pic 18

PulseSatellite: A collaborative tool using human-AI interaction to extract information from satellite imagery for humanitarian purposes

Summary

PulseSatellite is a collaborative web-based tool that combines cutting edge artificial intelligence with human expertise to extract the most relevant information from satellite imagery for use in humanitarian contexts. In humanitarian crises or conflict areas, accurate and timely satellite image analysis is key to supporting critical operations on the ground. The potential use cases are numerous, such as monitoring population displacement, settlement mapping, damage assessment, flood assessment and identifying the direct impact of earthquakes, volcanoes, cyclones and landslides. Until recently, this type of analysis was done by human analysts who spent hours in front of a map manually counting and classifying structures and other elements. The PulseSatellite tool aims to support this process by analysing satellite imagery assisted by mathematical models to enable optimal results. For now, the tool is available in beta version to UN entities; however, the plan is to make it available to other humanitarian practitioners going forward.

Results

PulseSatellite simplifies and reduces the process of analysing satellite images significantly while retaining a high accuracy in the results. The tool can save several days of an expert performing manual analysis of images, according to the project initiators. This increases time efficiency and reduces the amount of human resources required to use the potentials of satellite imagery in humanitarian operations, for instance, to identify the distribution of populations, destroyed infrastructure, and other relevant factors. Overall, this tool demonstrates an innovative way to better analyse satellite imagery, and – given the increased accessibility of such analysis due to lower costs – presents an opportunity for many humanitarian operations to explore the insights that satellite imagery can contribute to achieving their goals.

 

(Picture: © UN Global Pulse, PulseSatellite)

Last modified
27 Mai 2021