2021-11-19 –, Room 1
Language: English
From mid-2020 to 2021, a number of mappers in both Uganda and the DRC joined a participatory buildings footprint import. The data was produced by Ecopia based on an imagery mosaic provided by Maxar and contained more than 6.6 millions building footprints. During this session, we’ll explain what worked well and less well and what we learned from the process of integrating this data into the OpenStreetMap database.
The import process coordinated by the Humanitarian OpenStreetMap Team (HOT), together with HOT Uganda and OSM DRC, involved the use of the OSM wiki, the HOT Tasking Manager and OSM Sharp server. For editing, the contributors used JOSM Editor with several plugins to ease the conflation process. The instructions were developed to ensure that the import would result in data which would be as good, if not better, than what is usually produced by manual digitizing/traditional remote mapping. Each building proposed for import was to be compared with the project-specific Maxar mosaic and with the data already existing in OSM.
We’ll describe the OSM import guidelines that were followed, then the specific challenges met with this dataset extracted through machine learning will be explained. Considering the growing availability of such AI-related datasets, we’ll review common errors and how we adapted to them and to other issues such as imagery offsets, heterogeneity of existing data and other context specific challenges.
Eventually we’ll propose recommendations regarding this type of editing and aspects to consider before starting such imports.
