*The following is originally written by Russell Defner, Humanitarian OpenStreetMap Team. Read the Bahasa Indonesia version by changing the language preference at the upper right corner.*
The Humanitarian OpenStreetMap Team (HOT) began a concerted effort to create a basemap of Indonesia back in 2011 for disaster preparedness and resiliency. Since then, the OpenStreetMap community in Indonesia has continued to grow and expand their expertise into applications like InAWARE. However, these maps still lack a lot of the road network that exists outside of the major cities. [In consultation with local communities, we’re working with with Facebook to quickly identify more roads using artificial intelligence (AI) tools.]
In 2016, Facebook started researching how various machine learning algorithms could be integrated into a mapping workflow. At State of the Map 2016, Facebook shared initial results on how to use AI and new mapping techniques to detect roads from satellite imagery. We are excited to collaborate with Facebook to create quality road data, validated by trained mappers and local HOT and Indonesia OSM communities.
Leveraging the AI generated roads, tools and training, combined with the local knowledge of the HOT and Local OSM Indonesian community, we are confident that we can increase the coverage of roads with the appropriate classifications.