When two teenagers entered the surf off the northern NSW town of Lennox Head in Australia earlier this year, little did they suspect that they would run into trouble in the crashing waves – and even less so, that it would be a drone that saved them.
In a world first, the incident made news around the world, including here on Drone Below.
Now, the ‘Little Ripper Drone’ surf lifesaving drone developed by Westpac Rescue and The Ripper Group has been recognised by the Australian Information Industry Association at their annual iAwards for innovation.
Specifically, the drone and its software tech won the award for Artificial Intelligence or Machine Learning Innovation of the Year.
A result of the successful partnership between the two companies and the University of Technology Sydney (UTS), the Little Ripper Drone uses a specially developed SharkSpotter© AI algorithm.
In the case of the Lennox Head incident, the drone was used by dropping a lifesaving device to the teenagers.
But it was this algorithm, able to accurately detect and distinguish different marine creatures, that in particular won the award.
It is important because in Australia, the high number of ocean swimmers means that shark attack is a real risk (although still a lower risk than other mundane activities such as driving a car).
“It can detect really accurately the difference between a shark and a dolphin, sharks from so many types of marine life,” said Professor Michael Blumenstain from the UTS School of Software.
Blumenstain’s colleague Dr Nabin Sharma explains the cutting edge technology behind the drone’s algorithm.
“SharkSpotter automated system for detection and identification of sharks in particular, and marine life/objects more generally, uses cutting edge deep learning neural networks and image processing techniques for object recognition and classification,” he said.
With the Australian beach season quickly approaching, the drones and their shark detection systems will again start patrolling Aussie shores next month.
Providing a live feed to drone operators assisting the surf lifesavers who are present throughout the Australian summer to ensure the safety of swimmers, the 90% accurate shark-spotting software will allow risks of shark attack to be identified faster and more accurately than ever before.
“SharkSpotter© is an example of how an AI application assists humans as it has significantly higher rates of visual accuracy in shark detection at @90 per cent. At this level of accuracy the drone will certainly help us to improve detection rates, and to maintain safer conditions for water users,” Sharma concluded.
Following the national award event, the Little Ripper will now represent Australia in this category at the international awards in Guangzhou, China, from 9 to 13 October 2018.
A video produced in 2016 describes the potential that the Little Ripper offers for improving the safety of swimmers in Australian waters.