Review of Various Image Processing Techniques for Currency Note Authentication

© 2016, IJCERT All Rights Reserved Page | 119
International Journal of Computer Engineering In Research Trends
Volume 3, Issue 3, March-2016, pp. 119-122 ISSN (O): 2349-7084
Review of Various Image Processing
Techniques for Currency Note
Authentication
Trisha Chakraborty, Nikita Nalawade, Abhishri Manjre, Akanksha Sarawgi, Pranali P Chaudhari
Abstract:- In cash transactions, the biggest challenge faced is counterfeit notes. This problem is only expanding due to
the technology available and many fraud cases have been uncovered. Manual detection of counterfeit notes is time
consuming and inefficient and hence the need of automated counterfeit detection has raised. To tackle this problem, we
studied existing systems using Matlab, which used different methods to detect fake notes.
Keywords – Counterfeit, Image Processing, Matlab, Cloud, Map Reduce
——————————  ——————————
1. INTRODUCTION
Image processing is basically a technique in which an
image is converted into a digital form and there are
various operations performed on it in order to get an
enhanced image and to extract useful information
from it. The images are digitized with the help of a
scanner or by a video camera which is connected to
the frame grabber board in computers. This
digitization is done so that it can be stored in
computers memory or in a form of media storage like
CD-ROMs and hard disks. The image used as input in
this technique can be a video frame or photograph
and the output received may be an image or some
characteristics associated with that image. It is a type
of signal dispensation which treats an image as two-
dimensional signal by applying already set signal
processing methods. In today’s world, with rapid
growth of technology, image processing is being used
in various fields and is playing a very important role.
Image processing is done by following the below
three steps
 Importing the image with optical scanner or
digital photography.
 Image is then analyzed and manipulated for data
compression, image enhancement and pattern
recognition as they are not clearly visible to human
eyes.
 Output is the last stage in which result can be
altered image or report that is based on image
analysis.
Image Analysis is the extraction of meaningful
information from images; mainly from digital images
by means of digital image processing techniques.
Image analysis tasks can be as simple as reading bar-
coded tags or as sophisticated as identifying a person
from their face. Digital image analysis is when a
computer or electric device automatically studies an
image to obtain useful information from it. Apart from
a computer, the electric device may also be an electric
circuit, a digital camera or a mobile phone. Computer
Image Analysis largely contains the fields of computer
or ma-chine vision, and medical imaging, and this
field makes great use of techniques like digital
geometry, signal processing and pattern recognition.
The field of Image analysis was developed in the
1950s. It is the quantitative or qualitative
characterization of two-dimensional or three-
dimensional digital images. 2D images are, for
example, to be analyzed in computer vision, and 3D
images in medical imaging.
Available online at: www.ijcert.org

Trisha Chakraborty et al., International Journal of Computer Engineering In Research Trends
Volume 3, Issue 3, March-2016, pp. 119-122
In this fast-paced world with an ever expanding
market, there came a great need to make optimum use
of all the resources. With the massive storage of
images, there came the realization that images are an
indispensable source for the extraction of meaningful
information. It has been rightly said that “A picture is
worth a thousand words”. Image Processing is need
for Visualization to observe the objects that are not
visible, Image sharpening and Restoration to create a
better image, Image retrieval to seek for the image of
interest, Measurement of pattern that Measures
various objects in an image and Image Recognition to
distinguish the objects in an image. Analysis of images
can help us make strategic decisions in fields of
business and research. With the applications of big
data on the rise, it has been found that analyzing
digital images with the aid of computer techniques
helps to save 50% of investors’ time and reduce
human error.
2. LITERATURE SURVEY
We have seen a Grid Based Feature Extraction where
grids are applied to the images for processing them
further. The feature sets used in it are geometrical
shapes to recognize the denomination, governor
declaration and year of printing. There are various
algorithms used for finding out these feature set which
work as a part of the whole algorithm of processing.
Geometrical shapes for denomination are extracted by
feed forward neural network classifier, Governor
Declaration by region properties and year of print by
Optical Character Recognition technique. The
algorithm takes the images in RGB and then converted
in a binary image. A 3*3 grid is applied to the image
both on the top and back side of the image to create 18
blocks. Block 4 is selected for denomination of notes
where the neural network classifier extracts the shape
in it further classifying the amount of the note. Block 8
is selected for calculating the maximum and minimum
orientation in order to recognize the position of the
Governor declaration. Block 17 is selected for
recognition of year of printing, which segments the
characters and matches them with the standard
template using correlation technique. Correlation value
range from 0-1, if the vale is 0 is said to be a minimum
match and if 1 the maximum match. The experimental
setup is done on sample of 90 images, where 30 images
each are of rupees 100,500 and 1000.The average
success rate achieved is 92.30%.The best validation
performance of neural network classifier is 0.91181 at
epoch 59.The advantages of using these techniques
have been proved to be very efficient and accurate and
have overcome the shortcoming of other algorithms for
recognition of Indian currency.
In the second application, we have seen cloud
computing through android platform to send the
images to remote server. The feature sets used are
watermark of Mahatma Gandhi, security thread, see
through register, checking currency number series and
identification mark. The techniques used in this are the
watermark of Mahatma Gandhi has light and shade
effect and multidirectional lines in watermark window
which is matched with image in the database, the
security thread has the words “RBI” and “Bharat”
written on it which is checked with the database image,
the number series has 9 characters out of which first
three contain an a digit and two alphabets, the next 6
numbers are the serial numbers, this series is checked
by the server to check for the real results. Identification
mark is on the left side if the watermark window which
has different geometrical shapes for denominations of
notes, these shapes are checked with the database
image to give results. See through register has a floral
design on the left side between the vertical band both
on the front and back side, it has the denomination
printed in half on the front and the other half on the
back side, both of this is compared and checked by the
server. The experimental setup has the android phone
and the remote server. The android phone is of high
megapixel resolution which takes each of 3 images of
the feature sets mentioned and sent to the server
through cloud in jpg format. The server does the
processing, which converts the image into gray scale
and convolve with 3x3 Gaussian low-pass filters to
eliminate high frequency noise. Then canny edge
detection is applied to get the image in closed contours.
It then matches the feature sets with the standard
banknote images. If it matches appropriately, it results
into a real one. The advantages of this techniques are, it
is user friendly, available on the internet free of cost
and is time saving. The limitations though are the
android phone should be of good picture quality, the

images captured should be very clear and internet is
required throughout the process. Using this technique
can help common man in counterfeit should be very
clear and internet is required throughout the process.
Using this technique can help common man in
counterfeit of currency.
Manual testing of currency notes in transactions is an
untidy process as well as very time consuming which
may even involve tearing of notes while handling them.
In this paper, fake currency note detection technique
using MATLAB and feature extraction with HSV color
space is developed. The camera pictures the notes are
analyzed by MATLAB program installed on computer.
The project checks Indian currency notes of 100, 500,
1000 rupees. Methods to detect fake notes: See through
register, Water marking, optically variable ink,
Fluorescence, Security thread, Intaglio printing, latent
image, Micro lettering, Identification mark. The
experiment setup consisted of Operating system
interfaced with input device camera and output device
LCD display. First the Image Acquisition on which gray
scale conversion, edge detection, image segmentation,
Characteristic extraction, Comparison and then the
output is obtained. The low cost system, using effective
and efficient image processing techniques and
algorithms, provide accurate and reliable results at good
throughput as shown by experimental results. The
developed algorithm works for Indian denomination
100, 500, 1000.
Fake currency detection is a process of finding the
forgery currency. In this paper after we chose the image
apply pre-processing. In pre-processing the image is to
be cropped, smoothened and adjusted, and then the
image is converted into gray color. After the conversion,
image segmentation is done. The features are extracting
and reduce. Finally compare the image into original or
forgery. As a subcategory or field of digital signal
processing, digital image processing has many
advantages over analog image processing. It allows a
much wider range of algorithms to be applied to the
input data and can avoid problems such as the build-up
of noise and signal distortion during processing. With
image enhancement noise can effectively be removed by
sacrificing some resolution, but this is not acceptable in
many applications. In a Fluorescence Microscope
resolution in the z-direction is bad as it is. More
advanced image processing techniques must be applied
to recover the object. A Fake currency detection using
image processing and other standard methods by using
various methods like watermarking, latent image,
security thread and identification mark. By using
layman method the fake note is detected. This paper
presents the design and implementation of Indian paper
currency authentication system based on feature
extraction by edge based segmentation using Sobel
operator. The image is taken as Input on which pre-
processing, gray scale conversion, edge detection, image
segmentation, comparison and output is obtained.
Modules used in this paper are image acquisition, pre-
processing, gray scale conversion, edge detection, image
segmentation and feature extraction. Performance
Metrics taken into consideration are Peak Signal-to-
Noise Ratio (PSNR), Mean Squared Error Rate (MSE),
SSIM, Root Mean Squared Error Rate (MSE).PSNR
metric gave the highest accuracy compared to the other
performance metrics.
To detect the authenticity of currency note there are two
methods that is first line inspection method and second
line inspection method. First line inspection method
includes varied density watermarks, ultraviolet
fluorescence, intaglio printing, micro text and hologram
while the second line inspection methods include
isocheck/isogram, fiber based certificates of authenticity,
color and feature analysis. First line inspection method
is used for on-spot authentication of notes but is easier
to counterfeit than the second line inspection method
while the second line inspection method requires an
extra device to perform the inspection. The different
steps in currency recognition are image acquisition the
process of acquiring image from a currency note by
using digital camera on which image pre-processing is
done to enhance some image features important for
further processing and analysis. Further edge detection
for changing the image brightness, binarisation for
converting the image into binary image and feature
extraction takes place. The important features of an
image taken into consideration are size or area, aspect
ratio, color, Euler number, texture and matching
algorithm. The techniques for currency recognition used

are texture based, pattern based, color based and
currency localization techniques. Selection of feature
and selection of tools to recognize that particular feature
are critical. To overcome this limitation and building an
interactive system with high speed, accuracy and cheap
is a new challenge.
In this paper, we are designing a system that helps in
identification of Indian currency notes and to check
whether it is a valid or invalid. This is to differentiate
between the counterfeit notes and genuine notes.
Currency features such as See Through register, See
Through register symbol or Identification mark, Security
thread, Governor’s signature, Micro lettering, year of
print. These features are segmented using 3x3 grids.
This is done using SIFT technique which helps in
efficient matching of the features. The technique used in
this project is extracting unique features of the Indian
currency note using grid. Grid divides the currency into
nine parts on each side which reduces the time
complexity of the proposed model. Applying the pre-
processing to each block to recognize and extract
potential feature of Indian currency note. Grids 2, 6 and
15 are cropped and potential features of currency notes
are extracted. This paper we are extracting the features
by using SIFT algorithm. Based on the features it is gives
the result as genuine are fake note and it also identifies
the denomination of the currency note. The proposed
method is experimented on the dataset and the result is
obtained.
3. CONCLUSION
After studying the existing systems and different
algorithms and techniques used for processing and
analysis for identifying counterfeit in notes, we realized
that the different shortcomings in them can be overcome
using map reduce platform. The distributed platform
provided by existing systems like Matlab does not
provide an open source platform and for Cloud, we
need to provide internet connectivity throughout the
process .But since Hadoop is open source which
provides us with Map Reduce paradigm, these
shortcomings can be overcome. Also, efficiency of the
system can be improved using Map Reduce.
REFERENCES
[1] B. Sai Prasanthi and D. Rajesh Setty2,Indian Paper
Currency Authentication System using Image
Processing ,http://www.ijsret.org/pdf/121306.pdf
[2] Sowen Yoon, Altered Fingerprints: Analysis and
detection, IEEE Transactions on Pattern Analysis and
Machine Intelligence, VOL 34
http://www.cse.msu.edu/groups/biometrics/Fingerprints
/PAMI11.pdf
[3] Kanayama, K; Development of vehicle-license
number recognition system using real-time image
processing and its application to travel-time
measurement, Gateway to the Future Technology in
Motion,41stIEEE
[4] Steven Hurley; James C Wang; Reference
Architecture for Hadoop: Big Insights Reference
Architecture, IBM Corp 2013, 14.
[5] Jerey Dean and Sanjay Ghemawat; Map Reduce:
Simplified data processing on large clusters, Proceeding
OSDI04 Proceedings of the 6th Conference on
Symposium on Operating Systems Design and
Implementation-Volume 6, 2004.
[6] H. Hassanpour, A Yaseri and Ardeshiri; Feature
Extraction for Paper Currency Recognition, Signal
Processing and It’s Application P NO.1 4, ISBN.978-1-
4244-0778-1.
[7] Trupti Pathrabe, Swapnil Karmore, 2011; A Novel
Approach of Embedded System for Indian Paper
Currency Recognition, International Journal of
Computer Trends and Technology, ISSN.2231-2308.
8] Khushboo Khurana, Reetu Awasti; Techniques for
Object Recognition in Images and Multi-Object
Detection ,International Journal of advanced Research in
Computer Engineering Technology.
[9] Dong Ping Tian, A Review on Image Feature
Extraction, Representation Techniques ,International
Journal of Multimedia and Ubiquitous Engineering Vol
8,No 4,July,2013.http://www.sersc.org/journals/IJMUE
[10] X Munoz, J. Freixenet, X Cu, J.Marti ,Segmentation
of Images, Strategies for image segmentation combining
region and boundary information, Pattern Recognition
Letters 24(2003) http://www.cse.iitm.ac.in/
sdas/vplab/courses/lect-Segmen.pdf27

Review of Various Image Processing Techniques for Currency Note Authentication

More Related Content

What's hot (19)

Viewers also liked (8)

Similar to Review of Various Image Processing Techniques for Currency Note Authentication (20)

More from IJCERT (20)

Recently uploaded (20)

Review of Various Image Processing Techniques for Currency Note Authentication