least two directions are possible.

  • A) The module helps assess and visually communicate similarities/differences between two evaluative city images. Possible applications of the comparison module include: assessment of the temporal variability (collection of two evaluative city images at different points in time), the spatial variability (comparison of evaluative city images from two different neighborhoods, or cities), the statements made by the city images about the mapped regions, and the (spatial) accuracies of the city images.
  • B) The module relates an evaluative city image (qualitative data) with quantitative data available about the mapped neighbourhoods to identify conflicting statements and converging evidence for urban facts.

Evaluation: aspects to be evaluated could include the computational performance (how the module performs as the areas mapped by participants grow), the effectiveness of the visual communication of differences (how easy it is for users to perceive similarities and differences), or the usefulness of the comparison module for urban planners (what experts say about integrating the module into their workflows).

 

Readings

Barros, M. S., Degbelo, A. and Filomena, G. (2022) ‘Evaluative image 2.0: A web mapping approach to capture people’s perceptions of a city’, Transactions in GIS, 26(2), pp. 1116–1139. doi: 10.1111/tgis.12867.

Nasar, J. L. (1990) ‘The evaluative image of the city’, Journal of the American Planning Association, 56(1), pp. 41–53. doi: 10.1080/01944369008975742.

 

Demos

eImage Canvas: https://www.youtube.com/watch?v=z8dtaum_Xu0

eImage Viewer: https://www.youtube.com/watch?v=J-MDp9YItnc

 

Contact

Auriol Degbelo (auriol.degbelo@uni-muenster.de )

Simge Oktay (soezdalo@uni-muenster.de )   

 

Contact: Auriol Degbelo

SILDisambiguation questions for spatial search sessions

Current search engines return results in almost a fraction of a second to any query submitted by users. This often comes with the drawback of users navigating through a number of irrelevant results. This thesis will explore the impact of adding query disambiguation steps of overall user satisfaction during search, and investigate the extent to which users are willing to trade response time with quality of results. Tasks envisioned include:

 

1. Data preparation

 

2. Development of a prototypical search engine, which (a) asks users which place they mean when homonyms are available (e.g. Berlin, USA vs Berlin, Germany), (b) offer disambiguation functionalities before the display of results using e.g. the Geonames API 

 

3. Development of a prototypical search engine, which simulates the functionalities of the current state of the art offered by search engines


The evaluation will involve A/B testing, and gathering of participants’ feedback during a user study.

 

Contact: Auriol Degbelo

SILReplicability of wayfinding research

"Replication" refers to the process of re-creating an experiment published by other researchers in an effort of obtaining results pointing to the same conclusion. A "replication crisis" showed that many published research is not replicable. We can distinguish two types of replication:

- an "exact replication" is the attempt of recreating every detail of the original experiment

- a "conceptual replication" is the attempt of creating a similar experiment, with similar hypotheses, but perhaps with a different stimuli, instructions, or groups of participants.

This thesis focuses on a "conceptual replication" of navigation research.

Navigation research is usually performed in very specific spatial context (such as the city in which the paper's authors are based or the virtual environment that they have created). This introduces a challenge to generalizability and replicability of navigation research because we do not know whether classic research findings would be equally applicable in different spatial contexts (e.g., a different city).

This thesis focuses on replicating an existing wayfinding paper (to be chosen by the student) in Münster, or in a virtual environment available at ifgi.

The key challenge is finding a way to make the new spatial context (of Münster) comparable to that of the original paper.

Thesis co-supervised by Daniel Nüst (with technical support w.r.t. replicability).

 

Examples of papers that can be replicated:

https://doi.org/10.1080/17470218.2014.963131

https://doi.org/10.1016/j.cognition.2011.06.005

 

Contact: Jakub Krukar

SILCommunicating the effects of the areal unit problem in spatio-temporal data analysis

The modifiable areal unit (MAUP, see [1]) problem is a well-known issue to geographers and GI scientists working with spatially aggregated data. In essence, the MAUP states that the conclusions drawn about a phenomenon are strongly dependent upon the spatial granularity of the analysis. The same principle holds for the temporal granularity of analyses and has been called the modifiable temporal unit problem (MTUP, see [2]). Given an increasing amount of visualizations of geographical data on the Web (e.g. to communicate about the outcomes of political elections and the spread of diseases such as Covid-19), there is a need for effective ways to educate citizens about the effects of the MAUP and MTUP. This thesis aims to explore strategies to make the general public aware of MAUP and MTUP issues. Tasks of this work include:

 

Task 1: data collection (at different levels of spatial detail) about a topic. One of the aims of this work is to inform the public opinion about progress regarding the SDG 4.3. The Sustainable Development Goal 4.3 strives to “ensure equal access for all women and men to affordable and quality technical, vocational and tertiary education, including university”. Thus, we envision here data collection about gender equality in education in the Netherlands and/or Germany (e.g. statistics about how many females/males started a given subject of study & how many graduated). The data will be analyzed at different levels, e.g. city district, city, country state, and country.

 

Task 2: Develop and evaluate strategies to 1) maximize user experience and engagement of participants with the topic of the SDG 4.3, and 2) raise awareness to the general public about the effects of the MAUP and MTUP on the interpretation of the data. The prototype should be optimized for use on mobile devices, as we assume that citizens primarily inform themselves through mobile devices. Knowledge of technologies for mobile app development (e.g. Apache Cordova, React Native, or others) is expected.

 

Task 3: A user study to assess the impact of the different strategies devised on the user experience, and engagement with the topic. The study may also assess the impact of personality types on the preferences of users.

This work will contribute i) curated spatio-temporal data about SDG 4.3; ii) a mobile prototype that communicates a snapshot of the progress made on SDG 4.3; and iii) insight about visualization/interaction strategies that are most convenient to users while interacting with this important societal topic. The lessons learned in this work can inform about how to best communicate SDG data and help citizens engage with it more broadly.

 

[1] Fogarty, E. A. (2010) ‘Modifiable areal unit problem’, in Warf, B. (ed.) Encyclopedia of Geography. SAGE Publications, Inc., pp. 1935–1937. doi: 10.4135/9781412939591.n780.

 

[2] Coltekin, A., De Sabbata, S., Willi, C., Vontobel, I., Pfister, S., Kuhn, M. and Lacayo-Emery, M. (2011) ‘The modifiable temporal unit problem’, in ICC2011 Workshop. doi: 10.5167/uzh-54263.

 

Contact: Auriol Degbelo

SILWI-FI Based Traffic Monitoring

WI-FI is a promising technique to monitor traffic in different scenarios. As discussed in [2], WiFi sensing is beneficial because 1) low-cost, 2) not intrusive and 3) sensitive to lighting conditions. First tests done in [1] in Münster have yielded encouraging results for the monitoring of cars, but much is still needed to grasp the potential of the technique.  The aim of this thesis is to advance our understanding of the benefits and costs of WI-FI sensing for city traffic monitoring.

Tasks include:

  • Scenario definition, system setup and data collection
  • Prototype implementation for automated vehicle detection and classification
  • Evaluation of the classification accuracies

Possible directions include (non-exhaustive list)

  • Detection of vehicles on multiple lanes using Wi-Fi (e.g. two lanes, one direction; two lanes, two directions)
  • Effect of the spatial distance between sending/receiving units on the classification accuracy
  • Effect of the temporal resolution of the measurements on the classification accuracy
  • Effect of the use of channel state information (CSI) on the classification accuracy
  • Measurement of velocity and direction of vehicles (e.g. through the addition of additional sender/receiving units)
  • WIFI-Sensing for real-time traffic monitoring
  • Designing and developing prototype for real life implementation.

References

[1] Gupta, S., Hamzin, A. and Degbelo, A. (2018) ‘A low-cost open hardware system for collecting traffic data using Wi-Fi signal strength’, Sensors, 18(11), p. 3623. doi: 10.3390/s18113623.

[2] Ma, Y., Zhou, G. and Wang, S. (2019) ‘WiFi sensing with channel state information’, ACM Computing Surveys, 52(3), pp. 1–36. doi: 10.1145/3310194.

 

Contact: Auriol Degbelo

SILKnowledge Harvesting for Geovisualization Search

Geovisualizations are increasingly available and finding them manually is becoming increasingly challenging. There is thus a need for techniques to automate their discovery. This work will build and evaluate a prototype for the generation of knowledge graphs on top of geovisualizations. Tasks include:

 

Dataset preparation: visualizations from existing galleries will be extracted manually, or through a crawler.

 

Knowledge harvesting: approaches to generate knowledge graphs out of these visualizations will be discussed and implemented.

 

Evaluation: the approaches to knowledge graph generation will be tested via a user study. The baseline here will be keywords search.

 

Readings

 

Hu, Y., Janowicz, K., Prasad, S. and Gao, S. (2015) ‘Enabling semantic search and knowledge discovery for ArcGIS online: A linked-data-driven approach’, in Bacao, F., Santos, M. Y., and Painho, M. (eds) AGILE 2015 - Geographic Information Science as an Enabler of Smarter Cities and Communities.

 

Weikum, G., Hoffart, J. and Suchanek, F. (2019) ‘Knowledge harvesting: achievements and challenges’, in Steffen, B. and Woeginger, G. (eds) Computing and Software Science: State of the Art and Perspectives, pp. 217–235. doi: 10.1007/978-3-319-91908-9_13.

 

Contact: Auriol Degbelo

SILSummary of Graphical Annotations produced during Spatial Analysis

Annotations of visualizations is a useful task during exploratory data analysis. As illustrated in [1,2], users may want to record trends, clusters, distributions or outliers found on components of a geovisualization during a data analysis session. A key challenge here is that of summarizing the annotations produced by various users during their exploration activities. This thesis will look into computational and/or graphical means of producing these summaries. Tasks include: 

Prototyping: design of a web-based or mobile prototype to record annotations of web maps graphically.

Data collection: here graphical annotation of web-maps will be generated. These annotations could be synthetic or real, depending on the scope of the thesis. 

Summarization: this design, implementation and evaluation of summary approaches. 

 

References

[1] Heer, J., Viégas, F. B. and Wattenberg, M. (2009) ‘Voyagers and voyeurs: supporting asynchronous collaborative visualization’, Communications of the ACM, 52(1), pp. 87–97. doi: 10.1145/1435417.1435439.

[2] Lai, P.-C. and Degbelo, A. (2021) ‘A comparative study of typing and speech for map metadata creation’, in Partsinevelos, P., Kyriakidis, P., and Kavouras, M. (eds) Proceedings of the 24th AGILE Conference on Geographic Information Science (AGILE 2021), pp. 1–12. doi: 10.5194/agile-giss-2-7-2021.

Contact: Auriol Degbelo

STMLScalable data cube processing with containerization

Data cubes are an efficient representation for spatiotemporal data as from Earth observation satellites. By multidimensional chunking, they allow highly parallel execution of complex analysis such as time series change detection. The aim of the thesis is to create an architecture that allows for scaling such operations in a distributed computing environment using containerized processing (Docker) and tools for container orchestration such as Kubernetes. A prototypical implementation as extension to the gdalcubes library shall be developed and used for a detailed analysis of the scalability of the proposed architecture.

Contact: Marius Appel

SITCOMFreehand interaction with Spatial Data on Large Displays

Visualizations of spatial data on large displays could be very useful during the exploration of large spatial datasets. A key issue in this context is the design of natural interaction (e.g. interaction with freehand gestures and/or speech-based interaction) to support exploratory data analysis tasks (e.g. panning, zooming, selection, aggregation, annotation). This thesis aims to explore strategies to facilitate the interaction with spatial data on large displays using freehand gestures. Tasks include:

Task 1: implementation of a software module to reliably map the position of users’ hands to the screen’s positions.

Task 2: implementation of a software module that enables panning and zooming with maps via freehand gestures on a large display.

Task 3: evaluation of the software modules (e.g. performance, usability, usefulness).

The immersive video environment will serve as a prototypical large display during the study. Programming skills (C++, C# or VB.NET) will be an advantage.

 

Contact

Auriol Degbelo (auriol.degbelo@uni-muenster.de)

Samuel Navas Medrano (s.navas@uni-muenster.de)

 

Readings

Bartoschek, T., Pape, G., Kray, C., Jones, J. and Kauppinen, T. (2014) ‘Gestural interaction with spatiotemporal linked open data’, OSGeo Journal, 13(1), pp. 60–67.

Nancel, M., Wagner, J., Pietriga, E., Chapuis, O. and Mackay, W. (2011) ‘Mid-air pan-and-zoom on wall-sized displays’, in Tan, D. S., Amershi, S., Begole, B., Kellogg, W. A., and Tungare, M. (eds) Proceedings of the 2011 annual conference on Human factors in computing systems - CHI ’11. Vancouver, Canada: ACM Press, pp. 177–186. doi: 10.1145/1978942.1978969.

Contact: Auriol Degbelo

STMLInfluence of spatial dataset characteristics on machine learning model transferability

Assessing and improving the transferability of machine learning models is an important task for the large-scale mapping of environmental observations. In the context of spatial modelling and remote sensing, the transferability of a model relates to its ability to validly predict in geographic regions where training data is not available. These regions can be identified using the Area of Applicability method currently published and developed in the Uebersat Project (Meyer et al. 2021).
Understanding the influence of the spatial distribution of training data and predictor interactions on the models transferability is crucial for the interpretation of predictions. In Barbiero et al. (2020) the models transferability and generalization from different datasets was tested using a convex hull approach and related to structural characteristics of the dataset (e.g. correlations between predictors).
In this context, we offer a Master Thesis dealing with the following topics:

* Reproduce ideas of Barbiero et al. (2020) and assess model transferability with the Area of Applicability
* Complement the concept by using spatial data and refer model transferability to spatially intrinsic aspects of the data

The Thesis will be part of the Ubersat Project and conducted in a collaborative team from ILOEK and IFGI.

 

Literature:
Barbiero, Pietro, Giovanni Squillero, and Alberto Tonda. 2020. “Modeling Generalization in Machine Learning: A Methodological and Computational Study.” arXiv:2006.15680 [Cs, Stat], June. https://arxiv.org/abs/2006.15680.

Meyer, Hanna and Edzer Pebesma. 2021. “Predicting into Unknown Space? Estimating the Area of Applicability of Spatial Prediction Models.” Methods in Ecology and Evolution, June, 2041–210X.13650. https://doi.org/10.1111/2041-210X.13650.

Contact: Marvin Ludwig

STMLA lightweight platform to analyze Earth observation data cubes in the cloud

Cloud-computing services for analyzing larger collections of satellite imagery typically have a complex architecture, include strong requirements on the infrastructure, and are difficult to deploy for others, even if the underlying libraries are open-source. To facilitate reproducing analyses on smaller scale, a lightweight Earth observation platform that is easy to deploy and directly works on cloud computing platforms containing freely available satellite imagery such as Amazon web services, Google cloud platform, or Microsoft Azure would be needed.

This thesis aims at developing such a platform by interfacing the gdalcubes R package [1] and the openEO API [2]. The gdalcubes R package is a lightweight library to create and analyse on-demand data cubes from satellite image collections. Among others, it can be used to run user-defined R functions on data cubes and reading data from cloud storage directly. The openEO project in contrast provides a standardized REST-like API that makes analyses interoperable even among different cloud services.

Important questions for the thesis may include:

  • How can gdalcubes functions be mapped to openEO processes?
  • How to make the deployment as easy as possible?
  • How to interface the backend with available STAC API endpoints?
  • How to publish results as interactive web-service?
  • How to run user-defined R functions on data cubes?

Do not hesitate to contact Marius Appel or Edzer Pebesma if interested.

Contact: Marius Appel