Shark spotting is going high-tech with AI-powered drones.
Peter Parks/Getty ImagesAustralia is famous for a few things: sunshine, excellent beaches and a huge array of deadly animals that will bite/poison/sting you at a moment’s notice.
With that in mind, a group of researchers have created a new shark-detecting drone capable of finding the apex predators underwater, quicker than the human eye and with a higher level of accuracy.
The Little Ripper drone’s SharkSpotter technology uses deep learning to detect sharks below the waves.
Little RipperThe technology, known as SharkSpotter, uses an algorithm to detect sharks in a live video feed recorded in real time by a drone (known as the Little Ripper Livesaver) flying above the water. Using a world-first algorithm, developed using artificial intelligence and deep neural networks, SharkSpotter is able to distinguish sharks from dolphins, rays and other marine animals, and even surfers. Thanks to an onboard megaphone, the drone can also warn swimmers about what’s lurking in the water before they’ve even seen the threat.
Australia saw a total of 26 shark attacks in 2016, including two fatalities, according to statistics from Taronga Conservation Society Australia. But while that number is arguably low (particularly compared to the number of swimmers that hit Australia’s beaches every summer) the spectre of shark attacks looms large over the Australian consciousness.
But the new technology has the potential to more accurately detect the animals before a potential shark-human encounter even becomes a risk.
The drones are the brain-child of The Little Ripper Group, which worked in partnership with researchers from the University of Technology Sydney’s School of Software for more than a year to develop the shark-spotting technology.
UTS Professor Michael Blumenstein said the technology will have a real positive impact for the public by using technology to make beaches safer.
“The automated system for detection and identification of sharks in particular, and marine life/objects more generally, was developed using cutting edge deep neural networks and images processing techniques,” said UTS Professor Michael Blumenstein.
“The system efficiently distinguishes and identifies sharks from other targets by processing video feeds that are dynamic as well as images, where objects are static.”
The drones will commence regular patrols along beaches in New South Wales and Queensland on Australia’s east coast from September, for the start of the Surf Life Saving patrol season.
Tech Culture: From film and television to social media and games, here’s your place for the lighter side of tech.
Batteries Not Included: The CNET team shares experiences that remind us why tech stuff is cool.
