OpenAerialMap: Building a Global Hub for Free Aerial Imagery

Episode Summary In this episode, host Inong Benson sits down with Cristiano Giovando and DK to discuss **OpenAerialMap (OAM)**, an open repository for openly licensed aerial and satellite imagery. Described as "OpenStreetMap but for imagery," OAM provides a centralized platform for drone pilots, satellite companies, and researchers to share and access high-resolution data for the common good. The guests delve into the platform's humanitarian origins, the cutting-edge technology powering it, and the future of open imagery in the age of AI. --- ### **Key Topics and Highlights** * **What is OpenAerialMap?** OAM is a set of tools for searching, sharing, and using open aerial and satellite imagery. It was born out of the 2010 Haiti earthquake response when responders lacked a simple way to make donated imagery accessible to volunteer mappers. * **The Technology Stack** * **Cloud Optimized GeoTIFFs (COGs):** OAM uses COGs to allow large images to be rendered on the fly without needing to store massive pre-processed tile sets. * **Dynamic Tiling:** Leveraging AWS Lambda, the platform tiles images in real-time, making it highly efficient for web display. * **Federated Network:** While HOT hosts much of the data, OAM is designed as a "network of databases," allowing providers to host their own data via the Open Imagery Network (OIN) and connect it via API. * **Who is Using OAM?** * **GIS Analysts:** For downloading raw data to use in tools like QGIS for spatial analysis. * **Mappers:** To provide custom imagery layers for OpenStreetMap (OSM) via Tasking Manager. * **Humanitarian Responders:** Recent examples include mapping the Turkey and Syria earthquakes using Maxar imagery and flood risk assessments in Dar es Salaam through the **Dar Ramani Huria** project. * **The Future: OpenAerialMap v2** The team is currently working on modernising the platform’s infrastructure (containerisation and serverless tech) and implementing the **STAC (SpatioTemporal Asset Catalog) API** for better data interoperability. They also envision imagery becoming a primary input for machine learning to automatically extract features like buildings and roads. --- ### **Links and Resources Mentioned** * **OpenAerialMap Website:** [openaerialmap.org](https://openaerialmap.org) * **OpenStreetMap (OSM):** [openstreetmap.org](https://www.openstreetmap.org) * **Disaster Ninja (to view OAM mosaics):** [disaster.ninja](https://disaster.ninja) * **Humanitarian OpenStreetMap Team (HOT):** [hotosm.org](https://www.hotosm.org) * **GitHub Repositories:** [github.com/hotosm](https://github.com/hotosm) --- ### **How to Contribute** * **Upload Imagery:** Drone pilots and satellite providers can upload three-band GeoTIFFs directly to the platform. * **Join the Community:** Participate in the HOT Slack channel (slack.hotosm.org) or contribute to documentation and bug reporting on GitHub. * **Test & Feedback:** Provide user stories or participate in beta testing for OAM v2.