This AI locates dino tracks better than human paleontologists

• Scientists have used artificial intelligence to identify dinosaurs behind 3,300 preserved tracks at a fossil site in Queensland, Australia.
• humans Paleontologists It could correctly identify known dinosaur footprints 75 percent of the time, while the accuracy of the AI ​​reached 90 percent.
• Confident in the AI’s capabilities, the researchers let the machine analyze the puzzle FootprintsLanding on Ornithischians.

---

93 million years ago, dinosaurs still roamed the Earth, and Australia still clung to Antarctica as part of the Paleozoic Era. sub-continent Gondwana, and a very small Cretaceous drama that unfolded in the northeast corner of the Land Down Under. Located in modern day Queensland, Australia, near the town of Winton, the Lark Quarry Dinosaur Trackways is a remarkable fossil find containing about 3,300 fossils. dino tracks. Today, it is known as “Jurassic Park” in Australia.

For years, scientists have guessed about the meaning behind this one-of-a-kind discovery. Was this evidence of a rare dinosaur escaping from A t-rexLike theropods, or was this just a common crossing current? This question is further complicated by the fact that the three-toed (three-toed) footprints of herbivorous birds and theropod predators can be remarkably similar.

To help shed some light on the question (and take some human subjectivity out of the equation), a University of Queensland paleontologist Anthony Romelio recruits Artificial intelligence To help scientists decipher these ancient pathways. Thanks to the help of AI, Romelio and his team have confirmed that the footprints – which were originally thought to belong to… TheropodsThey actually belong to the more obedient and plant-eating ornithischians. The researchers published their findings this week in the journal Front of the Royal Society.

To get the tech ready for the big time fingerprint tracing, Romilio and His team trained the AI ​​software They are 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 footprints related to theropods and ornithischians–the two related dinosaurs that are 93 million years old. Co-author Jens Lalensac said: Royal Society.

Then, 36 new pathways were awarded to both AI and human researchers. While Romilio could only identify tracks with up to 75 percent accuracy, the AI ​​was able to correctly identify tracks Footprints 90 percent of the time, it depends Cosmos. Once they trusted the AI’s accuracy, they let the machine analyze the mystery footprints in Lark Quarry.

conclusion? tracks belong to Ornithischiansa group of dinosaurs dating back to Jurassic. This includes the likes of duck-billed hadrosaurs, horned ceratopsians, and armored Stegosauria, according to the University of California Museum of Paleontology.

This study is the first time that artificial intelligence has been used to analyze dinosaur tracks and is the most recent steps In Romelio’s 12-year journey to recreate the drama that unfolds near Winton. It’s a tough task considering Lark Quarry Dinosaur Trackways has another name –Dinosaur Stampede National Monument.

In 2010, Romelio began exploring a theory that these particular tracks weren’t from a stampede at all, but are leftovers from crossing a well-travelled stream. The idea This angered some of Romelio’s classmates At the time, he described him as an “icon warrior” who somehow fabricated his statements. Since then, Romelio has written a book about visualizing dinosaur tracks (no, like Literally wrote a book about it), and training AI to identify footprints could be the next big step in solving these Cretaceous mysteries.

“The 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 change in track that occurred after it was formed,” Lalensac said. Royal Society. “Going forward, we need robust quantitative methods rather than human intuition, and neural networks could be a real game-changer.”

It may take this growing technology in the future to explore the intimate mysteries of the past.