Researchers, led by New York University, have developed a new method based on machine-learning algorithms to distinguish between genuine and counterfeit versions of the same product. Such technology will be of value to businesses producing goods, to retail stores and to trading standards and customs officials.
The counterfeit trafficking of goods could represent seven percent of the world’s trade, equating to half a billion dollars according to OCED. China appears to be the single largest producing market of fake goods.
Machine learning solution
The research has been led by Professor Lakshminarayanan Subramanian and it fits with the research field of ‘knowledge discovery in databases’ (KDD). Outlining what the technology does, Subramanian explained in an interview: “The underlying principle of our system stems from the idea that microscopic characteristics in a genuine product or a class of products — corresponding to the same larger product line — exhibit inherent similarities that can be used to distinguish these products from their corresponding counterfeit versions.”
Three million scanned goods
The researcher has designed ‘The Entrupy Method’. This gives a non-intrusive solution to allow businesses and officials to readily distinguish authentic versions of the product produced by the original manufacturer and fake versions of the product produced by counterfeiters. This is achieved via a dataset made up of three million images, drawn across various objects and materials.
The device can differentiate between different types of fabrics and leather; as well as a scanning and interpreting medicines, electronic goods and toys. The accuracy rating is 98 percent.
The developments in machine learning were recently presented to the annual KDD Conference on Knowledge Discovery and Data Mining, which this year took place in Halifax, Nova Scotia.