This AI identifies dinosaur footprints better than human paleontologists

  • Scientists used AI to identify the dinosaurs behind 3,300 preserved footprints at a paleontological site in Queensland, Australia.
  • Human paleontologists it was able to correctly identify known dinosaur tracks 75 percent of the time, while the AI ​​achieved 90 percent accuracy.
  • Confident in the AI’s abilities, the researchers let the machine analyze the mystery. footprintslanding on the ornithischians.

Ninety-three million years ago, dinosaurs still roamed the Earth, Australia still clung to Antarctica as part of the ancient supercontinent Gondwana, and a very small Cretaceous drama played out in the northeast corner of the Land Down Under. Located in present-day Queensland, Australia, near the town of Winton, is the Lark Quarry Dinosaur Trackways, a remarkable paleontological discovery containing some 3,300 preserved specimens. dino footprints. Today, it is known as Australia’s “Jurassic Park”.

For years, scientists have guessed the meaning behind this one-of-a-kind discovery. Was this evidence of a rare stampede of dinosaurs fleeing a t rex-like a theropod, or was it just a popular stream crossing? This question is further complicated by the fact that the tridactyl (three-toed) footprints of herbivorous ornithischians and theropod predators can be remarkably similar.

To help shed some light on the question (and remove some human subjectivity from the equation), University of Queensland paleontologist Antonio Romilio recruited artificial intelligence to help scientists decipher these ancient footprints. With the help of the AI, Romilio and his team confirmed that the tracks, originally thought to belong to the theropods— actually belong to the more docile herbivorous ornithischians. The researchers published their results this week in the journal Royal Society interface.

Popular Mechanics preview All sections

To prepare the technology for the great moment of fingerprint tracking, Romilio and his team trained an AI program called deep convolutional neural networks, which use artificial neural networks to analyze large swaths of data across machine learning. The researchers then fed the AI ​​more than 1,500 dinosaur tracks related to theropods and ornithischians, the two dinosaurs associated with the 93-million-year-old puzzle. The AI ​​then combined functions to help determine categories and “possibly could find ways to distinguish these categories that humans haven’t thought of yet,” said co-author Jens Lallensack. the royal society.

Then, 36 new clues were given to both the AI ​​and the human researchers. While Romilio was only able to identify clues with up to 75 percent accuracy, the AI ​​was able to correctly identify footprints 90 percent of the time, according to Cosmos. Once they could trust the AI’s accuracy, they let the machine analyze the mysterious tracks in Lark Quarry.

The conclusion? The footprints belong to the ornithischiansa group of dinosaurs dating back to the Jurassic period. That includes duck-billed hadrosaurs, horned ceratopsia, and armored stegosauria, according to the University of California Museum of Paleontology.

This study is the first time AI has been used to analyze dinosaur footprints and the last step in Romilio’s 12-year quest to recast the drama that unfolded near Winton. It’s a tall order considering that the Lark Quarry Dinosaur Trackways has another name:Dinosaur Stampede National Monument.

footprints covered in dinosaur stampede

Covered footprints of stampeding dinosaurs about 95 million years ago. Nearly 4,000 dinosaur footprints are spread over 210 square meters. Dinosaur Stampede National Monument, Lark Quarry, West Central Queensland, Australia.

auscape//fake images

In 2010, Romilio began to explore the theory that these particular tracks were not from a stampede, but rather the remains of a heavily trafficked creek crossing. The idea outraged some of Romilio’s companions at the time, who called him an “iconoclast” who had somehow fabricated his data. Since then, Romilio has written the book on viewing dinosaur tracks (no, as He literally wrote a book about it.), and training AI to identify footprints could be the next big step in unraveling these Cretaceous mysteries.

“A track is the result of multiple factors, including the anatomy of the foot, the consistency of the substrate, the movements of the animal that made the track, and the alteration of the track that occurred after it was formed,” Lallensack said. the royal society. “To go further, we need powerful quantitative methods instead of human intuition, and neural networks can be a real game changer.”

This growing technology of the future may be needed to explore the inner mysteries of the past.

Leave a Reply

Your email address will not be published. Required fields are marked *