A new study suggests so Fingerprints Not really unique, which could have major implications for the world Forensic science.
The study, which was published in the journal Science Advances, used a type of… Artificial Intelligence (AI) Facial recognition is typically used to look for patterns and identifiers in fingerprints that humans might ignore.
The artificial intelligence was able to determine whether Fingerprints From different fingers of the same person they came from the same person with an accuracy of 75 to 90%.
According to researchers at Columbia University in the United States, the model was able to look beyond finger features known as “details”, which investigators have relied on for centuries.
The artificial intelligence algorithm was trained on a database of 60,000 fingerprints.
A new way to get to know
Instead, the AI analyzed the angles and curvatures of the swirls in the center of the fingerprints. In theory, these ideas could be used to link fingerprints of different types found at multiple crime scenes to the same person.
However, the researchers' claims have been met with skepticism by some. Forensic scientists.
Simon Cole, a professor of criminology, law and society at the University of California, told CNN that the findings are not necessarily new and are far from refuting current forensic techniques.
“The unproven, but intuitively true, statement that no two fingerprints are exactly alike cannot be refuted by the discovery of fingerprint similarity,” he said.
According to Cole, it has always been known that the fingerprints of different people, as well as the same person, are similar.
However, the researchers say their findings are “irrefutable.” They believe that artificial intelligence has the potential to revolutionize forensic science and other areas of analysis.
“We'll see people using AI to discover things that were literally hiding in plain sight, right in front of our eyes, like our fingers,” said Gabe Gu, lead author of the study.
Source: Byte Editorial
