Combining satellite imagery and machine learning to predict poverty

Neal jean, Marshall Burke, Michael Xie, W. Matthew Davis, David B. Lobell, Stefano Ermon


Poverty is the #1 UN Sustainable development goal

The elimination of poverty worldwide is the first of 17 UN Sustainable Development Goals for the year 2030. To track progress towards this goal, we require more frequent and more reliable data on the distribution of poverty than traditional data collection methods can provide.

In this project, we propose an approach that combines machine learning with high-resolution satellite imagery to provide new data on socioeconomic indicators of poverty and wealth. Check out the short video below for a quick overview and then read the paper for a more detailed explanation of how it all works.

Fine-grained maps of poverty

Using the final model that has been trained on survey data, we can estimate per capita consumption expenditure for any location where we have daytime satellite imagery. This allows us to create fine-grained maps of poverty—here are some that we put together for Nigeria, Uganda, Tanzania, and Malawi.


If you're interested in extending our work, here are some links that can get you started:


We gratefully acknowledge support from NVIDIA Corporation through an NVIDIA Academic Hardware Grant, from Stanford’s Global Development and Poverty Initiative, and from the AidData Project at the College of William & Mary. N.J. is supported by the National Defense Science and Engineering Graduate Fellowship Program and S.E. is partially supported by the National Science Foundation.