How Bayesian modelling could use the Big Tech competition in producing built-up maps: predicting population in data-scarce contexts
How Bayesian modelling could use the Big Tech competition in producing built-up maps: predicting population in data-scarce contexts
Satellite-imagery derived products represent an exciting opportunity to map and estimate current population with high spatial precision in context where traditional demographic data is not available. Bayesian methods have been developped to harness the potential of built-up maps as a proxy for human settlements, thereby predicting population sizes of hard-to-reach areas across sub-saharan Africa. The increased availibility of high-resolution satellite imagery has fostered the competition between high-profile institutions to produce global-scale built-up maps. Given the key role of built-up maps to estimate population sizes we need to understand (1) how do the different sources impact population predictions, (2) how do they compare with human-made maps and finally (3) how can they be articulated together.
remote-sensing, population, building footprint, bayesian analysis
Darin, Edith
868fa688-2567-4dbd-aa12-3dcc91f2aa8d
October 2022
Darin, Edith
868fa688-2567-4dbd-aa12-3dcc91f2aa8d
Darin, Edith
(2022)
How Bayesian modelling could use the Big Tech competition in producing built-up maps: predicting population in data-scarce contexts.
Bayesian Methods for the Social Sciences, Institut Henri Poincaré, Paris, France.
19 - 21 Oct 2022.
Record type:
Conference or Workshop Item
(Poster)
Abstract
Satellite-imagery derived products represent an exciting opportunity to map and estimate current population with high spatial precision in context where traditional demographic data is not available. Bayesian methods have been developped to harness the potential of built-up maps as a proxy for human settlements, thereby predicting population sizes of hard-to-reach areas across sub-saharan Africa. The increased availibility of high-resolution satellite imagery has fostered the competition between high-profile institutions to produce global-scale built-up maps. Given the key role of built-up maps to estimate population sizes we need to understand (1) how do the different sources impact population predictions, (2) how do they compare with human-made maps and finally (3) how can they be articulated together.
Text
bayes_shs_poster_DARIN
More information
Published date: October 2022
Venue - Dates:
Bayesian Methods for the Social Sciences, Institut Henri Poincaré, Paris, France, 2022-10-19 - 2022-10-21
Keywords:
remote-sensing, population, building footprint, bayesian analysis
Identifiers
Local EPrints ID: 472521
URI: http://eprints.soton.ac.uk/id/eprint/472521
PURE UUID: 2035c05d-3bc5-41a9-b2fd-cd07dc683962
Catalogue record
Date deposited: 07 Dec 2022 17:48
Last modified: 17 Mar 2024 04:00
Export record
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
