Facebook Researchers Map Africa’s Population Density Using AI
11 April 2019
Work around Artificial Intelligence (AI) has been growing by the day and it has been front lined by big tech companies. Facebook too has been experimenting with AI in a host of different ways and this next one is very interesting.
Facebook partnered with Columbia University’s Center for International Earth Science Information Network to ensure that they get the best available administrative data for all countries involved. The Boston based Facebook team used advanced computer vision and machine learning to combine satellite imagery from Digital Globe with public census data and other sources to create detailed population density maps in Africa.
Normal high resolution satellite imagery would have required countless hours for volunteers to comb through lots of pictures to identify which contained a tiny town or remote village.
Well the Facebook team used AI t solve this problem. The computer vision they used examined 11.5 billion individual images to determine whether they contained a building. It ended up finding approximately 110 million buildings in just a few days.
This technology has been used in such places like Malawi where the maps were used to filter out 97% of the uninhabited terrain and led to coordinate the efforts of 3,000 Red Cross volunteers in the country. It has also been used by the Bill and Melinda Gates Foundation, World Bank and the Humanitarian OpenStreetMap. It has also been used by Tanzania to kick-start efforts to bring renewable electrification to rural areas.
AI use will only become more and more useful in the future and training these neural nets to derive useful information from spatial data will become invaluable in the futrue.