Canadian researchers develop AI technology to identify counterfeit money

TapTechNews June 2nd news, according to a report by TheCanadianPress, researchers at Concordia University in Montreal, Canada, and their team have developed an innovative technology that can accurately identify counterfeit money with the help of artificial intelligence technology.

Currently, the technology of counterfeiters to make counterfeit money is getting more and more sophisticated, and sometimes even experienced experts are at a loss. Now the addition of AI is expected to change the rules of the game.

It is reported that Maryam Sharifi Rad, a postdoctoral fellow at the university's Pattern Recognition and AI Research Center, and her team developed this technology. The researcher said, This is a problem that very directly affects the economy and public finances. I am passionate about pushing the boundaries of technology to enhance the security and protection of the monetary system.

The research team digitized a large number of real and counterfeit banknotes at high resolution. With the machine learning and image mining technology developed by the researchers, the researchers analyzed the two-dimensional images. By searching for irregularities and flaws on each sample, the system can accurately distinguish between real and counterfeit banknotes. (Note: The irregularities and flaws mentioned above belong to the kind that are often impossible for the naked eye to detect.)

However, for this team, obtaining counterfeit banknotes from multiple countries is a considerable challenge, as most of these counterfeit banknotes are in the hands of police around the world.

Maryam Sharifi Rad said that it is fascinating that AI can detect extremely tiny parts. Thanks to advanced image processing methods, our accuracy rate has reached 99%.

Her partner, Saeed Khazaee, said that the Royal Canadian Mint is very interested in the team's work because there are a large number of counterfeit banknotes in the local market. The authorities especially want to know what the 'exact' specific characteristics of the counterfeit currency are in order to make the real currency more difficult to replicate.