Researchers at Concordia University have developed a new method to detect counterfeit coins more easily using advanced technology.
Despite being one of the oldest forms of money, coins are still vulnerable to counterfeiting, posing a threat to global currencies.
Recent incidents, like the breakup of a counterfeit ring in Spain by European police, highlight the ongoing issue.
The new method, described in the journal Expert Systems With Applications, uses image-mining techniques and machine learning to spot flaws in fake coins that might be missed by the naked eye.
The research was led by Professor Ching Suen from the Department of Computer Science and Software Engineering and involved postdoctoral fellow Maryam Sharifi Rad and research associate Saeed Khazaee.
The team’s approach relies on fuzzy association rules mining, a type of artificial intelligence that finds patterns that are not exact but still useful for detecting counterfeits.
Here’s how it works:
- Scanning the Coins: Coins suspected of being fake are scanned using high-tech equipment provided by law enforcement agencies.
- Identifying Regions of Interest: The scanned images are broken down into regions of interest, called “blobs.” These blobs are collections of localized, coherent regions that show visual similarities and compositions.
- Analyzing Blobs: The blobs are analyzed for their color, texture, shape, and size. This analysis helps researchers identify patterns and relationships among the blobs’ attributes.
- Applying Fuzzy Association Rules: Using these patterns, the fuzzy association rules mining technique extracts frequent patterns from the images. These patterns help researchers understand the images better and determine whether a coin is real or fake.
Professor Suen explains, “Using image technology, we scanned both genuine and counterfeit coins to look for anomalies, like letters or the face of the person on the coin.”
This method is not only about protecting the economy but also about advancing technology and improving security.
Maryam Sharifi Rad, the lead author of the paper, adds, “This framework pushes the boundaries of technology and helps improve security.”
The technique can identify tell-tale signs of forgery that are not visible to the naked eye, making it a powerful tool in the fight against counterfeiting.
The researchers believe their method can be used to detect a wide range of counterfeit items, not just coins. “This method can be applied to detect all kinds of fake goods,” says Suen. “It can also be used to spot fake labels on fruits, wines, liquor, and more.”
In summary, this new technology from Concordia University provides a sophisticated and effective way to detect counterfeit coins and other fake items, helping to safeguard economies and improve security worldwide.
