Fingerprint minutiae (left, traditional method) versus angles at the center of the fingerprint (right, new AI method). Credit: Columbia Engineering
A team of engineers from Columbia University and the University at Buffalo, SUNY say they have built and trained an AI that has proven not every fingerprint is unique.
It’s a well-accepted fact in the forensics community that fingerprints of different fingers of the same person—or intra-person fingerprints—are unique and therefore unmatchable. Now, a new study shows an AI-based system has learned to correlate a person’s unique fingerprints with a high degree of accuracy. According to the researchers, it does this by analyzing the curvature of the swirls at the center of the fingerprint rather than the minutiae, or endpoints in fingerprint ridges.
The paper has been accepted for publication in Science Advances, but not before being rejected by the journal previously, as well as a separate rejection from a forensic journal.
Findings and publication
Columbia Engineering undergraduate senior Gabe Guo, who had no prior knowledge of forensics, found a public U.S. government database of approximately 60,000 fingerprints. He fed them, in pairs, into an artificial intelligence-based system known as a deep contrastive network. Sometimes the pairs belonged to the same person (but different fingers), and sometimes they belonged to different people.
The researchers say the AI system, designed by modifying a state-of-the-art framework, got better over time at telling when fingerprints belonged to the same person and when they didn’t. The accuracy for a single pair reached 77%. When multiple pairs were presented, the accuracy was even higher.
Once the team verified their results, they sent the findings to a forensics journal, but received a rejection a few months later. The expert reviewer and editor said: “It is well known that every fingerprint is unique,” and therefore it would not be possible to detect similarities even if the fingerprints came from the same person.
The rejection did not sway Guo and his team, however. They fed their AI system even more data to help it improve. This time, the researchers submitted their paper and findings to a more general journal—Science Advances.
The paper was rejected again.
However, Hod Lipson, professor of innovation at Columbia and co-director of the Makerspace Facility, appealed the decision. After more back and forth, the paper was ultimately accepted for publication.
“I don’t normally argue editorial decisions, but this finding was too important to ignore,” said Lipson. “If this information tips the balance, then I imagine that cold cases could be revived, and even that innocent people could be acquitted.”
Fingerprint markers
One of the main questions in the approval process centered on how this AI system debunked decades of forensic analysis. After careful visualizations of the AI system’s decision process, the team concluded that the AI was using a new kind of forensic marker.
Minutiae-based fingerprint analysis is the most widely used and accurate technique available today.
Minutiae refers to the location and direction of the fingerprint ridge endings and bifurcations (splits) along a ridge path. The types of information that can be collected from a fingerprint’s friction ridge impression include the flow of the friction ridges, the presence or absence of features along the individual friction ridge paths and their sequence, and the intricate detail of a single ridge.1
According to the Columbia research team, the AI system instead used data related to the angles and curvatures of the swirls and loops in the center of the fingerprint to reach its conclusions.
The researchers do note that more careful validation needs to be done using datasets with broader coverage if this technique is to be used in practice.
“But this research is an example of how even a fairly simple AI, given a plain dataset that the research community has had lying around for years, can provide insights that have eluded experts for decades,” said Lipson. “Even more exciting is the fact that an undergraduate student, with no background in forensics whatsoever, can use AI to successfully challenge a widely held belief of an entire field. We are about to experience an explosion of AI-led scientific discovery by non-experts, and the expert community, including academia, needs to get ready.”
Sources
1. https://ucr.fbi.gov/fingerprints_biometrics/biometric-center-of-excellence/files/fingerprint-recognition.pdf