AI-powered shark drones help protect Australia’s beaches

admin32325 views
Spread the love

Using artificial intelligence to distinguish sharks from dolphins and surfers in real time, the Little Ripper drones are able to warn swimmers of what lies beneath.

When you think of Australia, you think of sunshine, excellent beaches, and a variety of deadly animals that could bite, poison, or sting you at any time.

A group of researchers had that in mind when they created a drone capable of detecting sharks underwater, quicker than the human eye can, and at a higher accuracy level.

The SharkSpotter technology uses algorithms to identify sharks in a live video feed that’s provided in real-time by the Little Ripper Livesaver drone that’s flying above the water. This world-first algorithm, developed using deep neural networks and artificial intelligence, allows SharkSpotter to tell apart rays, dolphins, and other marine animals from sharks – surfers too. The drone is equipped with an onboard megaphone, so is also able to warn swimmers below about what’s swimming below them – well before they themselves detect the threat.

In 2016 Australia saw 26 shark attacks, including two fatalities, according to Taronga Conservation Society Australia’s statistics. While arguably quite low numbers (especially when looked in the context of how many swimmers head to Australia’s beaches every summer), the spectre of shark attacks continues to loom over the collective Australian consciousness.

With this new technology, there’s the potential to detect creatures before a human-shark encounter comes close to being a risk.

The Little Ripper Group is the brainchild behind the drones, working in partnership with researches from the University of Technology Sydney’s School of Software over the past year to create this shark-spotting tech.

UTS Professor Michael Blumenstein said “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.”

“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 drone are expected to commence regular patrols along beaches in Queensland and New South Wales from September, just in time for the beginning of the Surf Life Saving patrol season.