Measuring Poverty with Machine Roof Counting

Pulse Lab Kampala

Project Description

In Uganda, the type of roof a house has, is used as a proxy-indicator for poverty by the Uganda Bureau of Statistics. Traditional thatched roofs harbor pests and disease and are high maintenance. And so as soon as someone can afford to, they will upgrade to a metal of tiled roof. At the same time, a wide variety of remote sensing image sources from highly light-sensitive satellites, high-resolution (<1m) image data, nano-satellites, balloon mapping, and low-cost unmanned aerial vehicles is available for measuring sustainable development.

Pulse Lab Kampala worked on a project to use satellite imagery to identify the roofs, and develop  image processing software to count the roofs and identify the type of material they are constructed from. A user-friendly tool (on-line dashboard) that provides proxies for poverty monitoring based on household’s roof counting will be built. This tool and methodology will complement existing statistical tools that use surveys and primary data collection to assess poverty levels. Analysing real-time data on individual expenditure for development is a new way to study population groups. Without the biases that can be derived by the design or implementation of household surveys, the new data generated with the automated roof top counting can provide new insights on household economies.


Did you find this project interesting? Share it with your networks!

Below you can find our latest examples of our collaborative research, prototypes and experiments, where we analyse digital data to advance global development, support humanitarian action, and promote peace. For more, go to the research projects page.

Pulse Lab New York

UN Open GIS Initiative: Geo-AI Working Group

UN Global Pulse has joined the UN Open GIS Initiative as co-chair of a newly formed Geo-AI Working Group alongside the Food and Agriculture Organization (FAO). UN Open GIS is

Scroll to Top