Population Mapping and the Future of Image Interpretation

I’ve always been interested in the world’s population and how we can continue to meet basic needs in every corner of the globe.


The image above and tile of the article “Facebook Is Making a Map of Everyone in the World” (http://www.theatlantic.com/technology/archive/2016/02/facebook-makes-a-new-map-of-everyone-in-the-world/470487/) caught my attention and I had to give it a read. This article explains how Facebook is attempting to identify where new potential users without internet access live and then using that information to adopt solutions to connect those populations to the internet. Through this process, Facebook’s Connectivity Labs has developed neutral-net algorithms that recognize what a building looks like from above (using aerial imagery) and then assume people live there. With satellite launches becoming cheaper, more companies providing imagery and advanced algorithms extracting information at a faster pace the cost will continue to drop giving more us information to utilize.

The criteria presented in this article to identify local populations through building sites leaves a large gap. Buildings, cities and regions can be broken down by zoning type (industrial, commercial, residential, etc.). These maps are great starting point to understanding exactly where the world population is living but not to the detail where it could be in the future for planning purposes


The article makes you think of what is on the horizon for image interpretation and mapping, and how the two are becoming more accurate and cheaper than ever before. While this article gives some insight on how Facebook plans to use these tools to gain more users, there is great potential for information that can be extracted from a highly accurate world population map. By providing accurate population data in undeveloped regions steps can be taken to plan for future utility development for clean water, how to plan for emergency management when disasters strike or predicting how much food needs to be produced in a region to support the surrounding population and where to distribute it. Having detailed accurate data can provide an endless supply of solutions.