Bing Maps and Microsoft Philanthropies have joined hands with Humanitarian OpenStreetMap( HOTOSM), under Microsoft’s AI for Humanitarian Action program, as per an official Bing Blog. The initiative aims to utilize the potential of AI in building open maps. The focus will be on incorporating design updates, integrating machine learning, and bringing new open building datasets into Tasking Manager.
Over the last decade, around 2 billion people have borne the brunt of natural disasters, as per the World Disasters report 2018. In 2017 alone, more than 200 million people were in need of humanitarian assistance and 18 million were displaced due to disasters. Many of these areas that are disaster-prone are not on the map, which makes it difficult to prepare for rescue, relief and rehabilitation efforts.
Ever since the Tasking Manager was created, the Humanitarian OpenStreetMap Team (HOT) has spectacularly mapped 11 million square kilometers in Africa. However, over 60% of the area still remains unmapped.
18 million footprints for Uganda and Tanzania
Leveraging their long experience in United States and Canada, Bing is releasing country-wide open building footprints datasets for Uganda and Tanzania. The datasets would be the first open building datasets for Africa and will be available for OpenStreetMap (OSM). A two-step process with semantic segmentation followed by polygonization resulted in 18M building footprints — 7M in Uganda and 11M in Tanzania, as per the Bing blog.
“Machine learning datasets for OSM need to be open. We need to go beyond identifying roads and buildings and open datasets allow us to experiment and uncover new opportunities. Open Building Dataset gives us the ability to not only explore quality and validation aspects, but also advance how ML data assists mapping.” says Tyler Radford, Executive Director, Humanitarian OpenStreetMap Team.
The data would be made available for download free of cost and can be used for research, analysis and OSM. Currently, there are 14M building footprints in Uganda and Tanzania.
“We will be making the data available on Github to download. The CNTK toolkit developed by Microsoft is open source and available on GitHub as well. The ResNet3 model is also open source and available on GitHub. The Bing Maps computer vision team will be presenting the work in Uganda and Tanzania at the annual International State of the Map conference in Heidelberg, Germany and at the HOT Summit”, says Bing Maps Team.