As the world continues to experience the impacts of climate change, the intensity and unpredictability of natural disaster are also increasing. Scientist, governments and several organizations around the world are concerned; they are looking for innovative ideas to provide timely damage assessment after a disaster occurs. Two of the latest innovations that have been considered are machine learning and drones. Our approach speeds up the process of collecting and extracting information from big data such as drone and satellite images.

Drone and satellite images are used for performing building damage assessments. For example, after the Haiti Earthquake in 2010, building damage information was extracted from high resolution images using crowd sourcing. The process took over 600 volunteers, working for several days, to identify damage buildings on satellite images. In contrast, machine learning approach learns using the volunteers’ approach and then performs the same tasks automatically, while improving quality.

The challenges of climate change and un-anticipated urban shocks now requires that city managers, engineers, and planners have immediate access to spatial data and the ability to extract information from them. EarthNumerics partners with cities and government agencies to face these challenges.