Researchers from the Department of Computer Science at the University of Warwick in the U.K. have developed a method for identifying individual bank notes based on their unique “fingerprints.”

Due to a rise in counterfeiting following the 2016 introduction of plastic polymer banknotes in the U.K., the University of Warwick team developed a method dubbed Polymer Substrate Fingerprinting to identify the unique coating patterns that appear on individual bank notes.

According to the researchers, this technique identifies imperfections in the bank notes incurred by their physical manufacturing process, wherein the opacity coating is applied unevenly, thereby resulting in a random and unique dispersion of impurities in the ink. When backlit by a light source, these random translucent patterns on the bank note are revealed.

A demonstration of the feature area from polymer banknote (top). Three images at the bottom show the random variation of the patterns for three different banknotes. Source: University of WarwickA demonstration of the feature area from polymer banknote (top). Three images at the bottom show the random variation of the patterns for three different banknotes. Source: University of Warwick

To capture those patterns, the team used a commodity negative film scanner, processing them into a 2048 bit feature vector to accurately identify each individual bank note. The researchers analyzed the images, focusing exclusively on a small region of the bank note where the random translucent patterns incurred by the coating layer are revealed without obstruction via security printing.

With 500 billion bank notes currently in circulation throughout the world, the researchers believe that counterfeit bank notes are a significant threat to the economy and security. However, the University of Warwick team believes that counterfeiting attempts could be reduced by Polymer Substrate Fingerprinting because the unique fingerprints of the bank notes cannot be recreated.

The study — Anti-Counterfeiting for Polymer Banknotes Based on Polymer Substrate Fingerprinting — appears in the journal IEEE Transactions on Information Forensics and Security.

To contact the author of this article, email mdonlon@globalspec.com