All eyes on koala conservation thanks to clever QUT algorithm
Over the past eighteen months, with the devastating bushfires estimated to have killed or displaced nearly three billion animals, saving koalas and other Australian wildlife requires urgent action to prevent further decline of populations and habitat.
As many of our native species inch closer to extinction at alarming rates, Associate Professor Grant Hamilton and his team offer the gift of hope through their latest research. The team has developed a state-of-the-art mapping technique which delivers a consistent and efficient method for managing wildlife populations.
Professor Hamilton’s team uses drones and artificial intelligence (AI) detection to spot koalas from above – a faster and more precise method than traditional search approaches.
‘To save threatened species, we need to be able to count them – to see if our actions are helping or harming the populations. Our work is making this more accurate and cost-effective to survey large areas,’ said Professor Hamilton.
The method for detecting koalas in bushland begins by drones fitted with infrared cameras covering an area in a “lawnmower” pattern early in the morning and during the colder months of the year, when koala heat signatures are easier for the algorithm to identify.
Thanks to the generosity of over 130 donors last year, Professor Hamilton’s team can survey more land, ultimately assisting koala recovery and informing conservation decisions.
‘Climate change is one of many threats that have the potential to cause immense harm to wildlife in Australia and globally – particularly as it leads to a greater number of more intense bushfires.
‘In the next six to twelve months, we hope to develop an algorithm that will detect multiple species with a single drone flight.’
As the team develops new processes that enable more species to be detected, this becomes an even more powerful approach for monitoring and saving our wildlife and the environment.
If you would like to make a real-world difference for Australian wildlife, visit our website.