Special, dedicated software which uses artificial intelligence distinguishes sharks, dolphins and surfers in real time with higher accuracy than human eye. The system’s algorithm was trained by using publicly available aerial videos of sharks.
The fleet of drones will be deployed this month along east coast of Australia covering New South Wales and Queensland beaches. In case of shark detection, swimmers will be warned by alert from megaphone and a life raft, an emergency beacon will also be dropped.
The Little Ripper Group has developed the solution cooperating with researchers from the University of Technology Sydney’s School of Software. The company is currently working on “electronic shark repellent”, revealed company co-founder Paul Scully-Power.
"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.
According to statistics provided by Taronga Conservation Society Australia, in 2016 there was 26 shark attacks, including 2 fatal accidents. The new technology is a part of safety enhancing program, which aims to reduce sharks incidents as much as possible.