Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

Detection and Counting of Fake Currency &

Genuine Currency Using Image Processing
Nimish Bodkhe1; Deep Bodkhe2; Ayush Tayade3; Sarvesh Makode4; Dr. Nikkoo Khalsa5
1,2,3,4,5
Department of Electronics & Telecommunications Engineering,
Prof. Ram Meghe Institute of Technology & Research, Badnera, Amravati

Abstract:- In today's digital age, the proliferation of approaches to counterfeit currency detection leveraging image
counterfeit currency poses a significant challenge to processing technologies. In recent years, the field of image
financial institutions and businesses worldwide. The processing has seen remarkable progress, driven by
detection and prevention of counterfeit currency advancements in machine learning and computer vision. These
transactions require advanced technological solutions to developments have opened new avenues for the rapid and
safeguard the integrity of financial systems. In our research, accurate identification of counterfeit banknotes, offering
we present a novel approach to tackle this problem by potential solutions to combat financial fraud.
leveraging image processing techniques. The proposed
system utilizes state-of-the-art image processing algorithms
to distinguish between fake and genuine currency notes
accurately. By analyzing various features such as texture,
color, and watermark patterns, our system can effectively
identify counterfeit currency notes with high precision.
Additionally, the system is capable of counting and
validating genuine currency notes, providing a
comprehensive solution for currency authentication and
management. The integration of image processing
technology into currency verification systems offers
numerous benefits, including enhanced accuracy, speed,
Fig 1: 500 Rs. Note with 12 identity marks
and reliability. Our research aims to contribute to the
development of robust and efficient solutions for combating
Traditionally, counterfeit currency detection has relied on
counterfeit currency fraud, thereby safeguarding the
manual inspection or specialized machines, which are often
integrity of financial transactions and promoting trust in
time-consuming and prone to human error. Moreover,
monetary systems.
counterfeiters continuously adapt their techniques to mimic the
Keywords:- Image Processing, Fake Currency Detection, intricate features of genuine currency, posing a persistent
challenge to conventional detection methods.
Genuine Currency Detection, Counterfeit Currency Detection ,
Financial Security.
In response to these challenges, the paper proposes a novel
approach that harnesses the capabilities of Google TensorFlow,
I. INTRODUCTION
Teachable Machine, Java, Embedded C language, Arduino, and
In today's interconnected world, the circulation of physical Android Studio software to achieve real-time detection and
counting of currency notes.
currency remains a cornerstone of economic transactions,
despite the increasing prevalence of digital payment methods.
However, alongside the convenience of cash transactions By integrating these cutting-edge technologies, the
comes the persistent threat of counterfeit currency, a challenge research paper seeks to address the limitations of existing
that continues to evolve with technological advancements. The counterfeit currency detection methods and pave the way for
ability to accurately detect and distinguish between genuine and more efficient and reliable solutions.
fake banknotes is not only crucial for maintaining the integrity
of financial systems but also for preserving trust and confidence The proposed system offers a comprehensive framework
in currencies worldwide. for simultaneous image processing, enabling seamless
communication between an Android mobile device and a
The research paper titled "Detection and Counting of Fake currency note counting machine. Through a detailed
Currency & Genuine Currency Using Image Processing" methodology, the paper outlines the steps involved in
addresses this pressing issue by exploring innovative initializing the hardware, establishing Bluetooth connectivity,

IJISRT24APR2080 www.ijisrt.com 2106

Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

and implementing image processing algorithms for counterfeit banknotes were stored in a database for paper currency
currency detection. identification.[2]

Automatic authentication of paper money has been

targeted. Indian bank notes are taken as reference to show how
a system can be developed for discriminating fake notes from
genuine ones. Image processing and pattern recognition
techniques are used to design the overall approach. The ability
of the embedded security aspects is thoroughly analyzed for
detecting fake currencies.[3]

This proposed system describes an approach for

verification of Indian banknotes. The currency will be checked
out by using image processing techniques. The approach
consists of a number of elements including processing of image,
Fig 2: 100 Rs. Note detection of edge, image segmentation, drawing out
characteristic, comparing both images. The image processing
Furthermore, the paper highlights the practical approach is discussed with MATLAB to verify the parameters
implications of the proposed system, emphasizing its potential of note. Image processing involves changing the nature of an
to streamline currency verification processes, enhance security image in order to improve its visual information for human
measures, and mitigate financial losses resulting from interpretation. The image processing software is a collection of
counterfeit currency circulation. By providing a deeper functions that extends the capability of the MATLAB numeric
understanding of the underlying principles of image processing computing environment. The result will be whether note is real
and machine learning, the research paper aims to contribute to or fake.[4]
the advancement of counterfeit currency detection techniques
and foster greater resilience against financial fraud in the digital This paper presents the various fake currency detection
age. techniques. Fake currency is imitation currency produced
without the legal sanction of the state or government. Producing
In essence, the research paper serves as a testament to the or using fake currency is a form of fraud or forgery. We have
transformative power of image processing technologies in reviewed different fake currency detection systems. The
safeguarding the integrity of global financial systems and systems are developed using different methods and algorithms.
upholding trust in currencies worldwide. Through continuous The benefits of this study for the reader are that this study will
innovation and collaboration, it is envisaged that such provide information about the different methods and algorithms
advancements will play a pivotal role in shaping the future of used for fake currency detection system. They can compare the
currency authentication and ensuring the resilience of detection systems. Detection ability depends on the currency
economies in an increasingly digital and interconnected world. note characteristics of particular country and extraction of
features.[5]
II. LITERATURE SURVEY
This research article gives information on how to identify
In this paper, author suggest a highly useful and effective the security features of the Indian currency. It has mentioned all
technique for detecting fake currency. Counting the quantity of front side as well as reverse side features of 2000 rupees notes
disruptions there in thread line is used to detect fake cash notes. and 500 rupees notes. The author mentioned around twenty
The number of interruptions can be used to find out if a note is features like note’s size, color, security threads with an
authentic or counterfeit. If there are no interruptions, it is inscription, number on the top left and bottom right side,
genuine; otherwise, it is a fake. emblem of Ashoka pillar on the right side, etc.[6]

They also compute the entropy of money notes to detect In this project we have made fake currency note detection
fraudulent currency notes more effectively. The bogus money technique using MATLAB and feature extraction with HSV
note is detected using MATLAB software.(Pallavi S1, Pooja color space and other applications of image processing. In the
N2, Yashaswini HR3, Varsha N4). [1] project setup, note is placed in front of camera to check whether
it is fake or genuine. The camera pictures of notes are analyzed
In described an approach for recognizing paper currency by MATLAB program installed on computer. The project is
that depended on Hidden Marko Model (HMM). HMM was meant to check Indian currency notes of 100, 500 and 1000
utilized to model the banknote texture as a random operation. rupees. If the note is genuine, the respective message is
The pre-processing stage was necessary to solve the problem of appeared on the screen and vice-versa. [7]
identifying dirty banknotes. The attribute vectors for various

IJISRT24APR2080 www.ijisrt.com 2107

Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

III. METHODOLOGY Finally, the app displays the total value of genuine notes
and provides visual feedback through an LED indicator on the
This project implements TensorFlow, a machine learning ESP32 . The process concludes by sending a stop signal to the
framework, on an Android device to detect fake currency. The counting machine, halting its operation. In essence, this
system involves both hardware and software components methodology utilizes TensorFlow on the Android device for
working together. image classification, while a separate system handles physical
dispensing and counting. Communication and control signals
First, all hardware components, including the ESP32 ensure a coordinated detection process with real-time feedback.
microcontroller, relay module, motor, and LED indicator, are
connected and powered. The Android application, equipped
with Bluetooth and TensorFlow integration for image The system proposed here work here on the image of
processing and classification, is developed. The ESP32 waits currency note under ultraviolet light acquired by a digital
for a Bluetooth connection, while the Android app initiates camera. The algorithm which is applied here is as follows:
Bluetooth and searches for the device. Once connected, the user
interface on the app allows users to initiate the detection  Acquisition of image of currency note under ultraviolet light
process. by simple digital camera or scanner.
 Image acquired is RGB image and now is converted to
Upon confirmation, the app captures images of currency grayscale image.
notes using the phone's camera. These images are preprocessed  Edge detection of whole gray scale image.
on the device and then fed into the TensorFlow model for  Now characteristics features of the paper currency will be
classification as genuine or fake. The app communicates with a cropped and segmented.
separate currency note counting machine via Bluetooth. When  After segmentation, characteristics of currency note are
the detection process starts, the counting machine dispenses extracted.
notes and calculates the total value of genuine ones. The  Intensity of each feature is calculated.
Android app continuously receives feedback from TensorFlow
 If the condition is satisfied, then the currency note is said as
about the classification of each dispensed note. If a fake note is
original otherwise fak.
detected, the user is alerted, and the app signals the relay
module to stop the motor, preventing further dispensing. For
genuine notes, the total value is updated on the app screen.

A. Block Diagram

Fig 3: Block Diagram of Detection & counting of Fake currency & Genuine Currency Using Image Processing

IJISRT24APR2080 www.ijisrt.com 2108

Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

 Block Diagram Description  Software Functionality

 Hardware Components:  Image Capture: The system captures an image of the

currency using a camera (not shown explicitly in the block
 Receiver ESP32 BLE: This is the main processing unit of diagram).
the system. It is a microcontroller (MCU) with built-in Wi-  Image Preprocessing: The captured image data is
Fi and Bluetooth Low Energy (BLE) capabilities. It is likely preprocessed to prepare it for the machine learning model.
responsible for capturing the image of the currency, This might involve resizing the image, converting it to
preprocessing the image data, performing the machine grayscale, or applying filters to reduce noise and enhance
learning inference to classify the currency as fake or relevant features.
genuine, and transmitting the results wirelessly  Machine Learning Inference: A pre-trained machine
 Relay Module: This module is likely used to control an learning model is loaded onto the ESP32. This model
external device, such as a solenoid or motor, based on the likely takes the preprocessed image data as input and
classification results. For example, it could be used to outputs a classification result, indicating whether the
activate a mechanism to reject fake currency or sort genuine currency is fake or genuine
currency into a separate bin.  Output and Control: Based on the classification result, the
 Motor: This component might be used to drive a mechanism system might display the result on the LCD screen,
for sorting or rejecting currency based on the classification activate the relay module to control an external device, or
results. transmit the results wirelessly via BLE.
 LCD Display with I2C: An I2C (Inter-Integrated Circuit)
LCD display is likely used to show the user information
about the system's operation, such as the number of genuine
and fake currencies detected.
 5V 2 Amp Power Supply: This component provides the
necessary power to operate all the other parts of the system.

B. Connection Diagram

Fig 4: Connection Diagram of Detection & Counting of Fake Currency & Genuine Currency Using Image Processing

IJISRT24APR2080 www.ijisrt.com 2109

Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

 Receiver Module (ESP32 Dev Kit C V2): This  Relay Module: This module is likely used to control an
microcontroller is the main processing unit. It likely external device based on the classification results. For
captures images from a camera, preprocesses the image instance, it could activate a mechanism to reject fake
data, performs machine learning inference to classify the currency or sort genuine currency into a separate bin.
currency as fake or genuine, and transmits the results  Power Module (5V): This component supplies power to all
wirelessly (if Bluetooth Low Energy (BLE) capabilities are the other parts of the system.
used).  LCD Display with I2C: This display shows information
about the system's operation, such as the number of genuine
and fake currencies detected.

Fig 5: Real Image of the Detection & Counting of Fake Currency & Genuine Currency Using Image Processing

IJISRT24APR2080 www.ijisrt.com 2110

Volume 9, Issue 4, April – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24APR2080

C. Advantages REFERENCES

 Accessibility: Leverages a widely available smartphone [1]. FAKE CURRENCY DETECTION , Pallavi S*1, Pooja
(Android) for currency detection, potentially reducing N*2, Yashaswini HR*3, Varsha N*4 , June-2022
reliance on specialized hardware. [2]. Reserve Bank of India, High Level RBI Group Suggests
 Machine Learning Power: Integrates TensorFlow, a Steps to Check Menace of Fake Notes, Press release
powerful framework, for accurate real-time classification of 2009-2010, Dated 11 August (2009) , Sabeeta Badkar
genuine and fake notes. [3]. Image-Based Processing of Paper Currency Recognition
 User-Friendly Interface: Provides a user-friendly Android and Fake Identification: A Review Sabat Salih
app for initiating and monitoring the detection process. Muhamad , Twana Nasih Ahmed ( 2018 )
 Portability: The system's reliance on a phone and https://www.techniumscience.com/
potentially wireless communication makes it portable and [4]. Android-Based Verification System for Banknotes
convenient for on-the-go use. Ubaid Ur Rahman 1 , Allah Bux Sargano 2, ID and
 Real-Time Feedback: Delivers real-time feedback on the Usama Ijaz Bajwa 2 , Published: 24 November 2017
authenticity of each dispensed note, allowing for immediate [5]. Machine Assisted Authentication of Paper Currency: an
action if a fake is detected. Experiment on Indian Banknotes Ankush Roy , Biswajit
 Visual Cues: Utilizes an LED indicator for visual Halder, Utpal Garain, David S. Doermann,
confirmation of the detection Result. arXiv:1401.0689v1 [cs.CV] 2 Jan 2014
[6]. Fake Currency Detection Using Image Processing
 Potential Cost-Effectiveness: By leveraging existing
Ankur Saxena1 ,Pawan Kumar Singh2 , Ganesh Prasad
smartphone hardware, the overall system cost might be
Pal3 , Ravi Kumar Tewari (International Journal of
lower compared to dedicated currency detection machines.
Engineering & Technology) January 2018
D. Applications [7]. A Neural Network-Based Model for Paper Currency
Recognition and Verification Angelo Frosini, Marco
Gori, Member, IEEE, and Paolo Priami, 1996
 Banking and Financial Institutions
[8]. Fake currency detection using image processing Tushar
 ATM Machines Agasti, Gajanan Burand, Pratik Wade and P Chitra
 Vending Machines School of Electronics Engineering, VIT University ,
 Teller Operations 14th ICSET-2017
 Cash Registers [9]. M. Thakur and A. Kaur, “VARIOUS FAKE
 Ticket Sales CURRENCY DETECTION TECHNIQUES,”
 Online Sellers Published year :- 2014
 Airport Kiosks [10]. Aoba, M., Kikuchi, T., Takefuji, Y.: Euro banknote
 Transaction Verification recognition system using a three-layered perceptron and
 Educational Events RBF networks. IPSJ Trans. Math. Model. Appl. 44, 99-
 Central Banks --109 (2003)
 School & Colleges [11]. An automatic recognition of fake Indian paper currency
note using MATLAB Binod Prasad Yadav, C. S. Patil,
IV. CONCLUSION R. R. Karhe, P.H Patil ,Jully 2014
[12]. An automatic recognition of fake Indian paper currency
A project focused on the detection of fake currency and note using MATLAB Binod Prasad Yadav, C. S. Patil,
counting of currency is a valuable endeavor with significant R. R. Karhe, P.H Patil ,Jully 2
benefits, especially in the context of financial institutions, retail
businesses, and currency exchange offices. Such projects
typically involve a combination of hardware components,
software algorithms, and security features to ensure accurate
and reliable results. The key advantages include improved
accuracy, efficiency, counterfeit detection capabilities, and user
convenience. Currency counting and counterfeit detection
systems play a crucial role in streamlining financial processes,
reducing the risk of financial losses due to counterfeit currency,
and enhancing overall security.

IJISRT24APR2080 www.ijisrt.com 2111