Detecting trafficked wildlife with 3D X-rays and AI

Australia prides itself on being home to some of the most diverse and unique wildlife in the world, but with this wealth comes a market for the illegal wildlife trade.

Existing techniques for detecting illegally trafficked wildlife include x-ray scanning, physical detection by border guards, and the use of biosecurity dogs.

Now Australian scientists have found that 3D X-ray technology and artificial intelligence (AI) algorithms can be used together to detect trafficked wildlife hidden in luggage or other cargo.

The team created a 3D scanned “reference library” for three types of wildlife – lizards, birds and fish – which they used to teach artificial intelligence (AI) algorithms to recognize the animals.

The AI ​​achieved an 82% detection rate with a miss rate of just 1.6%.

The study recently published in the journal Frontiers in Conservation Science Human-wildlife interactionsis the first to document the use of 3D X-ray CT scanning technology to protect wildlife.

READ:  Government Bill to remove EU retained laws branded ‘an attack on nature’ by Kent Wildlife Trust – The Isle Of Thanet News

Continue reading: Reptiles threatened by online trade.

“Taking animals from the wild poses risks to the conservation of species, local populations, habitats and ecosystems, and preventing the illegal wildlife trade into Australia protects our unique natural environment from exotic pests and diseases,” says Sam Hush, acting Deputy Minister for Environment Compliance at the Department for Climate Change, Energy, Environment and Water (DCCEEW).

wild animals discovered
Example of a successful lizard detection via the wildlife algorithm. This is a screenshot of the user interface alerting the operator to a spotted shingleback lizard (Tiliqua rugosa) over the green bounding box that labeled this a lizard. Source: Pirotta, V. et al./Boundaries

“It’s also extremely cruel. Smuggled animals often suffer from stress, dehydration or starvation, and many die in transit. We have worked with the Department of Agriculture, Fisheries and Forestry (DAFF) to test and validate the 3D x-ray and wildlife algorithms, both of which have proven to be very effective.”

READ:  Florida wildlife agency expected to shoot down commercialized turtle breeding proposal

There is also a biosecurity element. “Illegal wildlife trade poses a significant biosecurity risk for Australia as it could introduce pests and diseases that could impact the environment, human and animal health,” said Dr. Chris Locke, Associate Secretary of the Biosafety and Compliance Group at DAFF.

The research team used 3D X-ray CT technology with real-time tomography (RTT) in the study. This is a technique that uses X-rays to create a series of cross-sectional images through an object, in this case an animal, that can be manipulated to provide a 360-degree view of it.

The deceased specimens were scanned using a system used for explosives detection testing, currently deployed at international borders and at airports and mail freight facilities worldwide.

READ:  Hurricane Ian: Officials warn animals may be 'more visible' after storm

Read more: Should we keep native Australian animals as pets?

The library included 294 scans of 13 different species in different scenarios – from an animal alone to more complicated and realistic trading scenarios – which were then used to teach AI algorithms to recognize the animals.

“This paper demonstrates the limitless potential of the 3D X-ray algorithm in preventing the illegal trade in exotic wildlife, protecting Australia’s agricultural industry and unique natural environment from exotic pests and diseases.

“This innovative technology is an invaluable complementary platform to our existing biosecurity and wildlife detection tools at Australia’s international borders, with potential global applications in the future,” said Locke.

Source link