Koala 'facial recognition' in tests at Australia
Griffith University researchers are aiming to use artificial intelligence (AI) to identify specific furry species using wildlife road crossings in Queensland.
Underground paths and bridges are created near koala habitats near busy roads to provide them with a safer route away from automobiles.
The pilot study's chief researcher, associate professor Jun Zhou, expressed hope that AI would eliminate the need for manual camera checks to determine which species were using the bridges.
"Now, with artificial intelligence developing very quickly over the past 10 years, the technology is powerful enough to help recognize not only koalas generally, but which individual koalas are using the crossings," he said.
Koalas have previously been tracked using identity tags and GPS.
According to Griffith University, the researchers will collaborate with conservation organizations to train AI to recognize specific animals based on their looks and movements.
They believe that the information will help them better understand how koalas utilize wildlife crossings and whether they may be better placed to safeguard the animals from being hit by motorists.
In July, the government-funded initiative will begin at 20 camera locations around Brisbane.
Koalas are found along Australia's eastern coast, but they are endangered and face a variety of problems, including habitat loss due to logging, development, and bushfires, which ravaged large areas of the country in 2019-2020.