International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248

Volume: 4 Issue: 3 474 – 479

_______________________________________________________________________________________________
Counterfeit Currency Detection System

Anwar Wanjara, Priyanka Satija, Nikita Patil, Samiksha Chindhalore, Shubham Gaulkar, Prof. A. D. Gotmare

Bapurao Deshmukh College of Engineering Department of Computer Engineering

Abstract: The elevation of color printing technology has increased the rate of fake currency note printing on a very large scale. Years before, the
printing could be done in a print house, but now anyone can print a currency note with maximum accuracy using a simple laser printer. This
results the issue of fake notes instead of the genuine ones has been increased very largely. India has been unfortunately damned with the
problems like corruption and black money .The counterfeit of currency notes is also a big problem to it. This problem leads to design of a system
that detects the fake currency note in a less time and in a more efficient manner. The proposed system gives an approach to verify the
genuineness of Indian currency. Verification of currency note is done by the concepts of image processing. The above article describes
extraction of various features of Indian currency notes. MATLAB software is used to extract the features of the note. The proposed system has
got advantages like simplicity, high performance speed, portability. The result will predict whether the currency is counterfeit.

Keywords: Fake Currency,Currency Mapping, Counterfeit Detection

__________________________________________________*****_________________________________________________

1. INTRODUCTION1 1.1.Process of Image preprocessing:

It involves the operations required prior to data analysis and
Technology is growing rapidly. Accordingly the banking
information extraction. Here image resizing is done.
sector is also getting advance day by day. This bringsa
intense need of automatic fake currency detection in
1.2 Process of Image segmentation:
Automatic Teller Machine and Automatic Goods Seller
Image segmentation is done to retrieve only the currency
Machine. Many researchers are encouraged to develop
part of the image. This basically is removal of background
robust and efficient automatic currency detection machine.
and other unimportant visual information.
The technology of currency detection basically aims for
identifying and extracting visible and invisible features of
1.3 Features Extraction:
currency notes. Until now, many techniques have been
After the image segmentation various features is extracted
proposed to identify the currency note. But the best way is
with the help of canny edge Filter and pattern matching of
to use the visible features of the note. For example, color
those features with preexisting images.
and size. But this way is not helpful if the note is dirty or
torn. If a note is dirty, its colorcharacteristic are changed
1.4 Mapping:
widely. So it is important to extractfeatures of the image of
Mapping of features with the data available in dataset is
the currency note and apply proper algorithm to improve
done.
accuracy to recognize the note.
The image of the currency note is captured through a
2. EXISTING SYSTEMS
digital camera. The hidden features of the note are
Fake money scanner
highlighted. Now processing on the image is done on that
acquired image using concepts like image segmentation,
edge information of image and characteristics feature
extraction.MATLAB is the perfect tool for computational
work, and analysis.Feature extraction of images is
challenging task in digital image processing.It involves
extraction of invisible and visible features of Indian
currency notes. This approach consists of different steps like
dataset, edge detection, image preprocessing, feature
extraction, image, mapping and decision making.Here, the
image is captured by a simple digital camera such that all the
features are highlighted. Image is then stored for further
processing.

Figure 2.1 Fake Money Scanner App

474
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 4 Issue: 3 474 – 479
_______________________________________________________________________________________________
The Fake Money Scanner App which is available on
playstore doesnot provides required output.Though the
genuine note has been given as input but the output is false.

Figure2.3 Counterfeit Detector Pen

3. BACKGROUND STUDY

Figure2.2 Output of Fake Money Scanner App

In the following application input is been provided as

500Rs.note which is genuine but the following application
fails to identify whether the note is real or fake. As a result
the output of the application shows that the note is fake
which resembles the false output.
Figure 3.1 Features of 500Rs note
Counterfeit Detector Pen:
A counterfeit banknote detection pen is a device used to 1.See Through Register: The see through register is the
apply an iodine-based ink to banknotes in an attempt to small floral design on the front (hollow) and back (filled up)
determine their authenticity. Typically, genuine banknotes on the note in the middle of the note between the vertical
are printed on paper based on cotton fibers, and do not band and the watermark window. This floral design can be
contain the starches that react with iodine. While detecting a seen as one when the note is held in the light.
note, pen contains an iodine solution that reacts with the
starch in wood-based paper to create a black stain. When the 2.Latent Image: The vertical band which is present on the
solution is applied to the fiber-based paper used in real right side of the Mahatma Gandhi‟s portrait contains a latent
notes, no discoloration occurs. image that shows the respective denominational value of the
note in numeral. This feature is visible only when the note is
Features: held horizontally at the eye level.
 Detect notes printed on normal copier paper instead
of the fine papers used by genuine authority. 3.Denominational Numeral:This is a special security
 Effective against notes printed on standard printer feature. The numeral values of the banknotes are printed in
or photocopier paper. optically variable ink, i.e. color shifting ink. This feature
was introduced for Rs.500 and Rs.1000 notes in November
 If the note is real, the ink turns yellow.
2000. The color of the numerals 500/1000 on the obverse of
 If the note is counterfeit, it will turn a dark blue or
the note can be seen in blue color when the note is held flat
black. but when it is held at different angles, the color changes to
green.
Limitations:
 Not applicable for starch free papers.
4. Portrait: Portrait of Mahatma Gandhi present at the
 Effectiveness is much lower.
center of note.

5.SecurityThread: The security thread has a width of 3mm

and it also has „Bharat‟ in Hindi and RBI seal in scripted on
475
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 4 Issue: 3 474 – 479
_______________________________________________________________________________________________
it. The color of security thread changes from green to blue 5. DATA FLOW
when held at different angles. It is greenish on the reverse
side and fluorescent text is visible when the note is held
under ultraviolet light. The security thread is alternately Uploading the Currency Note
visible normally but a long line of thread can be seen when
seen in the light. Convert image into gray scale

6.MicroLettering: This feature can be seen between the Gray scale to binary image
Mahatma Gandhi watermark and the vertical band. For Rs.5
and Rs.10 notes it has „RBI‟ printed on it. This feature can Extraction of Feature from the
only be seen under a microscope. Currency Note

7.Watermark: The Mahatma Gandhi series of notes contain

the Mahatma Gandhi portrait with light and shade effect and Currency Detection Detection Detection
multidirectional lines in the obverse side of a note. Series Of of Silver of
Number Currency Line in Currency
8.Number Panel: It is a feature with numeral growing from Extractio Color Image Pattern
n
small to big on the top left and the bottom right sides of
note.
Apply Calculating Calculating Extract the
OCR on the RGB the Pattern
9. Intaglio Printing: The portrait of Mahatma Gandhi(4), Number Value of Frequency
the RBI Governor‟s signature(7), the guarantee and the series Currency White and
detection Black Pixel
promise clause, the Ashoka Pillar Emblem at the left are
printed in intaglio, i.e. raised prints. This feature can be felt
Generated
by touch. number Number of Compare with
Black pixel the Database
should be
greater than
10.IdentificationMark: This is also another special feature Threshold
The generated number
printed in intaglio on the left of the watermark window on is compared with Pattern detected
all the banknotes except on Rs.10 note. This feature is in blacklist numbers match Database
Pattern
different shapes for different notes. The identification mark
is vertical rectangle for Rs.20 note, square for Rs.50 note,
triangle for Rs.100 note, circle for Rs.500 note, H for Rs.200
note and horizontal rectangle for Rs.2000 note. This feature
is for the blind people.
Mapping If match is
11.Angular Bleed Lines:Seven angular bleed lines on the Features found
Database
left and right side in raised print.

No
Action
4. PROPOSED SYSTEM If match Alarm
is will be
notfound generated
Implementation of the Counterfeit currency detection
cover following below point
[1]Currency Number Extraction from the Note.
[2]Pattern finding for each currency and detection of Figure 5.1 Data Flow
fakeand genuine currency according to the currency number.
[3]Detection of sliver line in middle of the currency note. 6. ALGORITHMS
[4]Detection of the fake or Genuine Currency using the
Image Color as Every Currency note we have different RGB 1.OCR:
Value.
[5]Black List should be maintained using the server. User The goal of Optical Character Recognition
can click on the update Button so that the user will sync the (OCR) is to classify optical patterns (often contained in a
blacklist and have the latest one. digital image) corresponding to alphanumeric or other
characters. The process of OCR involves several steps
including segmentation, feature extraction, and

476
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 4 Issue: 3 474 – 479
_______________________________________________________________________________________________
classification. Each of these steps is a field unto itself, and is commercial environment. The key factors involved in the
described briefly here in the context of a Matlab implementation are: an optimal selection of features which
implementation of OCR. One example of OCR is shown categorically defines the details of the characters, the
below. A portion of a scanned image of text, borrowed from number of features and a low image processing time.
the web, is shown along with the corresponding (human The process of converting an image to editable document
recognized) characters from that text.A few examples of is separated to several steps; every step is a set of related
OCR applications are listed here. The most common for use algorithms which do a piece of OCR job.
OCR is the first item; people often wish to convert text
documents to some sort of digital representation. 2.IMPIXELINFO:
Impixelinfo creates a Pixel Information tool in the current
1. People wish to scan in a document and have the text of figure. The Pixel Information tool displays information
that document available in a wordprocessor. about the pixel in an image that the pointer is positioned
2. Recognizing license plate numbers over. The tool can display pixel information for all the
3. Post Office needs to recognize zip-code images in a figure.

The Pixel Information tool is a uipanel object, positioned

in the lower-left corner of the figure. The tool contains the
text label Pixel info: followed by the pixel information.
Before you move the pointer over the image, the tool
contains the default pixel information text (X,Y) Pixel
Value. Once you move the pointer over the image, the
information displayed varies by image type, as shown in the
following table. If you move the pointer off the image, the
pixel information tool displays the default pixel information
label for that image type.

Fig.6.1: Shows general steps used in ocr

Optical character recognition (OCR) is a process

ofconverting a printed document or scanned page into
ASCIIcharacters that a computer can recognize. Computer For example, for grayscale (intensity) images, the pixel
systems equipped with such an OCR system improve the information tool displays the x and y coordinates of the pixel
speed of input operation, decrease some possible human and its value, as shown in the following figure.
errors and enable compact storage, fast retrieval and other
file manipulations. The range of applications include postal
code recognition, automatic data entry into large
administrative systems, banking, automatic cartography and
reading devices for blind. Accuracy, flexibility and speed
are the main features that characterize a good OCR system.
If you want to display the pixel information without the
The performance of the systems has been constrained by the
“Pixel Info” label, use the impixelinfoval function.
dependence on font, size and orientation.The recognition
rate in these algorithms depends on the choice of features. impixelinfo(h) creates a Pixel Information tool in the
Most of the existing algorithms involve extensive processing figure specified by h, where h is a handle to an image, axes,
on the image before the features are extracted that results in uipanel, or figure object. Axes, uipanel, or figure objects
increased computational time.In this paper, we discuss a must contain at least one image object.
pattern matching based method for character recognition
that would effectively reduce the image processing time impixelinfo(hparent,himage) creates a Pixel Information
while maintaining efficiency and versatility. The parallel tool in hparent that provides information about the pixels
computational capabilities of neural networks [1, 3, 5] in himage. himage is a handle to an image or an array of
ensures a high speed of recognition which is critical to a image handles. hparent is a handle to the figure or uipanel
object that contains the pixel information tool.
477
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 4 Issue: 3 474 – 479
_______________________________________________________________________________________________
hpanel = impixelinfo(...) returns a handle to the Pixel The sendmail function does not support HTML-formatted
Information tool uipanel. messages. However, you can send HTML files as
attachments.
3.SYNCING THE FAKE CURRENCY If sendmail cannot determine your email address or
DATABASE: outgoing SMTP mail server from your system registry,
The Update Database of Fake currency database are store specify those settings using the setpref function. For
in Cloud Storage (Google Drive). Which are maintained by example:
the authorized organization. We are downloading the setpref('Internet','SMTP_Server','my_server.example.com');
Updated database by Clicking on Sync button in our project
setpref('Internet','E_mail','[email protected]');
So the current database is updated.
Function used For Downloading the Database : To identify the SMTP server for the call to setpref, check
urlwrite :- the preferences for your electronic mail application, or
urlwrite(URL,filename) consult your email system administrator. If you cannot
urlwrite(URL,filename,Name,Value) easily determine the server name, try 'mail', which is a
A. Description common default, such as:
urlwrite(URL,filename) reads web content at the setpref('Internet','SMTP_Server','mail');
specified URL and saves it to the file specified by filename
[filestr,status] = urlwrite(___) stores the file path in By default, the sendmail function does not support email
variable filestr, and suppresses the display of error servers that require authentication. To support these servers,
messages, using any of the input arguments in the previous change your system settings and set preferences for the
syntaxes. When the operation is successful, status is 1. SMTP user name and password, with commands in the
Otherwise, status is 0. following form:
props = java.lang.System.getProperties;
4. ALARM GENERATION: props.setProperty('mail.smtp.auth','true');
If the Image is Fake we are alarm the User by Showing
the Alert Message on the Screen " Fake Image Detected" setpref('Internet','SMTP_Username','myaddress@exa
and Sending the message to the authorized mail Id. mple.com');
For Connecting to the mail and Sending the Alert message setpref('Internet','SMTP_Password','mypassword');
we are using the
sendmail Function : To override the default character encoding, set the
Send email message to address list preference for email character encoding as follows:
sendmail(recipients,subject) setpref('Internet','E_mail_Charset',encoding);
sendmail(recipients,subject,message)
sendmail(recipients,subject,message,attachments) where encoding is a character vector specifying the
sendmail(recipients,subject) sends email to recipients with character encoding, such as 'SJIS'.
the specified subject. For a single recipient,
specify recipients as a character vector or a string. For RESULT
multiple recipients, specify recipients as a cell array of Mapping of features is been done and detection of
character vectors or a string array. subject must be a currency is been done. If fake note is detected then alarm
character vector or string. will be generated and mail will be send to the respected
authority.
sendmail(recipients,subject,message) includes the
specified message. If message is a character vector or a
string, sendmail automatically wraps text at 75 characters.
To force a line break in the message text,use 10, as shown in
the Examples. If message is a cell array of character vectors
or a string array, then each element represents a new line of
text.

sendmail(recipients,subject,message,attachments) attaches
the files listed in the attachments input
argument. attachments can be a character vector, cell array
of character vectors, or string array.
Figure.Real Note Detected
Point which we have to consider before sending the mail.
478
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 4 Issue: 3 474 – 479
_______________________________________________________________________________________________
[8]. Rubeena Mirza, Vinti Nanda 2012 IFRSA Int.J.
Computing 2 375-80
[9]. JunfangGuo, Yanyun Zhao and AnniCai 2010 Proc IEEE
Int. Conf Network Infrastructure and Digital
Content 359-363.
[10]. Deborah M, Soniya C and Prathap 2014Int J
InnovSciEngg& Tech1 151-57.

Figure. Sending of mail to the authorised person.

CONCLUSION
The counterfeit currency detection using image
processing was implemented on MATLAB. Features of
currency note like serial number, security thread,
Identification mark, Mahatma Gandhi portrait, Angular
bleed lines were extracted. The process starts from image
acquisition to calculation of intensity of each extracted
feature. The system is capable of extracting features even if
the note has scribbles on it. The algorithm processed here
works suitably for the newly introduced 500 and 2000
denomination. Hardware implementation of the proposed
system can also be done using suitable processor so that to
increase the speed of detection.

REFERENCES

[1]. Snigdha Kamal, Simarpreet Singh Chawla, Nidhi Goel,

Balasubramanian Raman (2015) “Feature Extraction and
Identification of Indian Currency Notes” pp. 978-1 2015
IEEE.
[2]. Sahana Murthy, JayantaKurumathur, B
RojaReddy(2016)“Design and Implementation of Paper
Currency Recognition with Counterfeit Detection” 2016
Online International Conference On Green Engineering
and Technologies(IC-GET)IEEE.
[3]. Li Liu, Yue Lu, Senior Member, IEEE, and Ching y.
Suen, Life Fellow, IEEE (May 2017) “An Image-Based
Approach to Detection of Fake Coins” pp. 1556-6013
2017 IEEE.
[4]. Ssbinod Prasad Yadav, C.S. Patil, R.R. Karhe, P.H.
Patil(July 2014)“An Automatic Recognition of Fake
Indian Paper Currency Note Using MATLAB”
International Journal of Engineering Science and
Innovative Technology(IJESIT) Volume 3, Issue 4, July
2014
[5]. TruptiPathrabe G and SwapniliKarmore 2011 Int. J.
CompTrendsTech 152-156
[6]. Tanaka M, Takeda F, Ohkouchi K and Michiyuk 1998
IEEE Tran on Neural Network 1748-
[7]. Jahangir N, Ahsan Raja Chowdhury 2007 IEEE 10th Int.
Conf. on Computer and Information Technology 1-5.

479
IJFRCSCE | March 2018, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________