Case study: Mapping potential informal settlements areas in Tegucigalpa\n with machine learning to plan ground survey

2020/06/25 by Federico Bayle, Bayle, Federico, Damian E. Silvani +1
Environmental Science · Social Sciences · #Computers and Society (cs.CY) #FOS: Computer and information sciences #Impact of Light on Environment and Health #Land Use and Ecosystem Services #Latin American Urban Studies #Machine Learning (cs.LG) #Regional Development and Innovation

paper · pdf · doi:10.48550/arxiv.2006.14490

openalex publication_date 2020/06/25 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

Data collection through censuses is conducted every 10 years on average in\nLatin America, making it difficult to monitor the growth and support needed by\ncommunities living in these settlements. Conducting a field survey requires\nlogistical resources to be able to do it exhaustively. The increasing\navailability of open data, high-resolution satellite images, and free software\nto process them allow us to be able to do so in a scalable way based on the\nanalysis of these sources of information. This case study shows the\ncollaboration between Dymaxion Labs and the NGO Techo to employ machine\nlearning techniques to create the first informal settlements census of\nTegucigalpa, Honduras.\n

Citations

• A Coefficient of Agreement for Nominal Scales

Related

• Population Mapping in Informal Settlements with High-Resolution Satellite Imagery and Equitable Ground-Truth
• Mapping Informal Settlements in Developing Countries with Multi-resolution, Multi-spectral Data
• Uncovering large inconsistencies between machine learning derived gridded settlement datasets
• Social cartography and satellite-derived building coverage for post-census population estimates in difficult-to-access regions of Colombia
• On the development of open geographical data infrastructures in Latin America: progress and challenges