Wildlife monitoring in Australia could get a boost from a new platform that uses AI and computer vision to speed up the processing of millions of camera trap images being collected across the country.
The national initiative named the Wildlife Observatory of Australia (WildObs) is a way to collect, store and share camera trap data at scale, while improving collaboration between scientists, governments and environmental groups, according to the WildObs website.
The platform is being developed by researchers at the University of Queensland (UQ), with backing from the Australian Research Data Commons, Queensland Cyber Infrastructure Foundation and the Terrestrial Ecosystem Research Network.
Camera traps are commonly used to monitor wildlife globally: they’re easy to set up and can be left at locations for long periods, providing an invaluable window into the natural world. Across Australia, thousands of projects collect millions of images, Matthew Luskin, associate professor at the UQ School of the Environment and director of WildObs, said in a statement.
However, processing the images and identifying species takes time, money and computing power. WildOBS plans to speed it up.
“In conservation, timing matters and detecting problems early can mean the difference between recovery and extinction,” Luskin said.
WildObs requires users of the platform to upload images, which get stored and processed in the cloud. The platform’s models have been trained specifically to identify species found in Australia and can help track biodiversity trends, monitor invasive species and identify conservation priorities, according to the UQ statement.
“In one collaborative space, the WildObs platform now hosts all of Australia’s AI computer vision models. These have been trained specifically for Australian animals and environments — they can identify hundreds of species in camera trap images, 10 times faster than people,” Luskin said.
Meredith Palmer, an expert in camera trapping and conservation tech at Yale University, U.S., who was not involved in the project, told Mongabay by email: “The fields of ecology and conservation science have suffered in the era of big data due to silos between organizations and institutions, so an infrastructure that helps break down these barriers, standardize information, and encourage data sharing is an impressive step forward in this space.”
Luskin said in the statement, “Better data use can directly improve conservation outcomes — more effective protection of threatened species, smarter investment in conservation, and stronger environmental reporting.”
The value of data on WildObs could extend beyond its direct research and conservation use in Australia. Roland Kays, a research professor at North Carolina State University, U.S., who was not involved in the project, told Mongabay by email that it’s “great to get Australian camera trappers ‘on the map’ since much of their data has been unavailable for global comparisons, leaving a blank space in those papers.”
Banner image: A purebred dingo caught on camera trap on K’gari, the world’s largest sand island. Image by Zachary Amir/University of Queensland.
Credits
