International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 8 Issue 5, Sep-Oct 2024 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Fake Currency Detection

Aniket Bhoyar1, Aditya Sormare2, Chirag Tabhane3, Karan Tembhare4, Prof. Shreya Bhanse5
1,2,3,4
School of Science, G H Raisoni University, Amravati, Maharashtra, India
5
Assistant Professor, G H Raisoni University, Amravati, Maharashtra, India

ABSTRACT How to cite this paper: Aniket Bhoyar |

The creation and circulation of counterfeit notes are on the rise right Aditya Sormare | Chirag Tabhane |
now, as a result of advances in color printing technology. This is a Karan Tembhare | Prof. Shreya Bhanse
serious issue that affects practically all of the nations. The economy "Fake Currency Detection" Published in
International
is impacted. Such fake money fuels evil intentions, typically
Journal of Trend in
involving terrorist actions. According to the research, this has had a Scientific Research
highly negative effect on developing nations like India. This research and Development
suggests a method for viewing the fake currency through its image. (ijtsrd), ISSN:
Pre-processing should be used after choosing an image. Then a 2456-6470,
conversion from RGB to GRAYSCALE is done. Banks and other Volume-8 | Issue-5, IJTSRD69360
trading places have equipment available to verify financial validity. October 2024,
Nevertheless, the normal individual does not have access to such pp.274-279, URL:
tools, which is why fake money detecting software that is usable by www.ijtsrd.com/papers/ijtsrd69360.pdf
regular people is required. We get at most 81% of accuracy
considering 50 items of Indian Currency notes of 500-rupee. This Copyright © 2024 by author (s) and
International Journal of Trend in
project provides a thorough explanation of a fake note detector that Scientific Research and Development
can be used by the average person. The suggested system employs Journal. This is an
image processing to identify genuine currency from counterfeit Open Access article
money. The Python programming language has been used to create distributed under the
the software in its entirety. terms of the Creative Commons
Attribution License (CC BY 4.0)
KEYWORDS: Fake Currency, Image Processing, Grayscale (http://creativecommons.org/licenses/by/4.0)
Conversion, Segmentation, pre-processing
#Technologies
1)PYTHON-Python is an interpreter, object-oriented, high level,
dynamically semantic programming language. 2)Python Libraries:
OpenCV, NumPY, VS Code.

I. INTRODUCTION
Money can serve as the driving force behind any The issue of counterfeit money is one that is
economic activity associated with manufacturing, discussed and debated throughout the world. Banks
circulation, consumption, etc. Capital information can lost Rs. 16,789 crores in the most recent fiscal year
be used to save money and make investments. Money due to frauds. The Reserve Bank reported that "the
is essential for everything in today's dynamic culture. amount that has been lost on account of frauds in the
There are also other factors that are shrinking the year 2016–17 was Rs. 16,789 crores," which was in
economy as it advances. One of those things is the accordance with the fraud monitoring report made by
creation and use of counterfeit currency. Due to the various banks and financial institutions. According to
widespread use of counterfeit currency in the the RBI's (Reserve Bank of India) annual report for
economy, the typical person is the group most 2021–22, there was an increase in the number of
negatively impacted by this activity. Everyone is counterfeit notes found in the denominations of Rs.
afraid of accepting banknotes in the denominations of 10, Rs. 20, Rs. 200, Rs. 500 (new design), and Rs.
Rs. 500 and Rs. 1,000 because the bulk of them are 2,000, respectively, of 16.4%, 16.5%, 11.7%,
nearly hard to distinguish from genuine banknotes, 101.9%, and 54.6%.[51].
from gas stations to the neighborhood vegetable
Inflation is the typical impact of counterfeiting on the
seller.
economy. The only tool now available to the average
person to identify fake money is the Fake Note

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 274
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Detection Machine. The majority of the time, this conclusion part that includes result and discussion
machine is only found in banks, which are not always part, which represents what we have got from our
accessible to the regular person. In order to prove the project with the advantages of our project and also
viability of suggested solutions to a particular includes limitations of our project and the scope at
problem, a lot of experimental work is required in the which we can improve our project in the future.
field of digital image processing.
Data Description-The banknote-authentication dataset
It includes operations whose inputs and outputs are is used to distinguish between genuine and counterfeit
images and operations that extract properties from banknotes. Images of real and fake banknote-like
photos, including the identification of specific specimens were used to extract data from the photos.
objects. The watermark on fake currency is created These photos were processed and number of lines on
using opaque ink, white solution, and stamping with a a thin strip are measured. A compressed version of
dye that has a picture of Mahatma Gandhi engraved the dataset from Kaggle was used in this experiment.
on it. Visitors are the most susceptible to phony There are 100 samples total. The model has been
currency because they lack the knowledge necessary trained using 500-rupee notes of cash from India. To
to distinguish between fake and genuine currency determine the dataset's input/output behavior for the
notes. system, an experiment was run. The sample dataset
These people will benefit from automatic currency utilized in the experiment is named and provided
identification using image processing techniques. below:
Also, it can be helpful in other workplaces. The Dataset Source Item Type
devised system to verify the 500- rupee Indian Indian currency Image
kaggle 50
currency notes. It will organize the predetermined Note 500 rupee dataset
arrangement of information and pre-process the Requirement Analysis-
digital images before differentiating in monetary The implementation requirement details are given in
forms. The approach for detecting Indian currencies this section. Requirement Analysis method is
suggested in this article is practical and affordable. intended in such a way that it takes fewer resources to
The user can determine whether the cash note is figure out work correctly. The minimum needs that
authentic or phony at the conclusion of the process. we’d like to take care of: The system would require a
Objective- minimum of 4 GB (Gigabyte) of RAM (Random
1. To examine the various security components of Access memory) to run all the options sleek and
Indian currency notes. unforeseen. It wants a minimum of 2 GHz
2. Using a scanner or camera to gather paper money. (Gigahertz) processor to run the system smoothly.
3. To extract characteristics from the captured image The system can be operated by common people as
by cropping and segmenting it. well as commercial people.
4. Creating a feature localization algorithm. Hardware Specification:
5. Designing an extraction and recognition of Processor 2 GHz Intel
features. Storage 512GB
6. To determine the right money denomination. Ram 4GB
II. RELATED WORK- Software Specification-
This project paper is based on Fake Currency Operating System Windows 7,8,10
Detection. At the first chapter it includes the Programming language Python
introduction and objective part which mainly describe IDE (Integrated Development
about our project topic, why this project is important VS code
Environment)
and which methods are used in our project to make it
successful. The second chapter includes literature Python-
review which mainly represents the summary of the Python is an interpreter, object-oriented, high-level,
different project paper related to our project topic, dynamically semantic programming language. It is
including of what they had achieved and by which particularly desirable for Rapid Application
process they had done their project etc. The third Development as well as for usage as a scripting or
chapter includes the methodology part which mainly glue language to tie existing components together due
represents the information about the used dataset, to its high-level built-in data structures, dynamic
software and hardware tools and tables the typing, and dynamic binding. Python's
description of the process with the flow-chart straightforward syntax prioritizes readability and
diagram, features of the currency by following we did makes it simple to learn, which lowers the cost of
our project. The fourth chapter includes the program maintenance. Python's support for modules

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 275
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
and packages promotes the modularity and reuse of NumPy- Many mathematical operations can be
code in programs. On all popular platforms, the carried out on arrays with NumPy. It provides a vast
Python interpreter and the comprehensive standard library of high-level mathematical functions that work
library are freely distributable and available in source on these arrays and matrices, as well as strong data
or binary form. structures that ensure efficient calculations with
Python Libraries- arrays and matrices. To install NumPy run the
OpenCV- command - pip install numpy.
OpenCV is a sizable open-source library for image VS Code-Debugging, task execution, and version
processing, machine learning, and computer vision. It control are supported by the simplified code editor
now plays a significant part in real-time operation, Visual Studio Code. It tries to give developers only
which is crucial in modern systems. With it, one may the tools they require for a short cycle of code-build-
analyze pictures and movies to find faces, objects, debugging and leaves more sophisticated processes to
and even human handwriting. IDEs with more features, like Visual Studio IDE.
To install OpenCV run the command - pip install Accuracy: The percentage of accurately classified
opencv-python. Python is able to handle the OpenCV data samples over all the data is known as accuracy.
array structure for analysis when it is integrated with Accuracy can be calculated by the following
different libraries, such as NumPy. We use vector equation.
space and apply mathematical operations to these
Accuracy = (TP+TN)/(TP+FP+TN+FN)
features to identify visual patterns and their various
features.
Features of Currency-
All features of Indian currency 500 showing in fig

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 276
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Portrait of Mahatma Gandhi at the Center- Identification Mark:
The intaglio printing of portrait of Mahatma Gandhi Just over the Ashoka's pillar symbol, there is an
at the center of the currency. identification mark.

Fig: Identification mark

Guarantee Clause: Located to the right of Mahatma
Gandhi's image, the guarantee clause is signed by the
Fig: Portrait of Mahatma Gandhi governor and includes a promise clause that is printed
in intaglio.
Security Thread-
When held up to the light, the security thread, which
has "RBI" and "Bharat" inscribed on it continually,
can be seen at the left side of the watermark. The
photo of the Mahatma has a security thread on one
side.

Fig: Guarantee Clause

Currency Numeral with the Rupees Symbol:
Fig-Security Thread Fluorescent ink will be used for printing. When
See through Register: The denomination numeral is viewed from different perspectives, the numerals
displayed in the see-through register. Both sides of change.
this register are printed. One side of the two sides is
hollow, and the other side is filled with material. The
micro lettering has been written horizontally along
this register. The note has a latent image on the left Fig: Currency Numeral with the Rupees Symbol
side. Moreover, this register is shown above the latent
image. When viewed in contrast to the light, this Bleed Lines:
register appears as a single design. The oblique lines that protrude from the sides of
banknotes are known as bleed lines.

Fig: See through Register

Ashoka Pillar: Fig: Bleed lines
On the right side of the coin there is a picture of the Latent Image of Denomination Numeral: The right
Ashoka pillar. side of Mahatma Gandhi’s portrait is bordered by a
vertical band on the opposite side of the
denomination. A latent image of the corresponding
denominational value is present in it. Its
denominational value is represented by a numerical
value. The latent picture can be seen when the coin is
held horizontally, and it should also be held at eye
level. While using counterfeit money, it is not
Fig: Ashoka pillar noticeable.

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 277
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Micro Lettering: Between the vertical band and the
image of Mahatma Gandhi, micro lettering is visible.
The term "RBI" and the denominational value are
written in tiny letters. The micro letters on counterfeit
money are incorrectly printed.
Fig: Latent Image of Denomination Numeral

Fig: Micro Lettering

Government of India: The words "Government of The major image acquisition components have been
India" are printed at the top of the one rupee note, mentioned below:
directly over the Devanagari-scripted number one. 1. Trigger
The smallest currency note now in use in India is 1 2. Camera
rupee, and it is the only one that was produced by the 3. Optics
Government of India rather than the Reserve Bank of 4. Illumination
India like the others. Because of this, it is the only
RGB to Grayscale: Taking the average of the red,
one with the Finance Secretary's signature rather than
green, and blue pixel values for each pixel to obtain
the RBI Governor's.
the grayscale value is a straightforward technique to
convert a color image's 3D array to a grayscale
image's 2D array.
This creates an approximate gray color by combining
the lightness or brightness contributions from each
color band.
Fig: Government of India
The Average method takes the average value of R, G,
Required Algorithm: Image acquisition: The act of
and B as the grayscale value.
obtaining an image from sources is known as image
acquisition. Hardware systems like cameras, Grayscale = (R + G + B ) / 3
encoders, sensors, etc. can be used to do this. It is The weighted method, also called the luminosity
without a doubt the most important phase in the MV method, weighs red, green, and blue according to
(Machine Version) workflow because a bad image their wavelengths.
would make the workflow ineffective as a whole. As
machine vision systems don't study the acquired The improved formula is as follows: Grayscale =
digital image of the object and not the object itself, 0.299R + 0.587G + 0.114B
acquiring an image with the proper clarity and Image Segmentation: Image segmentation is a
contrast is crucial. A set of photo-sensitive sensors technique for breaking up a digital image into smaller
turn an object's incoming light wave into an electrical groupings called image segments, which reduces the
signal during the image acquisition step. These little complexity of the image and makes each segment
components provide the function of accurately more easily processed or analyzed. Technically,
describing the object to your machine vision segmentation is the process of giving labels to pixels
algorithms. It's a frequent fallacy that with an MV in an image in order to distinguish between objects,
system, choosing the correct colors is crucial. persons, or other significant aspects. Object detection
However it's not always the case. Colors frequently is a frequent use of image segmentation. It is usual
increase noise and make detection more challenging. practice to first apply an image segmentation method
The main objective of an image acquisition system is to discover things of interest in the image before
to increase contrast for the important features. The processing the complete image. The object detector
ideal image is one in which the camera can clearly see can then work with a bounding box that the
the object of interest. segmentation algorithm has previously established.

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 278
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
By stopping the detector from processing the entire  This project is only able to detect the currencies
image, accuracy is increased and inference time is whether it is fake or not with denomination 2000
decreased. A crucial component of computer vision of Indian rupees.
technologies and algorithms is image segmentation.
Future Scope-
Feature Measurement: The process of "feature  This project cannot be able to detect the
detection" involves computing abstractions of image currencies of other countries except India. So in
data and locally determining whether or not each the future we can make this project possible to
image point contains an image feature of a specific detect the currencies of other countries also.
type. A fundamental aspect of image processing is
 This project is only able to detect the currencies
feature detection. This means that it is typically done
whether it is fake or not with denomination 2000
as the initial operation on an image and checks each
of Indian rupees. So in the future we can make it
pixel to see if a feature is present there. If this is a
possible that it will detect the currencies with all
component of a bigger algorithm, the algorithm will
denomination.
usually just look at the image where the features are.
The term "feature description" refers to a technique  In this project, we worked using a few features of
for describing the local attributes of an image at the currencies. So in the future we can be able to
identified key points in an image. These algorithms work with all features of currencies to increase
take advantage of key points discovered in the image the accuracy of the project.
data to extract interesting information. The Reference-
information produced by these feature description [1] Colaco, Rencita Maria, Rieona Fernandes, and
techniques is frequently organized by encoding it as S. Sowmya. "Efficient Image Processing
the constituent parts of a single vector, or feature Technique for Authentication of Indian Paper
vector. A feature space is the collection of all feasible Currency." 2021 International Conference on
feature vectors. Computer Communication and Informatics
Finding Correlation: (ICCCI). IEEE, 2021.
For finding Correlation of two images we have to [2] Vadnere, G. V., Khilari, S. A., Sonawane, H.
follow this steps: P., sharad Sanap, P., Koneri, A. M., & Shinde,
1. Load two images and extract their pixel-by-pixel M. J. Indian Currency Identification Using
information Image Processing. Journal homepage: www.
2. Normalize and down sample the pixel information ijrpr. com ISSN, 2582, 7421.
3. Calculate cross-correlation using the processed [3] Jamkhandikar, Dayanand, Sahana Kaveri, and
pixel information V. Poojashree Sudharani. "Indian Currency
Recognition System."(2021).
4. Generate visual summaries of cross-correlation,
highlighting areas of maximum image overlap. [4] Sangogi, Mrs Jyoti, Ms Prachi Patil, and Ms
Anuradha Jadhav. "FAKE CURRENCY
Conclusion-
DETECTION USING IMAGE
Advantages-
PROCESSING."
 The application will be very helpful in identifying
counterfeit money. [5] Shiby, Ashik, Fevitha Francis, Philip Jose, and
Rintu Augustine. "FAKE CURRENCY
 The software is simple to use and accessible.
DETECTION USING IMAGE
 It will lessen the user's effort and save time. PROCESSING."
 It provides the user methods that are more [6] Patil, Deepika P., Girija Varma, Shweta
affordable, accurate, and give accurate Poojary, Shraddha Sawant, and Aditya Sharma.
recognition of money notes. "Counterfeit Currency Detection based on AI."
Restriction- [7] Kudalkar et al., “Fake Currency Detection
 This project cannot be able to detect the Using Image Processing”. AIJR Preprints, 388,
currencies whether it is fake or not, of other Version 1, 2022.
countries except India.

@ IJTSRD | Unique Paper ID – IJTSRD69360 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 279