International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 05 | May 2025
p-ISSN: 2395-0072
www.irjet.net
AUTHENTINOTE: A Counterfeit Currency Detection System Nikita Gauda1, Ruchi Bhati2, Savi Band3, Kumud Wasnik4 1Student,Dept of Computer Science and technology,Usha mittal institute of technology,Mumbai, Maharashtra 2Student,Dept of Computer Science and technology,Usha mittal institute of technology,Mumbai, Maharashtra 3Student,Dept of Computer Science and technology,Usha mittal institute of technology,Mumbai, Maharashtra 4HOD,Dept of Computer Science and technology,Usha mittal institute of technology,Mumbai, Maharashtra
-----------------------------------------------------------------------***--------------------------------------------------------------------ABSTRACT-Counterfeit notes are a growing issue that damages economic stability and destabilizes financial institutions around the world. Conventional methods of detecting counterfeit banknotes, being mostly visual, are not only time-consuming but also susceptible to human error. For the ordinary individual in everyday life, there is typically no means of checking a currency note's authenticity, making them more susceptible to fraud and financial loss. This problem requires a solution that is feasible and economically viable to allow all, professionals or not, to easily identify the counterfeit notes. To solve this urgent problem, we present an advanced Counterfeit Currency Detection System that integrates YOLOv5's ability to detect the prominent security features on banknotes and CNN (mobileNet V2) in identifying whether they are genuine or counterfeit. Our system is tailor-made for Indian rupee values of 500, 200, and 100 with the purpose of creating a standardized, real-time platform for examining the security features that characterize authentic notes from the counterfeits. Through the use of advanced deep learning methods, this system is sure to be precise and effective and thus an effective tool for identifying counterfeits. To be user-friendly and accessible, the system has a facility of audio feedback for blind individuals who can use its services. Given its emphasis on reliability, accessibility, and efficiency, this solution empowers individuals and communities by minimizing their potential to become victims of counterfeit money, thus leading to a secure financial system.
identifying and verifying different currencies.Banknotes must be recognizable from both sides, even if they are damaged or torn, by these identification systems. The world is replete with money, and all of them have different patterns, texture, and size, hence identification is not a simple task, particularly for foreign exchange workers. Physical handling of money is prone to numerous risks such as misidentification and physical wear and tear. Having this in mind, the demand for automated systems that can identify and authenticate money effectively with minimal human interference is increasing. One of the major challenges faced by global economies, including the Reserve Bank of India, is the increasing circulation of counterfeit currency (Reserve Bank of India, 2023) [21]. Every year, huge quantities of fake or torn currency are found, and this creates immense problems for the banks. Although electronic transactions are on the increase, physical currency is still indispensable for day-to-day transactions. The systems used to identify banknotes are not always efficient, especially in shopping malls and banks where transactions take place frequently. The necessity of an effective, automated system to identify fake money has become urgent. The development in imaging technologies and machine learning offers the scope to create systems that are effective in determining the genuineness of banknotes, decreasing fake currency circulation and maintaining the validity of financial transactions
Keywords–Computer Vision, Counterfeit Currency Detection, Indian Currency, MobileNetV2, Real-Time Classification, Security Features, YOLOv5,Audio Output.
The core purpose of the current research work is to engineer and develop an efficient currency detection and verification system that can determine authentic and forged banknotes. The system would look to dig out prominent banknote security characteristics such as numerals, security thread, Guarantee clause, Ashoka Emblem and RBI logo and then try labeling them as fake or real in the most apt way possible. Using a deep learning object detection model, the system will then be able to recognize such features even on torn or creased banknotes.
1.INTRODUCTION Identification of money has the most significant use of computer vision, that is, identification and verification of a country's bank notes. Counterfeit currency is defined as “imitation banknotes produced without legal sanction, designed to replicate genuine currency”(Interpol, 2023) refer [17]. Because of globalization and increased crossborder transactions, currency exchange across countries has become rampant, hence calling for safe means of
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 276