Jarry R., Chaumont M., Berti-Equille Laure, Subsol G. (2021). Assessment of CNN-based methods for poverty estimation from satellite images. In :
Del Bimbo A. (ed.), Cucchiara R. (ed.), Sclaroff S. (ed.), Farinella G.M. (ed.), Mei T. (ed.), Bertini M. (ed.), Escalante H.J. (ed.), Vezzani R. (ed.). Pattern Recognition : proceedings, part VII. Cham : Springer, 550-565. (Lecture Notes in Computer Science ; 12667). ICPR.International Workshops and Challenges, [En ligne], 2021/01/10-15. ISBN 978-3-030-68786-1.
Titre du document
Assessment of CNN-based methods for poverty estimation from satellite images
Année de publication
2021
Type de document
Partie d'ouvrage
Auteurs
Jarry R., Chaumont M., Berti-Equille Laure, Subsol G
In
Del Bimbo A. (ed.), Cucchiara R. (ed.), Sclaroff S. (ed.), Farinella G.M. (ed.), Mei T. (ed.), Bertini M. (ed.), Escalante H.J. (ed.), Vezzani R. (ed.) Pattern Recognition : proceedings, part VII
Source
Cham : Springer, 2021,
550-565 (Lecture Notes in Computer Science ; 12667). ISBN 978-3-030-68786-1
Colloque
ICPR.International Workshops and Challenges, [En ligne], 2021/01/10-15
One of the major issues in predicting poverty with satellite images is the lack of fine-grained and reliable poverty indicators. To address this problem, various methodologies were proposed recently. Most recent approaches use a proxy (e.g., nighttime light), as an additional information, to mitigate the problem of sparse data. They consist in building and training a CNN with a large set of images, which is then used as a feature extractor. Ultimately, pairs of extracted feature vectors and poverty labels are used to learn a regression model to predict the poverty indicators.First, we propose a rigorous comparative study of such approaches based on a unified framework and a common set of images. We observed that the geographic displacement on the spatial coordinates of poverty observations degrades the prediction performances of all the methods. Therefore, we present a new methodology combining grid-cell selection and ensembling that improves the poverty prediction to handle coordinate displacement.
Plan de classement
Informatique [122]
;
Télédétection [126]
Localisation
Fonds IRD [F B010085560]
Identifiant IRD
fdi:010085560
