Map land cover use and building footprints at national scales.
Deliver scalable raster and vector maps online.
Maps of building footprints and croplands are crucial for many international aid and business projects with national scope, such as agricultural intensification, medical surveillance, vaccine delivery, and disaster relief. These maps enable field logistics to be carried out more efficiently, informing routing algorithms with geospatial data. Building footprints may also used in lieu of government census data which may be non-existent or unreliable.
Throughout the Global South, aid projects struggle to acquire these maps efficiently and affordably. Project implementers are often directly undertaking this challenge, executing surveys in-person with large crews of local staff, which is extremely time-consuming and expensive. Some national ministries of agriculture have also launched in-house efforts to manually label satellite imagery, but these efforts are often too slow to reach fruition in time for planned interventions.
To address these needs, QED (https://qed.ai) has built automated methods for mapping buildings and croplands across nations. By coupling deep learning with crowdsourced collection of precise training data, we can well-approximate building footprints produced by traditional manual methods. Our techniques have been executed for both public and private sector clientele and used for industrial purposes.
Maps are often the final results of geospatial studies, segmenting regions of interest and visualizing hotspots for intervention.
However, many organizations struggle with delivering their maps to general audiences, since rendering large raster and vector maps across the internet in a scalable fashion is a challenging task. This difficulty is compounded by the more limited digital infrastructure available to data consumers in the developing world.
After evaluating many tools including Geoserver and Geonode, QED designed a new solution to fulfill these needs. Features include:
Maps satisfies all of these objectives through many back-end and front-end innovations. We use Maps both to host existing gridded satellite imagery, such as that generated by AfSIS through CIESIN, as well as predictive maps generated by AI.
We express our thanks to the following donors for supporting us in the development of many of the tools described above.