AI Innovation Challenge AI Assisted Collaborative Mapping Tool for Disaster Management
  • The Results

AI Assisted Collaborative Mapping Tool for Disaster Management

Challenge Completed!

Build an open-source AI-assisted mapping tool where AI models are created and trained by end users in a live environment. In this 8-week challenge, collaborative team of 50 AI engineers worked together. 

 

The problem

Each year, disasters around the world kill nearly 100,000 and affect or displace 200 million people. Many of the places where these disasters occur are literally ‘missing’ from any map and first responders lack the information to make valuable decisions regarding relief efforts. 

There is a need for map data that revolutionizes disaster management, reduces risks, contributes to the achievement of the Sustainable Development Goals, and helps humanitarian organizations that are trying to meet the needs of vulnerable people.

Our Partner – HOT develops open source apps and tools for collaborative mapping and geospatial data collection. Their tools are free for all to use and leveraged by partners such as Red Cross societies, Médecins Sans Frontières, UN agencies and programmes, government agencies, and local NGOs and communities.

 

The project goals

The project goal is to achieve an open-source AI-assisted mapping tool where AI models* are created and trained by end users in a live environment. Training datasets would be built and placed on defined locations, AI models are then created and evaluated (precision/recall) and those models will be used to detect features and push them into OpenStreetMap. End users will be able to select the Open Aerial Map imagery OAM and define a small region of the area of interest and grab the labels (with ability to modify).

* AI models in this context are features extraction (binary semantic segmentation or others) on open aerial imagery (raster) based on the training dataset & its labels.

 

The data

The data will be provided by the Partner. The Partner will deliver multiple training datasets.

Each training dataset would have the following format:

1. A list of raster images named as the following

SOURCE-x-y-z.png where, 

SOURCE refers to the source of the image, generally it is OAM

x refers to the x coordinate of the tile image in EPSG:3857

y refers to the y coordinate of the tile image in EPSG:3857

z refers to the zoom level of the tile image

The image coordinates would be convertible to geo reference for each time since its x,y & z are known.

2. Single labels.geojson file which will include the labels polygons in EPSG:4326 coordinates 

The sample training dataset is shown in the following pic:

 

Why join? The uniqueness of Omdena AI Challenges

A collaborative experience you never had in your working life! For the next eight weeks, you will not only build AI solutions to make a real-world impact but also go through an entire data science project lifecycle. This covers problem scoping, data collection, and preparation, as well as modeling for deployment.

And the best part is that you will join a global and collaborative team of changemakers. Omdena AI Challenges are not a competition or hackathon but a real-world project that will take your experience of what is possible through collaboration to a new level.

 

Find more information on how an Omdena project works

 

This challenge is hosted with our friends at

Want to work with us?

Learn more