Forensic authentication of banknotes on mobile phones

Relatively cheap and high quality consumer scanners and printers have enabled the rise of the casual money counterfeiter. One who passes along home-made fake bills of low denomination in busy environments where the receiver is not likely to authenticate a bill. While this may be negligible on macro-economic scale, it does hurt consumers. In this paper we investigate several methods to identify counterfeit bills using an ordinary hand held mobile phone without any modification or special lighting. We demonstrate using a database of Swiss and Euro notes that variations of statistics along edges between a printing press, a laser and an inkjet are distinguishable with a mobile device. Furthermore, we show how random printing variations in the production of true banknotes can be used as a unique non-cloneable identifier for that particular bill. Introduction The forgery of banknotes is no longer limited to organizations or even countries that have the means to invest in professional printing equipment. Cheap high quality consumer scanners and printers have enabled the casual counterfeiter. Someone who copies a low denomination bills and passes them off in environments where people are not likely to check the authenticity, e.g, small shops during rush hour. For instance, counterfeiter Art Williams Jr. produced millions of dollars using commercially available paper, ink, printers and photo-editing software [25]. Similarly, Wesley Weber forged the CAD 100 bill using HP DeskJet printers and he is believed to have produced more than $6-million in fake $100 bills that made it into circulation [8]. Casual counterfeits are removed relatively quickly from circulation as they don’t stand up to close scrutiny. The main victim is the last consumer to be handed the bill or the local retailer. In 2002, over 95% of the detected fake notes of $20 and smaller denominations, which are those most commonly used by U.S. consumers, were casual counterfeits. The 5 remaining percent of high-quality counterfeits only accounted for less than $220’000 in total [17]. Obviously, there are numerous security features embedded in banknotes to deter counterfeiting, be it high or casual low grade. Central banks and manufactures typically disclose as subset to enable the public to authenticate bills. Disclosed security features for the Swiss Franc, Euro and US Dollar include Intalgio printing, watermarks, Guilloch patterns holograms etc. In practice however, very few members of the public are fully aware of all security features further illustrating why end consumers are the most likely victims of causal counterfeiting. It is therefore in this work, that we demonstrate a simple forensic tool that allows a consumer to verify the authenticity of a Swiss banknote using an ordinary mobile phone. The deployed method has three main stages. A geometrical alignment of the suspect Swiss banknote followed by two types of feature extraction to ascertain the characteristics of the manufacturing device. The first feature is build up on the fact that Swiss banknotes exhibit random fluctuations between printing stages. These fluctuations can be measured and tied in to the serial number of the bill. Secondly, we show how a relatively simple feature vector build from the edge and contour characteristics can be used to discover the used printing technique.