In what is understood to be a world-first trial, the first real-time personal shark alert system will go live in a test above Illawarra beaches.
The SharkEye system, part of the SMART Infrastructure Facility’s Digital Living Lab at the University of Wollongong, uses a camera mounted on a blimp to send alerts direct to people wearing smart watches in the water.
Researchers at the University of Wollongong used a Shark Management Strategy grant from the NSW Department of Primary Industries to develop an algorithm capable of recognising shark shapes in the water and to tell people nearby via an app. SharkEye utilises this algorithm and brings together a range of monitoring projects in the region.
Project leader, Associate Professor Robert Gorkin said uses of the system went far beyond shark detection.
“We know the communications work and we now have a platform that can be used to continuously monitor beaches,” he said. “We are also training the algorithm for other types of animals such as rays.
“There are many other possible uses, from looking for swimmers and surfers who are in trouble, to monitoring pollution, weather events, surf conditions and surfing events – the possibilities are endless.
“As far as I know, this is the first system that provides personal alerts to smart watches, and allows surfers or swimmers to communicate directly with people on the beach.”
The SharkEye platform represents years of work from multiple shark monitoring projects that have been developed across the region.
The aerial image acquisition utilised as part of SharkEye, was developed by Kye Adams a doctoral student in the UOW biology department, on the Aerial Inflatable Remote Shark Human Interaction Prevention blimp or Project AIRSHIP for short. Mr Adams, who also works as a lifeguard, has operated the blimp for the last two summers over Kiama to understand shark activity.
The SharkMate app (available in the Apple app store for iPhone) is also utilised as part of SharkEye as the notification hub for the project. The app was developed by an Illawarra teenager, Sam Aubin. Sam started developing his app at the age of 13 as a means of promoting shark conservation and to increase the public’s knowledge of shark locations. The app uses a range of real-time data – air temperature, water temperature, time of day, the presence of lifesavers and the proximity to wastewater treatment plants – to predict the likelihood of a shark attack.
The two technologies were connected to an algorithm for real-time automatic shark detection trained on images collected by the blimp, which was developed by UOW’s Advanced Multimedia Research Lab (AMRL) led by Professor Wanqing Li, an expert in computer vision and machine learning.
Dr Matthew Berryman Honorary Senior Fellow at the SMART Infrastructure Facility and CEO of Across the Cloud, created the architecture for the cloud platform and integrated the hardware and software components together. It was Dr Berryman who also enlisted Amazon Web Services (AWS) as another partner on the work.
“This project is possible because of the rapid advancements and availability of Machine Learning & AI, cloud & edge computing and the Internet of Things (IoT), said Professor Gorkin.
“There are a host of new options for minimising harm to beachgoers, sharks and other wildlife alike – particularly with new imaging and drones. With over 50,000kms of coastline to monitor in Australia, the solution will likely be a mix of current and future technologies.”
“This week we are demonstrating the viability of our platform to deploy within this emerging ecosystem and once the algorithm is fully tested, it could provide a platform for use in environments as diverse as agriculture, security and fire management.
The SharkEye project is a core part of SMART’s Digital Living Lab, which is harnessing the Internet of Things to enhance community development and wellbeing.