Scientists applaud a breakthrough in monitoring British animals that “only AI made it possible.”
In hundreds of hours of recordings, technology demonstrates its ability to identify dozens of species.
In order to identify animals and birds and to track their movements in the wild, researchers have constructed arrays of AI-controlled cameras and microphones. This technology, according to the researchers, could assist address Britain’s rising biodiversity crisis.
The robot monitors have been tested at three different places and have recorded noises and photos that computers have used to map the positions of various species and identify them. Numerous bird species were distinguished based on their songs, and AI analysis was used to locate and identify foxes, deer, hedgehogs, and bats. There are no human observers present.
Pipstrelle Bat Specie
The size of the operation is significant, according to Anthony Dancer, a conservationist with the Zoological Society of London (ZSL). “From these test areas, we have collected tens of thousands of data files and thousands of hours of audio, and we have recognised a wide variety of creatures from them. Using human observers at that size would not have been possible. It was only made feasible by AI.
The project’s test locations were selected on land next to train lines in London’s Barnes, Twickenham, and Lewisham. The sections are gated off to prevent people from wandering onto the tracks and are only sometimes visited by rail maintenance personnel. They are owned by Network Rail, which was instrumental in the project’s conception.
Therefore, it was simple to access relatively untamed territory, which was crucial for commencing our project, according to Dancer.
And now that we have shown the technology to deliver on its promise, we may move on to other domains.
More than 52,000 hectares of land are owned by Network Rail, and many of these parcels are important for preserving the country’s biodiversity.
Consider species like the Eurasian blackcap, blackbird, and great tit, suggested Neil Strong, Network Rail’s biodiversity strategy manager. All three species were identified by AI from the acoustic signals gathered by our sensors at our three test sites. “All three species demand healthy surroundings, including sufficient quantities of berries and nuts. That’s good and offers crucial standards for future biodiversity measurements.
Six different species of bats, including the common pipistrelle, were among the other critters the AI monitors located.
Dancer told the Observer that “Bats almost certainly use railway bridges for roosting.” Therefore, we can aid in their protection if we can use AI monitors to get more precise information on the precise locations of their roosts.
Strong emphasised this point. “In the past, we’ve had to make educated guesses about the numbers of local species based on the dead creatures that have been left by the track or the side of the road, such badgers. In this method, we are considerably better able to estimate population numbers.
The investigation also found that the hedgehog travels on UK train lines on a regular basis. Because they are caged in, hedgehogs are severely restricted to certain areas, according to Strong. But there are workarounds for that issue. Hedgehog highways are being built on rail lines in Scotland by making tiny holes in the bases of all newly installed fences to allow for hedgehog passage but prevent bigger animals from entering.
The New Forest and Chobham in Surrey will now be included in the expansion of the usage of AI monitors by ZSL and Network Rail. “We were pleasantly surprised with the relatively healthy levels of wildlife we found in London,” said Dancer. “On the sites that we have already tested, we found signs of more than 30 species of bird and six species of bat, as well as foxes and hedgehogs.” “However, it wasn’t actually the project’s major goal.
The goal was to show how AI-driven technology may be used to effectively survey wildlife on Network Rail land as well as in other UK locations by combining acoustic and video traps. It will detail how species are adapting to climate change and how we may manage vegetation in locations outside only close to train lines.
The main thesis is that artificial intelligence (AI) will be crucial for maintaining biodiversity as the country’s climate changes. Strong claims that “this technology will require the analysis of tens of thousands of hours of recordings and hundreds of thousands of images.” “Really, the only thing computers can do for us is that.”