Researchers at the University of Dundee in the U.K. are developing an early-warning system for flood-prone communities using a combination of Twitter, citizen science, and artificial intelligence techniques. The researchers have demonstrated how AI can be used to extract data from Twitter and crowdsourced information from mobile devices to create hyper-resolution monitoring of urban flooding. The team found that social media and crowdsourcing can be used to improve datasets based on traditional remote sensing and witness reports of flood-prone areas.
"We were particularly interested in the increased incidence of what we call sunny day flooding — flooding that occurs in the absence of any extreme weather event due to the mean sea level being higher," says University of Dundee researcher Roger Wang.
The researchers used computer-vision techniques on data collected from the MyCoast crowdsourcing app to automatically identify images of flooding posted by users.
From University of Dundee
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found