The quality of remote sensing based land monitoring is often assessed statistical methods which gives an overall indication on how good the prediction is. However, this overall quality indicator does not inform
inform about spatial variability in map accuracy, yet predictions errors are not distributed evenly across the map. For regional users, it is important to know if the quality of the map is good for their region of interests.
Spatial variation of map accuracy can be modelled using different techniques such as indicator kriging  and geographically weighted regression (GWR) . However, these methods are known to be time consuming and require a large number of reference sample sites with a good geographical spread. Several machine learning based approaches such as random forest spatial interpolation and geographic random forests were recently introduced and their efficiency and quality for spatial prediction of map quality has not been studied.
This study aims to investigate the efficiency of the random forest based spatial predictions in assessing the local accuracy of maps. depending on the data availability, the region of interests can be in Europe or some parts of Africa. The variables of interests can be land cover or soil types.
This thesis is linked to the ESA funded project, in which WU is a partner: https://esa-worldcover.org
- Assess and compare random forest based algorithms for spatial accuracy of maps.
- Tsendbazar, N.E., de Bruin, S., Fritz, S., & Herold, M. (2015). Spatial Accuracy Assessment and Integration of Global Land Cover Datasets. Remote Sensing, 7, 15804-15821
- See, L., Fritz, S., Perger, C., Schill, C., McCallum, I., Schepaschenko, D., Duerauer, M., Sturn, T., Karner, M., & Kraxner, F. (2015). Harnessing the power of volunteers, the internet and Google Earth to collect and validate global spatial information using Geo-Wiki. Technological Forecasting and Social Change, 98, 324-335
- Stefanos Georganos, Tais Grippa, Assane Niang Gadiaga, Catherine Linard, Moritz Lennert, Sabine Vanhuysse, Nicholus Mboga, Eléonore Wolff & Stamatis Kalogirou (2021) Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling, Geocarto International, 36:2, 121-136, DOI: 10.1080/10106049.2019.1595177
- Sekulić, A., Kilibarda, M., Heuvelink, G.B.M., Nikolić, M., & Bajat, B. (2020). Random Forest Spatial Interpolation. Remote Sensing, 12, 1687
- Møller, A.B., Beucher, A.M., Pouladi, N., & Greve, M.H. (2020). Oblique geographic coordinates as covariates for digital soil mapping. SOIL, 6, 269-289
- Spatial modelling and statistics
Theme(s): Integrated Land Monitoring