Please register to access this content.
To continue viewing the content you love, please sign in or create a new account
Dismiss
This content is for our paying subscribers only

UAE Environment

Dubai invites start-ups for ‘flood detection’ climate challenge

Challenge is open to start-ups offering rewards of AWS credits worth US$40,000



The Mohammed Bin Rashid Space Centre (MBRSC) announces the launch of ‘Climate Change Challenge: Floods Detection’ for start-ups.
Image Credit: MBRSC

Dubai: The Mohammed Bin Rashid Space Centre (MBRSC) has announced the launch of an innovative initiative ‘Climate Change Challenge: Floods Detection’ in collaboration with Amazon Web Services (AWS).

Th aims is to encourage start-ups to create innovative solutions to tackle pressing environmental issues, at the upcoming Dubai Air Show 2023, scheduled for 13-17 November.

The challenge aims to spur development of cutting-edge tools leveraging artificial intelligence, satellite imagery, and data analytics to detect and monitor the impacts of flooding caused by sea level increases. The competition is open to start-ups from around the world and offers AWS credits of US$25,000 for the first place, US$10,000 for the second place, and US$5,000 for the third place.

Read More

The challenge

Rising sea levels have significantly increased the vulnerability of coastal regions, leading to severe floods, damage to infrastructure, loss of lives, and displacement of communities. Coastal ecosystems, vital for storm surge protection and habitat sustenance, are also under threat. Efficient detection and monitoring tools are essential in addressing the looming challenge.

Advertisement

The challenge calls on start-ups to develop an intuitive tool harnessing the capabilities of satellite imagery and AI to provide precise flood impact visualisations. The tool should differentiate flood patterns from satellite imagery, analysing changes in land cover, water levels, and infrastructure.

How to participate

Startups can access the publicly available flood mapping data set, “OMBRIA,” which includes Sentinel-1 and Sentinel-2 imagery sets of pre and post-flood events. The data set is available at https://github.com/geodrak/OMBRIA. Submissions should present results as a binary mask, differentiating flooded from non-flooded regions. Proposals will be assessed based on precision, recall, overall accuracy, F-measure, Intersection over Union (IOU), processing time, and innovation.

Entries can be sent no later than 31 October. The finalists will be announced on 6 November, who will be invited to pitch their solutions on 16 November during the Dubai Airshow 2023.

Advertisement