Mapping slums and model population density using earth observation data and open source solutions

Abstract

This paper presents a collection of frameworks aiming at mapping land cover, land use and estimate population densities from very-high resolution images and relying on open-source software. Using height information and landscape metrics, slums location and extent can be accurately extracted from the rest of the city. Moreover, the processing chain developed can deal with large amount of data and produce useful pieces of geographical information citywide. All the results, methods and computer code are available in open-access for anyone and any purpose.

High-resolution estimates of social distancing feasibility, mapped for urban areas in sub-Saharan Africa

“…Data on all features used in defining the boundaries of the custom spatial units was extracted from OpenStreetMap 17 and is detailed in Table 1. Similar approaches utilising OpenStreetMap (OSM) data to create spatial units based on recognisable features have been used in sub-Saharan African cities for studies on urban land use classification 18,19 , slum mapping 20 and semi-automated approaches to create census enumeration units 21 .…”

Leveraging OSM and GEOBIA to Create and Update Forest Type Maps

“…OSM data is used regularly in studies, e.g., to update existing datasets [45,46] or to create new datasets based on OSM data [41][42][43], but rarely do these projects generate feedback to the OSM database. Ideally, the investigation would result in an assessment of the feasibility to use GEOBIA to identify subdivisions of existing OSM forest polygons that can subsequently be integrated back into the OSM database with updated keys on leaf_type.…”