Australia (Commonwealth Union) – The analysis of large quantities of data continues to be made simpler with advanced data analysis tools. Artificial intelligence (AI) has demonstrated that prehistoric footprints assumed to be of a vicious dinosaur predator are actually from a timid herbivore.
A global partnership where University of Queensland paleontologist Dr Anthony Romilio applied AI pattern recognition to re-evaluate footprints of the Dinosaur Stampede National Monument in Central Queensland.
According to Dr Romilio, huge dinosaur footprints were first seen back in the 1970s at a track site known as Dinosaur Stampede National Monument, and for several years it was assumed that they were left by a predatory dinosaur, such as Australovenator, with legs nearly 2m long. He also stated that the mysterious tracks were assumed to be left behind during the mid-Cretaceous Period, approximately 93 million years back. “But working out what dino species made the footprints exactly – especially from tens of millions of years ago – can be a pretty difficult and confusing business,” he explained. “So, to crack the case, we decided to employ an AI program called Deep Convolutional Neural Networks.”
The program applied 1,500 dinosaur footprints, all of which were theropod or ornithopod in origin – the dinosaur groups applicable to the Dinosaur Stampede National Monument prints.
Deep convolutional neural networks are most frequently applied in identifying patterns in images and video. A valuable tool in analyzing large quantities of images. The findings were certain that the tracks were from the herbivorous ornithopod dinosaur.
Lead author of the Liverpool John Moores University, Dr Jens Lallensack stated that the computer assistance was essential, as the team was originally at a deadlock. “We were pretty stuck, so thank god for modern technology,” said Dr Lallensack, adding that their research team of three, found one to be pro-meat-eater, the other undecided, and the third one was a pro-plant-eater.
