Fake Currency Recognition System Using Edge

Detection
P. Ashok babu P. Sridhar Rajeev Ratna Vallabhuni, MIEEE
Department of Electronics and Department of Electrical and Application Developer
Communication Engineering Electronics Engineering Bayview Asset Management, LLC
Institute of Aeronautical Engineering Institute of Aeronautical Engineering Florida, USA
Dundigal-500043, Hyderabad, India Dundigal-500043, Hyderabad, India [email protected]
[email protected] [email protected]

Abstract—In this paper, we propose a system for currency Our system introduced to solve the issues efficiently
recognition system and the detection of fake Indian currency through a more robust algorithm and works appropriately.
banknotes using image processing techniques. It is hard for Enabling things to be easily accessed and used by most of the
people to perceive monetary forms from various nations. Our population [6]-[11].
2022 Interdisciplinary Research in Technology and Management (IRTM) | 978-1-6654-7886-1/22/$31.00 ©2022 IEEE | DOI: 10.1109/IRTM54583.2022.9791547

point is to help individuals with taking care of this issue. In any

case, money acknowledgement frameworks that are in light of According to the CIA's survey, there are approximately 180
picture investigation are entirely not adequate. Our framework or more currencies in use around the world [12]-[16]. This
depends on picture handling and makes the procedure holds true for each of those currencies: In different ways, they
programmed and vigorous. Our aim to assist those folks that are range in size, color, and texture [17]-[21]. Thanks to
not ready to recognize which country’s currency note was. We globalization, international trade and commerce have soared to
use banknotes which are currency, may differ the size, texture, new levels. Banks' need to acquire currency knowledge on all
color. Our system helps in Indian currency identification, which currencies has been critical [22]-[26]. Understanding each note
is a fake currency or not. In India,’ currency’ is Transaction, so of a conversation is difficult for even for humans. It is
there is more value for currency in our social and economic imperative that an excellent note-recognition system is
development. We have used MATLAB software to recognize developed in the future. Here we are using the terms “dollar,”
other country currencies and Indian fake currency. “pound,” and “yen” to gauge our foreign currency reputation
Modernization within the money-related framework ensures
[27]-[33].
financial improvement, and nowadays, the Indian government
has become cognizant about this. Hence, Rs 1000 and Rs 500 As we probably know, in India, the Ministry of Finance and
notes' demonetization is the foremost up-to-date case of it. the RBI (the central bank) is assigned the job of providing cash
However, we have Rs 2000 as another benefit showcase. In light notes and coins in addition to the country's central bank [34]-
of the top elevated worth note, quite possibly degenerate [37]. He might, perhaps, be because he may, degenerate the
individuals will try to make it a fake. The real target of this unselected masses to counterfeit money with up-to-to-date
project is to contemplate distinctive key highlights of recent equipment and techniques. The mark of an expert sleuth is
certifiable money and utilize such systems to acknowledge and spotting the fake currency from the outset [38]-[41]. Typically,
confirm new cash circled by India's depository institution. There
currency systems are utilized in banks, shopping malls, hotels,
are different strategies which are utilized to acknowledge fake
business organizations, railway stations, cities, cities, and the
notes and certifiable one. By utilizing various parts of Digital
Image handling, such as picture preparing, Image Segmentation, government [42]-[45]. One person may understand that
feature extraction, and viewing pictures, we will then remove the $64,000 is a great deal of money, but everyone knows that
highlights of certified notes. It is a problematic errand for there is plenty of money, and no one has enough for anything.
recognizing counterfeit money. It is why we permit it in our system of open malpractice. To
date, many researchers have invested their time in the search
Keywords—Currency recognition; feature extraction; Image for methods to separate the real from the fraudulent currency.
Processing; Image Segmentation.
We are using software called MATLAB for the
computational and image analysis in this project. To
I. INTRODUCTION distinguish and eradicate the currency notes that are not likely
Technology is increasing fast these days. Consequently, the to be counterfeited, to drive their adoption in the past, many
banking sector is getting modern day to day [1]-[5]. We have to techniques have been suggested to tell the bill apart from the
address this project to solve problems facing related to rest of the currency. They will continue to be so in the future
currencies that common people are challenging issues are: [50]. A creative option would be to put the highlights of the
note to work. For instance, shading and size. Additionally, it is
1. Recognition of other country currency notes undesirable for a note to be too dirty or ripped. When a note's
2. Fake Indian currency notes lyrics are filthy, its value goes up dramatically. So, we must
restore the natural color of the cash note to make the change

978-1-6654-7886-1/22/$31.00 ©2022 IEEE

Authorized licensed use limited to: Marathwada Mitra Mandal's College of Engg. Downloaded on August 22,2024 at 08:48:51 UTC from IEEE Xplore. Restrictions apply.
significant All of the features are extracted through MATLAB image pre-processing are those that are typically performed
and digital image processing features. before the most significant amount of knowledge extraction.
3) Edge Detection: Edge detection is widespread in image
II. LITERATURE SURVEY processing because it relies on highly quantitative calculations
One of the primary tools to discover currency note and refers to any abrupt brightness changes in an image to
detection systems was initially expected to be developed in the points of increasing or decreasing brightness. This is typically
early 90s. They do not include aspects of authentication in their a relative brightness measurement, with accurate segments
system. Thus, the notes and picture sections are desired [43]. known as lines called “edges.” An edge detection process
As a result, it is notable to note that the system requires a set could be applied to finding the boundaries of images. It uses
angle and distance for all images used. Before images are discontinuities to gauge Edge detection is used in fields like
evaluated, the system runs a set of processing steps on the input image segmentation and data extraction.
pictures to arrive at the hue, saturation, and price parameters. It 4) Image Segmentation: Division of an image into
then calibrates these values to calculate geometry and
compares the results to check for deviation. Although this multiple sections More generally, to the point, the image is
methodology attempts to establish an overall equation, it would broken into multiple segments to make the analysis more
be inaccurate to say that there are equivalent degree notes [45]. straightforward and more accessible.
5) Feature Extraction and Comparison: Feature
The primary goal of the system is to create a simulacrum of extraction could be a reduction that is quantitative rather than
currency. There are scores of machines on the market that aid
categorical. If an algorithmic programme can process an input
those highly flexible when it comes to currencies. The cashier
must maintain multiple anti-forgery labels for common file, but it's found to be incredibly redundant, the file will be
currencies [16]. However, all parties must include commonly simplified into a subset of algorithms. Extracting features from
used return currency options and anti-counterfeiting labels to the input file is known as re-expression. Fastidious extraction,
have a reference work. You cannot count on manual coupled with a focus on the outcome, is expected to deliver
recognition to be correct. To find out the actual currency, we relevant results rather than accurate input.
must avoid interference with the results. Conjoined, it can save 6) Output: When it comes to currency recognition, the
time for police and administration by police quality currencies output will be in the text.
in less time [33].
B. Algorithm:
As it stands, the current banknote design is composed of
optoelectronic devices that rely on solar refraction of light for 1) Acquisition of image of currency note is chosen from
its power [36]. Several currency recognition machines are the folder.
available that use image recognition or neural networks. The 2) Image non inheritable is RGB image and now could be
primary function of existing currencies is to use image reborn to gray scale image.
processing and neural networks. Gaussian and Sigmoid appear 3) Edge detection of whole grey scale image.
in the hidden and output layers instead of the Gaussian process 4) Currently characteristics options of the paper money are
is shown to be more practical than the sigmoid for two cropped and segmented.
methods: a method shown to be practical and obscure [22]. 5) Once segmentation, characteristics of currency note
area unit extracted.
III. METHODOLOGY 6) Intensity of every feature is calculated.
To recognize the different countries currency notes, this 7) If the condition is glad, then the currency note is
system is developed. This system follows the following alleged as original otherwise fake.
methodology for recognizing currency notes which are given
for testing. IV. RESULTS AND DISCUSSIONS

A. Block diagram of the system: A. Results for Currency Recognition

The currency recognition system recognizes the currency
Image Image Pre- Edge
Acquisition processing Detection
name of which country it belongs to. The following is the steps
for the recognition of the currency name. The system displays
the currency name if it found. The input image is uploaded
Comparison Feature Image
through a file folder in MATLAB acquired for the system to
and output Extraction Segmentation recognize the currency name to which country it belongs.
We are testing the specific currencies for testing the
Fig. 1. Block diagram of the system. currency recognition system. Here, we are assuming the
American dollar, pound, rupee, and yen. The first step is to take
1) Image Acquisition: Image acquisition is capturing an an input banknote for currency recognition. We have divided
image from any source like a digital camera, scanner. each section into functions to perform and obtain the accurate
2) Image Pre-processing: Our primary objective is to results of currency recognition.
make the images and to facilitate the manipulation of datasets.

Authorized licensed use limited to: Marathwada Mitra Mandal's College of Engg. Downloaded on August 22,2024 at 08:48:51 UTC from IEEE Xplore. Restrictions apply.
 The following are the parameters used for this currency
recognition system to check whether the input image is
matched to the degraded image or not by calculating the
parameters like PSNR, mean square error and similarity index.

TABLE I. RESULTS OF PARAMETERS FOR CURRENCY RECOGNITION

Testing Input Image Mean square error PSNR Structural
with Output Image (MSE) In (dB) Similarity Index
Rupee currency 936.9132 18.4138 0.4807 Fig. 5. Sobel Edge Detected Image.
Pound currency 2437.9966 14.2605 0.0436
Dollar currency 2101.5129 14.9055 0.0012
Yen currency 937.9294 18.4091 0.3674

We are comparing to all country’s currency notes taken for

the input. We observed that the structural similarity index for
the rupee is the highest among them. The quality of the rupee
currency is high compared to the other currency inputs.

B. Results for Fake Currency Recognition

Fig. 6. Canny Edge Detected Image.
The results are obtained for fake Indian currency detection
for 2000, 500 notes are as follows. Certain features are
extracted and detected when the original image and fake are D. Output
given are inputs. In MATLAB, the results will be showing in
figures.

Fig. 7. Real note in output window.

Fig. 8. Fake note in output window.

Fig. 2. Acquired Image.

C. Pre-Processing steps:

Fig. 9. Comparison of real and fake.

V. CONCLUSION
Fig. 3. Gray Scale Image.
As we stated earlier, our currency recognition system works
for some country currencies, so this system’s future work is to
identify all country currency names. Our system helps people
who cannot recognize other country currency names. The
proposed system can implement for all countries currency
notes and for recognition of which country note it is and for
fake currency detection system includes remaining Indian
notes. The strategy is straightforward and simple to actualize.
This implementation is versatile so that it can be applied in real
life, day-to-to-day settings. The procedure begins with
Fig. 4. Sharpened Image.

Authorized licensed use limited to: Marathwada Mitra Mandal's College of Engg. Downloaded on August 22,2024 at 08:48:51 UTC from IEEE Xplore. Restrictions apply.
obtaining an image, proceeds through the highlights' layers, Operational Transconductance Amplifier with Bias Current,”
and ends at the correlation of those layers. Therefore, this International Journal of System Assurance Engineering and
Management, vol. 12, iss. 6, 2021, pp. 1173-1179.
undertaking will be created for those who know nothing about
[18] Sadulla Shaik, Anil Kumar Kurra, and A. Surendar, “Statistical Analysis
money (or financial issues in general) about money. These of Arbiter Physical Unclonable Functions using Reliable and Secure
historic financial history highlights like a sequential number, Transmission Gates,” International Journal of Simulation--Systems,
security string, Mahatma Gandhi's picture, and Identification Science & Technology, vol. 19, no. 4, 2018, pp. 6.1-6.6.
Mark were extracted from the money note. [19] R. R. Vallabhuni, P. Shruthi, G. Kavya and S. Siri Chandana,
“6Transistor SRAM Cell designed using 18nm FinFET Technology,”
2020 3rd International Conference on Intelligent Sustainable Systems
REFERENCES (ICISS), Thoothukudi, India, 2020, pp. 1584-1589, doi:
[1] Trupti Pathrabe G and Swapnili Karmore 2011 Int. J. Comp Trends Tech 10.1109/ICISS49785.2020.9315929.
152-156. [20] Vallabhuni Vijay, and Avireni Srinivasulu, “A low power waveform
[2] Vipin Kumar Jain and Ritu Vijay, “Indian currency denomination generator using DCCII with grounded capacitor,” International Journal
Identification using image processing technique,” International Journal of Public Sector Performance Management, vol. 5, 2019, pp. 134–145.
of Computer Science and Information Technologies, 2013, pp. 126-128. [21] Rajeev Ratna Vallabhuni, A. Karthik, CH. V. Sai Kumar, B. Varun, P.
[3] R. Bhavani, A. Karthikeyan, “A Novel Method for Counterfeit Banknote Veerendra, and Srisailam Nayak, “Comparative Analysis of 8-Bit
Detection,” IJCSE, vol .2, iss. 4, pp. 165-167, April 2014. Manchester Carry Chain Adder Using FinFET at 18nm Technology,”
2020 3rd International Conference on Intelligent Sustainable Systems
[4] Andrew S. Glassner. Principles of Digital Image Synthesis. Morgan
(ICISS), Thoothukudi, India, 2020, pp. 1579-1583, doi:
Kaufmann Publishers, 1995.
10.1109/ICISS49785.2020.9316061.
[5] A.Ms. Trupti Pathrabe and B.Dr.N.G. Bawane, “Paper Currency
[22] V. Vijay, J. Prathiba, S. Niranjan reddy, Ch. Srivalli, and B. Subbarami
Recognition System Using Characteristics Extraction and Negativity
reddy, “Performance evaluation of the CMOS Full adders in TDK 90 nm
Correlated NN Ensemble,” Int. Journal of Latest Trends in Computing,
Technology,” International Journal of Systems, Algorithms &
2010.
Applications, vol. 2, iss. 1, 2012, pp. 711.
[6] Rafael C. Gonzalez, Richard E. Woods. Digital Image Processing 2nd
[23] Anil Kumar Kurra, Usha Rani Nelakuditi, “A Reliable Current Starved
edition. Pearson Education, New York, 2001.
Inverter based Arbiter Puf Architecture for Iot Applications,”
[7] Mark S. Nixon and Alberto S. Aguado. Feature Extraction and Image International Journal of Engineering and Advanced Technology
Processing. Academic Press, 2008. (IJEAT), vol. 9, iss. 1S5, 2019, pp. 163-167.
[8] Deborah M, Soniya C and Prathap 2014 Int J Innov Sci Engg & Tech 1 [24] Rajeev Ratna Vallabhuni, J. Sravana, M. Saikumar, M. Sai Sriharsha,
151-57. and D. Roja Rani, “An advanced computing architecture for binary to
[9] Vallabhuni Vijay, C. V. Sai Kumar Reddy, Chandrashaker Pittala, thermometer decoder using 18nm FinFET,” 2020 Third International
“System and Method to Improve Performance of Amplifiers Using Bias Conference on Smart Systems and Inventive Technology (ICSSIT),
Current,” The Patent Office Journal No. 43/2019, India. International Tirunelveli, India, 20-22 August, 2020, pp. 510–515.
classification: C12Q1/6869. Application No. 201941042648 A. [25] Vallabhuni Vijay, “Second Generation Differential Current Conveyor
[10] Kurra, A. K., and S. Sadulla, “Analysis of physical unclonable functions (DCCII) And Its Applications,” Vignan's Foundation for Science,
(PUFS) for secure key generation on smartcard chips,” Journal of Technology & Research (Deemed to be University), Guntur, 2017.
Advanced Research in Dynamical and Control Systems, vol. 9, 2017, pp. [26] Mohammad Khadir, Kancharapu Chaitanya, S. Sushma, V. Preethi, and
1735-1745. Vallabhuni Vijay, “Design Of Carry Select Adder Based On A Compact
[11] Rajeev Ratna Vallabhuni, G. Yamini, T. Vinitha, and S. Sanath Reddy, Carry Look Ahead Unit Using 18nm Finfet Technology,” Journal of
“Performance analysis: D-Latch modules designed using 18nm FinFET Critical Reviews, vol. 7, iss. 6, 2020, pp. 1164–1171.
Technology,” 2020 International Conference on Smart Electronics and [27] V. Vijay, J. Prathiba, S. Niranjan Reddy, V. Raghavendra Rao, “Energy
Communication (ICOSEC), Tholurpatti, India, 10-12, September 2020, efficient CMOS Full-Adder Designed with TSMC 0.18μm Technology,”
pp. 1171–1176. International Conference on Technology and Management (ICTM-
[12] Ch. Srivalli, S. Niranjan reddy, V. Vijay, J. Pratibha, “Low power based 2011), Hyderabad, India, June 8-10, 2011, pp. 356-361.
optimal design for FPGA implemented VMFU with equipped SPST [28] P. Saritha, J. Vinitha, S. Sravya, V. Vijay, and E. Mahesh, “4-Bit Vedic
technique,” National Conference on Emerging Trends in Engineering Multiplier with 18nm FinFET Technology,” 2020 International
Application (NCETEA-2011), India, June 18, 2011, pp. 224-227. Conference on Electronics and Sustainable Communication Systems
[13] Vallabhuni Vijay, C. V. Sai Kumar Reddy, Chandrashaker Pittala, (ICESC), Coimbatore, India, 2020, pp. 1079–1084.
Sonagiri China Venkateswarlu, “System for Reducing Crosstalk Delays [29] Vallabhuni Rajeev Ratna, M. Saritha, Saipreethi. N, V. Vijay, P.
In Electronic Devices Using A CMOS Inverter,” The Patent Office Chandra Shaker, M. Divya, Shaik Sadulla, “High Speed Energy
Journal No. 43/2019, India. International classification: H03B5/18. Efficient Multiplier Using 20nm FinFET Technology,” Proceedings of
Application No. 201941042515 A. the International Conference on IoT Based Control Networks and
[14] Chandra Shaker Pittala, and Vallabhuni Vijay, “Design of 1-Bit FinFET Intelligent Systems (ICICNIS 2020), Palai, India, December 10-11, 2020,
Sum Circuit For Computational Applications,” Lecture Notes in pp. 1-8. http://dx.doi.org/10.2139/ssrn.3769235
Networks and Systems, 2021. [30] Usha Rani Nelakuditi, Anil Kumar kurra, Haritha Dasari,
[15] Vallabhuni Vijay, C. V. Sai Kumar Reddy, Chandrashaker Pittala, P “Implementation of High-Performance Wallace Tree Multiplier Using
ASHOK BABU, “System to Obtain Finite Gain and Noise of an Compressors,” Journal of Critical Reviews, vol. 7, iss. 2, 2020, pp. 977-
Electrocardiogram Amplifier,” The Patent Office Journal No. 43/2019, 981.
India. International classification: H03F3/38. Application No. [31] Rajeev Ratna Vallabhuni, D.V.L. Sravya, M. Sree Shalini, and G. Uma
201941042674 A. Maheshwararao, “Design of Comparator using 18nm FinFET
[16] Bandi Mary Sowbhagya Rani, Vasumathi Devi Majety, Chandra Shaker Technology for Analog to Digital Converters,” 2020 7th International
Pittala, Vallabhuni Vijay, Kanumalli Satya Sandeep, Siripuri Kiran, Conference on Smart Structures and Systems (ICSSS), Chennai, India,
“Road Identification Through Efficient Edge Segmentation Based on 23-24 july, 2020, pp. 318–323.
Morphological Operations,” Traitement du Signal, vol. 38, no. 5, Oct. [32] Vallabhuni Vijay, Pittala Chandra shekar, Shaik Sadulla, Putta Manoja,
2021, pp. 1503-1508. Rallabhandy Abhinaya, Merugu rachana, Nakka nikhil, “Design and
[17] Vallabhuni Vijay, C. V. Sai Kumar Reddy, Chandra Shaker Pittala, performance evaluation of energy efficient 8-bit ALU at ultra low
Rajeev Ratna Vallabhuni, M. Saritha, M. Lavanya, S. China supply voltages using FinFET with 20nm Technology,” VLSI
Venkateswarlu and M. Sreevani, “ECG Performance Validation Using Architecture for Signal, Speech, and Image Processing, edited by

Authorized licensed use limited to: Marathwada Mitra Mandal's College of Engg. Downloaded on August 22,2024 at 08:48:51 UTC from IEEE Xplore. Restrictions apply.
 Durgesh Nandan, Basant Kumar Mohanty, Sanjeev Kumar, Rajeev using 18nm FinFET for Memory Storing Applications,” Proceedings of
Kumar Arya, CRC press, 2021. the 2nd International Conference on IoT, Social, Mobile, Analytics &
[33] Rajeev Ratna Vallabhuni, Jujavarapu Sravana, Chandra Shaker Pittala, Cloud in Computational Vision & Bio-Engineering (ISMAC-CVB 2020),
Mikkili Divya, B.M.S.Rani, Vallabhuni Vijay, “Universal Shift Register 2020, pp. 1-10.
Designed at Low Supply Voltages in 20nm FinFET Using Multiplexer,” [40] B.M.S Rani, Divyasree Mikkili, Rajeev Ratna Vallabhuni, Chandra
Lecture Notes in Networks and Systems, 2021. Shaker Pittala, Vijay Vallabhuni, Suneetha Bobbillapati, H. Bhavani
[34] Vallabhuni Vijay, and Avireni Srinivasulu, “A Novel Square Wave Naga Prasanna, “Retinal Vascular Disease Detection from Retinal
Generator Using Second Generation Differential Current Conveyor,” Fundus Images Using Machine Learning,” Australia patent 2020101450.
Arabian Journal for Science and Engineering, vol. 42, iss. 12, 2017, pp. [41] Venkateswarlu, S.C., Kumar, N.U., Kumar, N.S., Karthik, A., and Vijay,
4983–4990. V., “Implementation of Area optimized Low power Multiplication and
[35] Rani, B.M.S., Ratna, V.R., Srinivasan, V.P. et al. Disease prediction Accumulation,” International Journal of Innovative Technology and
based retinal segmentation using bi-directional ConvLSTMU-Net. J Exploring Engineering (IJITEE), vol. 9, iss. 9, 2019, pp. 2278–3075.
Ambient Intell Human Comput, 2021. https://doi.org/10.1007/s12652- [42] P. Chandrashekar, R. Karthik, O. Koteswara Sai Krishna, and Ardhi
021-03017-y Bhavana, “Design of low threshold Full Adder cell using CNTFET,”
[36] Babu, P. Ashok, et al. “Realization of 8 x 4 Barrel shifter with 4-bit International Journal of Applied Engineering Research, vol. 12, no. 12,
binary to Gray converter using FinFET for Low Power Digital 2017, pp. 3411–3415.
Applications,” Journal of Physics: Conference Series, vol. 1714, no. 1, [43] Rajeev Ratna Vallabhuni, S. Lakshmanachari, G. Avanthi, and
IOP Publishing, 2021. Vallabhuni Vijay, “Smart Cart Shopping System with an RFID Interface
[37] Ch. Srivalli, S. Niranjan reddy, V. Vijay, J. Pratibha, “Optimal design of for Human Assistance,” 2020 3rd International Conference on
VLSI implemented Viterbi decoding,” National conference on Recent Intelligent Sustainable Systems (ICISS), Thoothukudi, India, 2020, pp.
Advances in Communications & Energy Systems, (RACES-2011), 165-169, doi: 10.1109/ICISS49785.2020.9316102.
Vadlamudi, India, December 5, 2011, pp. 67-71. [44] Rajeev Ratna Vallabhuni, M. Saritha, Sruthi Chikkapally, Vallabhuni
[38] Vijay, V., J. Prathiba, S. Niranjan Reddy, and P. Praveen Kumar, “A Vijay, Chandra Shaker Pittala, and Sadulla Shaik, “Universal Shift
REVIEW OF THE 0.09 µm STANDARD FULL ADDERS,” Register Designed at Low Supply Voltages in 15nm CNTFET Using
International Journal of VLSI Design & Communication Systems, vol. 3, Multiplexer,” Lecture Notes in Networks and Systems, 2021.
no. 3, 2012, p. 119. [45] Manchala Sreeja, Vallabhuni Vijay, “A Unique Approach To Provide
[39] Rajeev Ratna Vallabhuni, K.C. Koteswaramma, B. Sadgurbabu, and Security For Women By Using Smart Device,” European Journal of
Gowthamireddy A, “Comparative Validation of SRAM Cells Designed Molecular & Clinical Medicine, vol. 7, iss. 1, 2020, pp. 3669-3683.

Authorized licensed use limited to: Marathwada Mitra Mandal's College of Engg. Downloaded on August 22,2024 at 08:48:51 UTC from IEEE Xplore. Restrictions apply.