Modelling urban socio-economic inequalities - living conditions - deprivation (slums) using machine learning and earth observation (topic has several sub-topics)

IDEAMAPS Network (with a research team from Nigeria, Kenya, UK, US and NL): https://ideamapsnetwork.org [1,2]
ADEAtlas (ESA project) (with a research team from Austria, NL and associated partners in Argentina, Brazil, Mexico, Colombia, Kenya, Belgium, US, Nigeria, Indonesia and India): https://slummap.net/index.php/ideatlas
SPACE4ALL https://www.itc.nl/space4all/ (research team at ITC and BMS - several African countries)
ONEKANA https://onekana.ulb.be/ and follow-up project starting 2025 (DynEO4SLUMS: Space-Time Dynamics of Slums and Vulnerable Communities Exposed to Multiple Hazards) (several African and Latin American countries included)

CONTEXT:

Deprived areas (commonly called slums) are a socio and economic by-product of rapid urbanization in many countries of the Global South (but also newly developing refugee camps in the Global North share similar characteristics) [3]. Deprived areas are associated with poor living and housing conditions, overcrowding and tenure insecurity [3]. Development processes in many cities are often very rapid, and planning authorities do not have updated base data. Many city administrations do not have slums on official maps. Thus, global data sets are very inconsistent and outdated but urgently required to support local and global initiatives and policies (e.g., SDG indicator 11.1.1) [4]). Current Machine-Learning (ML) based Remote Sensing algorithms have the potential to map such areas via their morphological characteristics. However, such areas are rather complex and diverse, with often fuzzy boundaries. Official statistics and data do not provide reliable data on the number of inhabitants or living conditions. To support improvement strategies, accurate and up-to-date locational and population data are required. However, data are commonly inconsistent, updated and underestimate the population.

RS studies [e.g., 2, 4-12] have shown the potential of satellite imagery to provide consistent, accurate and timely information on the location of deprived areas. Very-high-resolution (VHR) or high-resolution images can map and characterize deprivation [9], largely drawing on locally adapted image features. Advanced machine-learning methods are overcoming the need to define locally adapted image features [10] and are, therefore, more transferable across countries and cities. However, most studies neither produce city-level delineations of deprived areas (due to image and computational costs), nor do studies produce an assessment of the variations of the socio-economic conditions at city scale or provide population estimates in support of policy-relevant information (due to the unavailability of bottom-up estimation models). Furthermore, variations in socioeconomic conditions can be related to information on climate change-related risks (e.g., urban heat, floods, landslides), lack of infrastructure (e.g., electricity, street light), or health outcomes.

Supervisors: Monika Kuffer, Mariana Belgiu, Mila Koeva, Claudio Persello; Caroline Gevaert, Raian Vargas Maretto, Jon Wang

References

1. Abascal, A., Rodríguez-Carreño, I., Vanhuysse, S., Georganos, S., Sliuzas, R., Wolff, E., & Kuffer, M. (2022). Identifying degrees of deprivation from space using deep learning and morphological spatial analysis of deprived urban areas. Computers, Environment and Urban Systems, 95, 101820. doi:https://doi.org/10.1016/j.compenvurbsys.2022.101820
2. Kuffer, M., Thomson, D. R., Boo, G., Mahabir, R., Grippa, T., Vanhuysse, S., . . . Kabaria, C. (2020). The Role of Earth Observation in an Integrated Deprived Area Mapping “System” for Low-to-Middle Income Countries. Remote Sens., 12(6), 982.
3. UN-Habitat. State of the world's cities 2010- 2011: Bridging the urban divide. Nairobi, Kenya, 2010.
4. Kuffer, M.; Pfeffer, K.; Sliuzas, R. Slums from space—15 years of slum mapping using remote sensing. Remote Sens. 2016, 8, 455.
5. UN-Habitat. Slums Almanac 2015-16. Tracking Improvement in the Lives of Slum Dwellers. Nairobi, 2016.
6. Duque, J.C.; Patino, J.E.; Betancourt, A. Exploring the Potential of Machine Learning for Automatic Slum Identification from VHR Imagery. Remote Sens. 2017, 9, 895.
7. Wurm, M.; Taubenböck, H. Detecting social groups from space – Assessment of remote sensing-based mapped morphological slums using income data. Remote Sens. Lett. 2018, 9, 41-50.
8. Taubenböck, H.; Kraff, N.J.; Wurm, M. The morphology of the arrival city - A global categorization based on literature surveys and remotely sensed data. Appl. Geogr. 2018, 92, 150-167.
9. Persello, C.; Stein, A. Deep Fully Convolutional Networks for the Detection of Informal Settlements in VHR Images. IEEE Geosci. Remote Sens. Lett. 2017, 14, 2325-2329.
10. Kuffer, M.; Pfeffer, K.; Sliuzas, R.; Baud, I. Extraction of slum areas from VHR imagery using GLCM variance. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2016, 9, 1830–1840.
11. Kuffer, M.; Pfeffer, K.; Sliuzas, R.; Baud, I.; van Maarseveen, M. Capturing the Diversity of Deprived Areas with Image-Based Features: The Case of Mumbai. Remote Sens. 2017, 9, 384.
12. Ajami, A.; Kuffer, M.; Persello, C.; Pfeffer, K. Identifying a Slums’ Degree of Deprivation from VHR Images Using Convolutional Neural Networks. Remote Sensing 2019, 11, 1282.
13. Boanada-Fuchs, A., Kuffer, M., & Samper, J. (2024). A Global Estimate of the Size and Location of Informal Settlements. Urban Science, 8(1), 18. https://www.mdpi.com/2413-8851/8/1/18
14. Linares Arroyo, H., Abascal, A., Degen, T., Aubé, M., Espey, B. R., Gyuk, G., Hölker, F., Jechow, A., Kuffer, M., Sánchez de Miguel, A., Simoneau, A., Walczak, K., & Kyba, C. C. M. (2024). Monitoring, trends and impacts of light pollution. Nature Reviews Earth & Environment, 5(6), 417-430. https://doi.org/10.1038/s43017-024-00555-9

Topic objectives and methodology

The focus of this research topic will be on employing state-of-the-art remote sensing algorithms or geospatial models for modelling living conditions, and in specific deprivation, linked to socio-economic and environmental conditions within cities, this will be combined with local data (either existing data, data collected in the field or via collaborating organizations). There are several possible foci (topics):
1. Development of generalizable (transferable) deep learning to model the variations of socio-economic conditions
2. Modelling of environmental conditions in cities (e.g., urban heat exposure, floods)
3. Comparing state-of-the art deep learning methods to make bottom-up population estimates
4. Settlement-based mapping of the urban morphology using 3D modelling
5. Analysing the climate change-related risks of deprived areas in different regions (e.g., of selected cases in Africa, Latin America and Asia).
6. Co-design and development of spatial models that integrate citizen science data with open geospatial data to analyze the living conditions in cities.
7. Mapping access and stability of electricity in Low-and Middle-Income Countries from Night Light Remote Sensing.
8. Relating deprivation with health outcomes and food security

References for further reading

2. Kuffer, M., Thomson, D. R., Boo, G., Mahabir, R., Grippa, T., Vanhuysse, S., . . . Kabaria, C. (2020). The Role of Earth Observation in an Integrated Deprived Area Mapping “System” for Low-to-Middle Income Countries. Remote Sens., 12(6), 982.

3. Ajami, A.; Kuffer, M.; Persello, C.; Pfeffer, K. Identifying a Slums’ Degree of Deprivation from VHR Images Using Convolutional Neural Networks. Remote Sensing 2019, 11, 1282.

4. Abascal, A., Rodríguez-Carreño, I., Vanhuysse, S., Georganos, S., Sliuzas, R., Wolff, E., & Kuffer, M. (2022). Identifying degrees of deprivation from space using deep learning and morphological spatial analysis of deprived urban areas. Computers, Environment and Urban Systems, 95, 101820. doi:https://doi.org/10.1016/j.compenvurbsys.2022.101820

5. Georganos, S., Hafner, S., Kuffer, M., Linard, C., & Ban, Y. (2022). A census from heaven: Unraveling the potential of deep learning and Earth Observation for intra-urban population mapping in data scarce environments. International Journal of Applied Earth Observation and Geoinformation, 114, 103013. doi:https://doi.org/10.1016/j.jag.2022.103013

6. Kyba, C. C. M., Altıntaş, Y. Ö., Walker, C. E., & Newhouse, M. (2023). Citizen scientists report global rapid reductions in the visibility of stars from 2011 to 2022. Science, 379(6629), 265-268. doi:doi:10.1126/science.abq7781

7. Owusu, M., Kuffer, M., Belgiu, M., Grippa, T., Lennert, M., Georganos, S., & Vanhuysse, S. (2021). Towards user-driven earth observation-based slum mapping. Computers, Environment and Urban Systems, 89, 101681. doi:https://doi.org/10.1016/j.compenvurbsys.2021.101681

How can topic be adapted to Spatial Engineering

We work with various stakeholder groups, including governments (local and national) and the topic of making advanced modelling outputs relevant and access to diverse user groups might interest Spatial Engineering students.