AI helps detect smuggled marine wildlife in airport luggage
Researchers in Australia have developed an artificial intelligence (AI) system capable of detecting smuggled marine wildlife products, including shark fins, dried seahorses, and sea cucumbers, within airport luggage scans. The research utilises existing 3D X-ray CT scanners at airports and neural network models to identify contraband hidden in luggage with 92 per cent accuracy, according to a press release from the journal Frontiers in Ocean Sustainability, which published the study on Monday (8/6).
The illegal trade in marine wildlife, estimated to generate billions of dollars annually, poses a significant threat to marine ecosystems. Unlike more widely known wildlife trafficking crimes, such as the trade in ivory or rhino horn, the illegal trade of marine species is more difficult to detect because items are often hidden within the luggage or parcels of ordinary passengers, the press release stated.
To train the algorithm, a team led by scientists from Macquarie University, Australia, conducted almost 300 scans using samples from seized illegal marine wildlife trade cases. The process simulated various methods commonly used by smugglers and real-world conditions, such as wrapping items in aluminium foil or clothing, or hiding them inside toys.
The system was able to achieve detection rates of 95 per cent for shark fins, 96 per cent for seahorses, and 86 per cent for sea cucumbers, according to the researchers.
They stated that while the technology could support border law enforcement efforts, they emphasised that it is designed to complement, rather than replace, existing detection methods. The researchers also noted several limitations, including the potential for false positives and the uneven access to advanced 3D scanners.
“We can only simulate real-world smuggling scenarios based on previously detected cases. AI is not a magic solution for detection, nor is it a replacement for human detection methods and sniffer dogs,” said Vanessa Pirotta from Macquarie University, the study’s lead author.