Sign up for CNN’s Wonder Theory Science newsletter. Explore the universe with news on exciting discoveries, scientific advances and more..
CNN
–
“Do you think every fingerprint is actually unique?”
That’s a question a professor asked Gabi Guo during a casual conversation while he was stuck at home during the Covid-19 lockdown, waiting to start his freshman year at Columbia University. “Little did I know that communication would become the center of my life for the next three years,” Gu said.
Gu, now an undergraduate senior in Columbia’s computer science department, led a team that conducted a study on the topic, with University at Buffalo professor Wenyao Xu as one of its authors. Published this week in the journal Science Advances, the paper appears to reveal a long-accepted truth about fingerprints: They are not, Gu and his colleagues say, all unique.
In fact, journals rejected the work several times before the team appealed and finally accepted it into Science Advances. “Initially there was a lot of pushback from the forensics community,” recalled Gove, who had no forensics background before the study.
“For the first iteration or two of our paper, he said it’s a well-known fact that no two fingerprints are alike. I guess that really helped refine our study, because we Just kept putting more data into it, (increasing accuracy) until the evidence became irrefutable,” he said.
To achieve its surprising results, the team used an artificial intelligence model called a deep convolutional network, which is commonly used for tasks such as facial recognition. The researchers added their own twist and then fed it a US government database of 60,000 fingerprints in pairs that sometimes belonged to the same person (but from different fingers) and sometimes belonged to different people.
As it worked, the AI-based system found strong similarities between fingerprints from different fingers of the same person and was therefore able to tell when the fingerprints belonged to the same person and when they did not. , with precision for a couple. Reaching 77% – apparently proving that every fingerprint is “unique”.
“We found a rough explanation for why this is: the angles and curvature at the center of the fingerprint,” Go said.
He added that over hundreds of years of forensic analysis, people have been looking for various features called “minutiae,” the branches and endpoints in fingerprint ridges that would have been used as traditional markers for fingerprint identification. are “They are great for fingerprint matching, but not reliable for finding correlations between fingerprints of the same person,” Gu said. “And that’s the insight we had.”
The authors stated that they were aware of potential biases in the data. Although they believe the AI system works similarly across genders and races, for the system to be usable in real forensics, the research requires more careful analysis of large and extensive databases of fingerprints. Verification is required.
However, Gu said he believed the discovery could improve criminal investigations.
“The most immediate application is that it can help generate new leads for cold cases, where the fingerprints left at the crime scene are of different fingerprints than those on file,” he said. “But on the other hand, it’s not just going to help catch more criminals. It’s actually going to help innocent people who no longer have to be investigated unnecessarily. And I think that society is a win.
Applying deep learning techniques to fingerprint images is an interesting topic, according to Christophe Champaud, a professor of forensic science at the School of Criminal Justice at the University of Lausanne in Switzerland. However, Champod, who was not involved in the study, said he did not believe the work revealed anything new.
“Their argument that these patterns are somehow connected between fingers goes back to the earliest days of fingerprinting, when it was done manually, and has been documented for years,” he said. Is.” “They oversold their paper, I think, because of a lack of knowledge. I’m glad they’ve rediscovered a well-known thing, but basically, it’s in a teacup. There’s a storm.”
In response, Gu said that no one had systematically measured or used the similarity between fingerprints from different fingers of the same person to the extent that the new study did. Is.
“We are the first to clearly identify that the similarity is due to the orientation of the ridge at the center of the fingerprint,” Gu said. “Furthermore, we are the first to attempt to match fingerprints from different fingerprints of the same person, at least with an automated system.”
Gabi Go/Columbia Engineering
The system used in the study to identify similarities between fingerprints could be useful in crime scene analysis, the authors said.
Simon Cole, a professor of criminology, law and society at the University of California, Irvine, agreed that the paper was interesting but said its practical utility had been overstated. Cole was also not involved in the study.
“We weren’t ‘wrong’ about the fingerprints,” he said of the forensic experts. “The unproven but intuitively true claim that no two fingerprints are ‘exactly the same’ does not deny that fingerprints are identical. Fingerprints from different people, as well as from the same person, are always identical. They seem to be the same.
The newspaper said the system could be useful at crime scenes in which fingerprints are from different fingers than those in police records, but Cole said that would only be in rare cases. It can, because all 10 fingers and often the palms are routinely recorded when prints are taken. “It’s not clear to me when they think law enforcement will have only some, but not all, of an individual’s fingerprints on record,” he said.
The team behind the study says it is confident in the results and has open-sourced the AI code for others to check, a decision praised by both Champod and Cole. But Gu said the significance of the study goes beyond fingerprints.
“It’s not just about forensics, it’s about AI. Humans have been looking at fingerprints for as long as we’ve existed, but no one noticed the similarity until we started using our AI. Don’t analyze. It just speaks to the power of AI to automatically recognize and extract relevant features,” he said.
“I think this study is just the first domino in a larger sequence of things. We’ll see people using AI to discover things that are hidden right in front of our eyes, in plain sight like our fingers. were.”