International Journal For Technological Research In Engineering

Volume 1, Issue 11, July-2014 ISSN (Online): 2347 - 4718

VARIOUS FAKE CURRENCY DETECTION TECHNIQUES

Megha Thakur1, Amrit Kaur2
Department Of Electronics and Communication
Punjabi University
Patiala, Punjab, India

Abstract: This paper presents the various fake currency assess the menace of fake currency, puts the total amount of
detection techniques. Fake currency is imitation currency fake currency in circulation in India at about Rs 1,69,000
produced without the legal sanction of the state or crore as of 2000 (in other words, eight per million were
government. Producing or using fake currency is a form fake). The magnitude of the problem, therefore, is immense
of fraud or forgery. Over the past few years, as a result of [15]. To distinguish between fake and real currency notes has
the great technological advances in colour printing, become increasingly difficult mainly due to the fact that
duplicating and scanning, counterfeiting problems have counterfeits are now printed with state of the art technology
become more and more serious. Therefore the issue of using security paper [15]. Due to great technological
efficiently distinguishing counterfeit banknotes from advancement counterfeiting problems have become more and
genuine ones via automatic Fake currency detection system more serious. Therefore the issue of efficiently
has become more and more important. Fake currency distinguishing counterfeit banknotes from genuine ones via
detection system can be used in places such as shops, banks automatic machines has become more and more important
counter and automated teller machine, auto seller machines [9]. The fake currency detection system is developed to
etc. We have reviewed different fake currency detection detect the fake currency by applying different techniques and
systems. The systems are developed using different methods methods on currency note. The fake currency detection
and algorithms. The benefits of this study for the reader are system should be able to recognize the note quickly and
that this study will provide information about the different correctly. The fake currency detection system should be able
methods and algorithms used for fake currency detection to recognize currency note from any side. Currency
system. They can compare the detection systems. Detection recognition system can be used in places such as shops,
ability depends on the currency note characteristics of banks counter and automated teller machine, auto seller
particular country and extraction of features. machines etc [12]. We have reviewed different fake currency
Keywords: Fake currency, Digital image processing, detection systems. The systems are developed using different
counterfeit detection. methods and algorithms. The benefits of this study for the
reader are that this study will provide information about the
I. INTRODUCTION different methods and algorithms used for fake currency
Fake currency is imitation currency produced without the detection system. They can compare the detection systems.
legal sanction of the state or government. Producing or using Detection ability depends on the currency note
Fake currency is a form of fraud or forgery. Counterfeiting is characteristics of particular country and extraction of
almost as old as money itself. Before the introduction features [11].
of paper money, the most prevalent method of counterfeiting
involved mixing base metals with pure gold or silver. A form II. DIFFERENT FAKE CURRENCY DETECTION
of counterfeiting is the production of documents by TECHNIQUES
legitimate printers in response to fraudulent instructions. A. Commonly Used Methods to Detect Fake Currency
Some of the ill-effects that counterfeit money has on society 1. See through Register
include a reduction in the value of real money; and increase The small floral design printed both on the front (hollow)
in prices due to more money getting circulated in the and back (filled up) of the note in the middle of the vertical
economy- an unauthorized artificial increase in the money band next to the Watermark has an accurate back to back
supply; a decrease in the acceptability of paper money; and registration. The design will appear as floral design when
losses, when traders are not reimbursed for counterfeit money seen against the light.
detected by banks, even if it is confiscated. According to
figures disclosed in Parliament, during the 2006-09, 7.34 lakh 2. Water marking
of Rs 100 notes, 5.76 lakh of Rs 500 notes and 1.09 lakh of The Mahatma Gandhi Series of banknotes contain the
Rs 1000 notes, all fakes, have been seized. The number of Mahatma Gandhi watermark with a light and shade effect
fake notes per million have increased from 4.4 in 2007-08 to and multi-directional lines in the watermark window.
7.51 in 2011-12. For higher denominated notes (Rs 500 and
Rs 1000) the increase was almost double: from 9.7 in 2007- 3. Fluorescence
08 to 18.2 in 2011-12. This is considered only as the “tip of Number panels of the notes are printed in fluorescent ink.
the iceberg” when compared to the total unseizured notes in The notes also have optical fibres. Both can be seen when the
the Indian market. The Nayak Committee, appointed to notes are exposed to ultra-violet lamp.

www.ijtre.com Copyright 2014.All rights reserved. 1309

International Journal For Technological Research In Engineering
Volume 1, Issue 11, July-2014 ISSN (Online): 2347 - 4718

4. Security Thread 9. Optically Variable Ink

The Rs.500 and Rs.100 notes have a security thread with This is a new feature included in the Rs.1000 and Rs.500
similar visible features and inscription „Bharat‟ (in Hindi), notes with revised color scheme introduced in November
and „RBI‟. When held against the light, the security thread on 2000. The numeral 1000 and 500 on the obverse of Rs.1000
Rs.1000, Rs.500 and Rs.100 can be seen as one continuous and Rs.500 notes respectively is printed in optically variable
line. The Rs.5, Rs.10, Rs.20 and Rs.50 notes contain a ink viz., a color-shifting ink. The colour of the numeral
readable, fully embedded windowed security thread with the 1000/500 appears green when the note is held flat but would
inscription „Bharat‟ (in Hindi), and „RBI‟. The security change to blue when the note is held at an angle.
thread appears to the left of the Mahatma's portrait.
B. Digital Image Processing Method To Detect Fake
Currency
The design flow of fake currency detection system includes
eight stages: Image acquisition, pre-processing, gray scale
conversion, edge detection, image segmentation, feature
extraction, comparison and output [11]. This system is works
on two images, one is test currency image on which
authentication is to performed and other is the original
currency image.

Fig.1 Original Indian 100 denomination

5. Intaglio Printing
The portrait of Mahatma Gandhi, the Reserve Bank seal,
guarantee and promise clause, Ashoka Pillar Emblem on the
left, RBI Governor's signature are printed in intaglio i.e. in
raised prints, which can be felt by touch, in Rs.20, Rs.50,
Rs.100, Rs.500 and Rs.1000 notes.

6. Latent image
On the obverse side of Rs.1000, Rs.500, Rs.100, Rs.50 and
Rs.20 notes, a vertical band on the right side of the
Mahatma Gandhi‟s portrait contains a latent image showing
the respective denominational value in numeral. The latent
image is visible only when the note is held horizontally at eye
level.

7. Micro lettering
This feature appears between the vertical band and Mahatma
Gandhi portrait. It always contains the word „RBI‟ in Rs.5
and Rs.10. The notes of Rs.20 and above also contain the
denominational value of the notes in micro letters. This
feature can be seen well under a magnifying glass.

8. Identification Mark
Each note has a unique mark of it. A special feature in
intaglio has been introduced on the left of the watermark
window on all notes except Rs.10/- note. This feature is in
different shapes for various denominations (Rs. 20-Vertical
Rectangle, Rs.50- Square, Rs.100-Triangle, Rs.500-Circle,
and Rs.1000- Diamond) and helps the visually impaired to Fig.2 Flow Chart of Digital Image Processing Method To
identify the denomination. Detect Fake Notes

www.ijtre.com Copyright 2014.All rights reserved. 1310

International Journal For Technological Research In Engineering
Volume 1, Issue 11, July-2014 ISSN (Online): 2347 - 4718

1. Image Acquisition regions or objects. The level to which sub division is carried
There are various ways to acquire image such as with the depends on the problem being solved. Segmentation
help of camera or scanner. Acquired image should retain all algorithm for monochrome images generally are based on
the features [11]. one of the two basic properties of image intensity values-
 Discontinuity
2. Pre-Processing  Similarity.
Pre-processing of image are those operations that are In the first category, the approach is to partition an image
normally required prior to the main data analysis and based on abrupt changes in intensity such as edges in an
extraction of information. The aim of image pre-processing is image. The approach in the second category is based on
to suppress undesired distortions or enhance some image partitioning an image into regions that are similar according
features that are important for further processing or analysis. to a set of predefined criteria [9].
It includes
6. Feature Extraction
2.1 Image Adjusting In pattern recognition and in image processing, feature
When we get the image from a scanner, the size of the image extraction is the special form of dimensionality reduction. It
is so big. In order to reduce the calculation, we decrease the is the method of capturing the visual content of images for
size of image. Image Adjusting is done with the help of indexing and retrieval. When the input data to an algorithm
image interpolation. Interpolation is the technique mostly is too large to be processed and it is suspected to be
used for tasks such as zooming, rotating, shrinking, and for notoriously redundant (much data but not much information)
geometric corrections. then the input data will be transformed into a reduced
representation set of features (also named feature vector). If
2.2 Image smoothening the attributes extracted are carefully chosen, it is expected
When using a camera or a scanner and perform image that the attributes set will extract the relevant information
transfers, some noise will appear on the image. Image noise from the input data in order to perform the desired task using
is the random variation of brightness in images. Removing this reduced representation instead of the full size input.
the noise is an important step when image processing is being Feature extraction involves simplifying the amount of
performed. However noise may affect segmentation and resources required to describe the large set of data.
pattern matching. When performing smoothing process on a
pixel, the neighbor of the pixel is used to do some
transforming. After that a new value of the pixel is created.
The neighbor of the pixel is consisting with some other pixels
and they build up a matrix, the size of the matrix is odd
number, the target pixel is located on the middle of the
matrix. Convolution is used to perform image smoothing.
Also image smoothening can be done with the help of
median filter which more effective than convolution when
goal is to simultaneously reduce the noise preserving edges.
Median filter replaces a pixel via the median pixel of all the
neighborhoods [11].

3. Gray-scale conversion:
The image acquired is in RGB color. It is converted into gray
scale because it carries only the intensity information which
is easy to process instead of processing three components R
(Red), G(Green), B(Blue) [7].

4. Edge detection
Edge detection is a fundamental tool in image processing and
computer vision, particularly in the areas of feature detection
and feature extraction, which aim at identifying points in a
digital image at which the image brightness changes sharply
or, more formally, has discontinuities. Edge detection is one
of the fundamental steps in image processing, image analysis,
image pattern recognition, and computer vision techniques
[9].

5. Image segmentation
Image segmentation sub divides the image into its constituent Fig.3 Feature Extraction Approach

www.ijtre.com Copyright 2014.All rights reserved. 1311

International Journal For Technological Research In Engineering
Volume 1, Issue 11, July-2014 ISSN (Online): 2347 - 4718

Visual attributes of images are of two types- 2. How counterfeiters defeat this pen:
 Domain specific attributes which include The iodine in the pen reacts with starch that makes white
fingerprints, human faces. paper look brighter. Most unless bleached and starched. If
 General attributes which include color, texture, and there is no starch present in the paper then the pen will
shape. indicate – by remaining amber - that the note is original [6].

There are two types of attributes categorized under the shape E. Other techniques:
attribute extraction- The other anti-counterfeit device for the money is an
 Global attributes include moment invariant, aspect Ultraviolet counterfeit detection scanner. Best used in highly
ratio and circularity. lit point of sale locations, the UV detector identifies the
 Local attributes include boundary segments [7]. ultraviolet security features present in most currencies. By
simply placing the note in the detector, counterfeit currency
7. Comparison is immediately identified, without the need for an employee
Lastly the extracted features of test currency image are to closely examine the note [6].
compared with the extracted features of original currency
image, if it matches then the currency is original otherwise III. CONCLUSION
fake [7]. In this study, we discussed various fake currency detection
techniques, each one has its own significance. By using
C. MATLAB technique: mentioned methods we have observe that good results can be
In this technique one can split the red, blue, green obtained quickly and correctly. The benefits of this study for
components of a picture and name them as r1, g1, b1 which the reader are that this study will provide information about
correspond to image i.e. original currency note. Consider the different methods and algorithms used for fake currency
second image that is note to be tested. Split this image to detection system. They can compare the detection systems.
components r2, b2, g2. Construct a new image with
components as r1, g2, b1 or r2, g1, b1 or b2, g1, b1. But REFERENCES
r1,g2,b1 combination is most preferred because human eye is [1] Yufeng Kou, Chang-Tien Lu, Sirirat
sensitive to green component and most of our images Sinvongwattana S. ans Yo-Ping Huang, Survey of
contains maximum green component so that our output Fraud Detection Techniques, IEEE International
image will be much easier to identify the fake note more Conference on Networking, Sensing & Control, 0-
efficiently. After that compare newly constructed image with 7803-8193-9/04/$17.0020 2004 IEEE
image1.Calculate the threshold value of equivalence by [2] Jae-Kang Lee AND 11-Hwan Kim, New
calculating the standard deviation. If equivalence is above Recognition Algorithm for Various Kinds of Euro
40% then one can consider it as original note. Here consider Banknotes, 0-7803-7906-3/03/$17.00 Q2003 IEEE
40% value because note may be damaged. Parameters for [3] M. Tanaka, F. Takeda,K. Ohkouchi and Y.
measure of comparing images are Mean Square Error (MSE), Michiyuki, Recognition of Paper Currencies by
Peak Signal to Noise Ratio (PSNR in dB), and structural Hybrid Neural Network, 0-7803-4859-1/98
Content (SC).When combine two various components of two $10.0001998 IEEE
images then if note to be tested is original then only at the [4] Fumiaki Takeda AND Sigeru Omatu, A Neuro-
place of number we get variation. But in case of fake note Paper Currency Recognition Method Using
after applying the same code, one can observe that the image Optimized Masks by Genetic Algorithm, 0-7803-
overlapping is not done correctly. One can also see that the 2559-1195 $4.00 0 1995 IEEE
resultant image is blurred indicating fake note. So one can [5] Fumiaki Takeda and Sigeru Omatu, High Speed
confirm that it is a fake note [6]. Paper Currency Recognition by Neural Networks,
IEEE Transactions On Neural Networks, Vol. 6,
D. Counterfeit Detection Pen: No. 1, January 1995.
A counterfeit pen is simply an inexpensive device that is [6] D. Alekhya, G. DeviSuryaPrabha and G. Venkata
designed to determine if a currency note is original or fake. Durga Rao, Fake Currency Detection Using Image
The pen contains a tincture of iodine as ink which, when Processing and Other Standard Methods,
drawn over a note, will remain amber or brown. According to International Journal of Research in Computer and
one manufacturer the ink will turn black if the note is fake. Communication Technology, Vol 3, Issue 1,
January- 2014
1. Working of counterfeit pen: [7] Rubeena mirza and veenti nanda, Design and
The iodine in the pen reacts with starch, which is the primary Implementation of Indian Paper Currency
component that makes white paper look brighter. Most Authentication System Based on Feature Extraction
commercial paper, made from wood pulp, is brown unless by Edge Based Segmentation Using Sobel Operator,
bleached and starched. If there is no starch present in the IJERD, Volume 3, Issue 2 (August 2012), PP. 41-
paper then the pen will indicate - by remaining amber- that 46.
the note is original. [8] Amol A. Shirsath, S. D. Bharkad, A Review of

www.ijtre.com Copyright 2014.All rights reserved. 1312

International Journal For Technological Research In Engineering
Volume 1, Issue 11, July-2014 ISSN (Online): 2347 - 4718

Paper Currency Recognition System, IOSR Journal

of Computer Engineering (IOSR-JCE) e-ISSN:
2278-0661, p- ISSN: 2278-8727Volume 10, Issue 1
(Mar. - Apr. 2013).
[9] Rubeena Mirza and Vinti Nanda, Paper Currency
Verification System Based on Characteristic
Extraction Using Image Processing, International
Journal of Engineering and Advanced Technology
(IJEAT) ISSN: 2249 – 8958, Volume-1, Issue-3,
February 2012.
[10] Hanish Aggarwal and Padam Kumar, Indian
Currency Note Denomination Recognition in Color
Images, International Journal on Advanced
Computer Engineering and Communication
Technology Vol-1 Issue:1 :ISSN 2278 – 5140
[11] Amol A. Shirsath and S. D. Bharkad, Survey Of
Currency Recognition System Using Image
Processing, International Journal of Computational
Engineering Research, Vol, 03, Issue 7.
[12] Ahmed Ali Abbasi, A Review on Different
Currency Recognition System for Bangladesh India
China and Euro Currency, Research Journal of
Applied Sciences, Engineering and Technology
7(8): 1689-1690, 2014.
[13] Rumi Ghosh and Rakesh Khare, An Elegant Neural
Network based draw near for currency Recognition,
JECET; June – August-2013; Vol.2.No.3, 876-882.
[14] Kishan Chakraborty, Jordan Basumatary, Debasmita
Dasgupta, Jagadish Chandra Kalita and Subra
Mukherjee, Recent Developments In Paper
Currency Recognition System, IJRET, Volume: 02
Issue: 11 | Nov-2013
[15] N. Manoharan, Counterfeit Currency as a Threat to
India‟s Internal Security.

www.ijtre.com Copyright 2014.All rights reserved. 1313