Résumé
In this paper we investigate a local implementation of Random Forest (RF), named Geographical Random Forest (GRF) to predict population density with Very-High-Resolution Remote Sensing (VHHRS) data. As an independent variable we use population density at the neighborhood level from the 2013 census of Dakar, while as explanatory features, the proportions of three different built-up types in each neighborhood derived from a VHHRS land cover classification. The results demonstrated, that by using an appropriate geographic scale to calibrate GRF, we can maximize prediction accuracy due to the incorporation of spatial heterogeneity in the estimates. Additionally, since GRF is an ensemble of local sub-models, the results can be mapped, highlighting local model performance and other interesting spatial variations. Consequently, GRF is suggested as an interesting exploratory and explanatory technique to model remotely-sensed spatially heterogeneous relationships.
langue originale | Anglais |
---|---|
titre | 2019 Joint Urban Remote Sensing Event, JURSE 2019 |
Sous-titre | May, 22-24, 2019, Vannes, France |
Editeur | Institute of Electrical and Electronics Engineers Inc. |
ISBN (Electronique) | 9781728100098 |
Les DOIs | |
Etat de la publication | Publié - 1 mai 2019 |
Evénement | 2019 Joint Urban Remote Sensing Event, JURSE 2019 - Vannes, France Durée: 22 mai 2019 → 24 mai 2019 |
Série de publications
Nom | 2019 Joint Urban Remote Sensing Event, JURSE 2019 |
---|
Une conférence
Une conférence | 2019 Joint Urban Remote Sensing Event, JURSE 2019 |
---|---|
Pays/Territoire | France |
La ville | Vannes |
période | 22/05/19 → 24/05/19 |
Financement
ACKNOWLEDGMENT This work was supported by the Remote sensing for Epidemiology in African Cities (REACT : http://react.ulb.be/) project, funded by the STEREO-III program of the Belgian Science Policy (BELSPO, SR/00/337). The population data were provided by the ASSESS project, funded by the ARES-CDD.