It wasn’t long after Hurricane Laura struck the Gulf Coast Thursday that individuals started flying drones to tape the damage and publishing videos on social networks. Those videos are a valuable resource, state researchers at Carnegie Mellon University, who are working on ways to utilize them for rapid damage assessment.
By utilizing artificial intelligence, the scientists are developing a system that can immediately determine structures and make a preliminary decision of whether they are damaged and how severe that damage may be.
” Existing damage control are primarily based upon people finding and documenting damage to a building,” said Junwei Liang, a Ph.D. student in CMU’s Language Technologies Institute (LTI). “That can be slow, expensive and labor-intensive work.”
Satellite imagery does not supply sufficient information and reveals damage from just a single viewpoint– vertical. Drones, nevertheless, can gather close-up info from a variety of angles and viewpoints. It’s possible, of course, for first responders to fly drones for damage control, however drones are now widely readily available among homeowners and consistently flown after natural catastrophes.
” The number of drone videos offered on social networks soon after a catastrophe means they can be a valuable resource for doing timely damage control,” Liang stated.
Xiaoyu Zhu, a master’s student in AI and Innovation in the LTI, stated the preliminary system can overlay masks on parts of the buildings in the video that appear damaged and determine if the damage is slight or major, or if the building has actually been damaged.