Workshop  /  December 12, 2022  -  December 16, 2022

Workshop Within the Project Edda

Project Edda: Artificial Intelligence Supports Faster Evaluation of Drone Images

After disasters such as earthquakes, tsunamis or hurricanes, relief supplies have to be distributed as quickly as possible in crisis areas and needy people rescued or supported. However, destroyed infrastructures pose a logistical problem and make it difficult for aid organizations to assess the extent of the disaster, the number of people in need of help and possible rescue routes. The EDDA project aims to use artificial intelligence and unmanned drones to provide real-time inspection of disaster areas with as much coverage as possible.

One of the biggest challenges in the EDDA project is the acquisition of a data set so that the EDDA software can evaluate the situation in the disaster area. 

As part of this project, our colleagues Damjan Hatic and Dr. Petra Gospodnetic will be holding a workshop in Maputo, Mozambique, where Cyclone Idai (tropical storm) originated in 2019. The event is organized in coordination with our project partners, WFP (World Food Program) and INGD (Mozambique National Institute for Disaster Risk Management and Reduction). 
 

The Workshop Has the Following Objectives:

  •  Promote collaboration with our project partners
  •  Promote the development carried out under the EDDA project.
  • Prepare the data set received by WFP following the occurrence of Cyclone Idai in Mozambique in March 2019.

The outputs of the workshop will be ordered and labeled datasets of ortho-mosaic images with their respective building footprints and damage classification labels. These datasets will be used as part of the EDDA project to improve building segmentation and building damage classification models. In addition, they will be published to support further research in this area.