DeepWealth: A generalizable open-source deep learning framework using satellite images for well-being estimation

Author(s)

Ben Abbes, Ali ; Machicao, Jeaneth ; Correa, Pedro L. P. ; Specht, Alison ; Devillers, Rodolphe ; Ometto, Jean P. ; Kondo, Yasuhisa ; Mouillot, David

• Document type: Journal article
• Source: SOFTWAREX; v. 27, p. 8-pg., 2024-06-21.

Abstract

Measuring socioeconomic indices at the scale of regions or countries is required in various contexts, in particular to inform public policies. The use of Deep Learning (DL) and Earth Observation (EO) data is becoming increasingly common to estimate specific variables like societal wealth. This paper presents an endto-end framework 'DeepWealth' that calculates such a wealth index using open -source EO data and DL. We use a multidisciplinary approach incorporating satellite imagery, socio-economic data, and DL models. We demonstrate the effectiveness and generalizability of DeepWealth by training it on 24 African countries and deploying it in Madagascar, Brazil and Japan. Our results show that DeepWealth provides accurate and stable wealth index estimates with an R 2 of 0.69. It empowers computer-literate users skilled in Python and R to estimate and visualize well-being-related data. This open -source framework follows FAIR (Findable, Accessible, Interoperable, Reusable) principles, providing data, source code, metadata, and training checkpoints with its source code made available on Zenodo and GitHub. In this manner, we provide a DL framework that is reproducible and replicable. (AU)

• FAPESP's process: 22/14429-8 - Building new tools for data sharing and re-use through a transnational investigation of socioeconomic impacts of protected areas
• Grantee: Ali Ben Abbes
• Support Opportunities: Scholarships in Brazil - Technical Training Program - Technical Training
• FAPESP's process: 17/22269-2 - Transition to sustainability and agriculture-energy-water nexus: exploring an integrated approach with case studies in the Cerrado and Caatinga
• Grantee: Jean Pierre Henry Balbaud Ometto
• Support Opportunities: Research Program on Global Climate Change - Thematic Grants
• FAPESP's process: 18/24017-3 - Building new tools for data sharing and re-use through a transnational investigation of the socioeconomic impacts of protected areas (PARSEC)
• Grantee: Pedro Luiz Pizzigatti Corrêa
• Support Opportunities: Regular Research Grants
• FAPESP's process: 20/03514-9 - Evaluation the effects of Brazilian protected areas in local communities based on the use and re-use of biological, environmental and socioeconomic data
• Grantee: Marina Jeaneth Machicao Justo
• Support Opportunities: Scholarships in Brazil - Post-Doctoral