Comparing spatial and spatio-temporal paradigms to estimate the evolution of socio-economical indicators from satellite images
Fiche du document
16 juillet 2023
- handle: 10670/1.jwob76
- isbn: 979-8-3503-3174-5
- hal: hal-04268542
- doi: 10.1109/IGARSS52108.2023.10282306
- IRD: fdi:010090482
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1109/IGARSS52108.2023.10282306
info:eu-repo/semantics/OpenAccess
Mots-clés
Zanzibar Tanzania Deep learning Time series analysis Estimation Predictive models Satellite images Standards Remote sensing ZanzibarSujets proches
Hours (Time)Citer ce document
Robin Jarry et al., « Comparing spatial and spatio-temporal paradigms to estimate the evolution of socio-economical indicators from satellite images », HAL-SHS : économie et finance, ID : 10.1109/IGARSS52108.2023.10282306
Métriques
Partage / Export
Résumé
In remote sensing, deep spatio-temporal models, i.e., deep learning models that estimate information based on Satellite Image Time Series obtain successful results in Land Use/Land Cover classification or change detection. Nevertheless, for socioeconomic applications such as poverty estimation, only deep spatial models have been proposed. In this paper, we propose a test-bed to compare spatial and spatio-temporal paradigms to estimate the evolution of Nighttime Light (NTL), a standard proxy for socioeconomic indicators. We applied the test-bed in the area of Zanzibar, Tanzania for 21 years. We observe that (1) both models obtain roughly equivalent performances when predicting the NTL value at a given time, but (2) the spatio-temporal model is significantly more efficient when predicting the NTL evolution.