International Journal of Electronics Communication and Computer Technology (IJECCT)

Volume 2 Issue 3 ( May 2012)

ISSN:2249-7838 IJ ECCT | www.ijecct.org 101

Coin Recognition Using Circular Hough Transform


Dr. Neelu Jain
Assistant Professor, PEC University of Technology,
Chandigarh. India
[email protected]

Neha Jain
Assistant Professor, Rayat Bahra College of Engg. & Bio-
Technology for Women, Kharar, Mohali, India
[email protected]


Abstract-- This paper represents algorithm for recognition of the
coins of different denomination. The proposed system first uses a
canny edge detection to generate an edge map, then uses CHT
(Circular Hough transform) to recognize the coins and further
find the radii of them. Based on the radius of the coin, the coins
of different denomination are classified. The experimental result
shows that the Hough transform is an effective tool for coin
detection even in the presence of noise.

Keywords-Circular Hough Transform (CHT); circle detection;
coin recognition
I. INTRODUCTION
Coin Recognition is a difficult process in machine
intelligence and computer vision because of its various
rotations and widely changed input pattern. Coin detection and
recognition in noisy and cluttered images also poses a great
challenge. The goal of this research work is to develop a
system to classify the coins of different denominations by
finding their radii in real time despite the above mentioned
challenges. In this system, CHT (Circular Hough transform) [1,
2] is used to detect the presence of circular shapes like coins
from the input image because it has the robustness to deal with
the noises in the image. CHT is a kind of HT (Hough
transform) [3] that can extract circular objects from an image.
The Hough Transform was first introduced by Paul Hough [4]
in 1962 to detect straight lines in bubble chamber data, the
transform consists of parametric description of a feature at any
given location in the original images space. The HT
essentially consists of two stages. In the first stage, edge map
of the image is calculated then each edge point contributes a
circle to an output accumulator space. In the second stage, the
output accumulator space has a peak where these contributed
circles overlap at the center of the original circle and then
define the coordinates of the circle.
The CHT has been used in several researches in detecting
iris and pupil boundaries for face recognition [5], fingertips
position detection [6] and automatic ball recognition.
The main advantage of using HT is high reliability and it
gives ideal result even in the presence of noises. Also the HT
provides parameters to reduce the search time for finding
objects based on a set of edge points.
In spite of its advantages, the HT has some disadvantages
when it deals with large size image.
When the large size image is used, the quantity of data has
become large and processing will be slow. The large amount of
storage and high cost of computation are two major drawbacks
of using HT in real time applications.
To tackle these problems, many modifications to the CHT
have been widely implemented to either increase the detection
rate of the coins or more commonly to reduce its computational
complexity [7]. In this proposed work, the edge orientation
information is used to increase the CHT performance. This
method was first suggested by kimme et.al. [8] noting that the
orientation of an edge point on the boundary of a circle is in the
direction of its centre. This modification reduces computational
requirements by plotting arcs in the accumulator space.
II. PROPOSED SYSTEM
The set-up includes a low end digital video camera and a
computer. The camera having the resolution of 320 240
pixels has been used throughout this study. The camera is
placed at the distance of 20cm from the coin so that the image
of the coin can be captured easily. Fig. 2 shows the image
captured by the camera.
However, due to the positioning of the camera or that of the
coin, the coin image is at times slightly distorted or only a
portion of the coin is visible, but nevertheless it still resembles
to a circle. CHT has the capability of identifying the coin even
when only portion of the coin is visible. Fig. 1 shows the block
diagram of the detection and recognition system.






Figure 1. Coin recognition process
Various preprocessing techniques have been used to make
the input image suitable for the algorithm. In order to reduce
Input
Pre-
processing
Edge-
detection
Input
Pre-
processing
Edge-
detection
Result Circular
Hough
transform
International Journal of Electronics Communication and Computer Technology (IJECCT)
Volume 2 Issue 3 ( May 2012)


ISSN:2249-7838 IJ ECCT | www.ijecct.org 102

the noise from the image, Gaussian filter has been used to
smooth the image, and then an appropriate gray level
thresholding is done to obtain a binary image.
A. Edge Detection
Hough transform is based on feature points extracted from
the original image and usually, edges are used as the feature
points. Various edge detection methods have been used for
different applications.
If Sobel filter [9] is used to a coin image, large number of
edge points are obtained from texture of the coin can be
regarded as noise, which will induce a huge overhead in the
execution time of the Hough transform and most importantly
will produce measurement errors, so technique to reduce the
unwanted edge is sought. Result of applying Sobel filter to an
image is shown in Fig. 3.
The canny edge detector [10] is very powerful tool for
detecting edges in a noisy environment. Canny edge detector
can remove most of the edge points. Canny gives thin edge
compared to the Sobel. Hence, canny edge detector has used
for eliminating the unwanted edges that can result from Sobel.
Based on the smoothed image, derivatives in both the x and y
direction are computed, these in turn are used to compute the
gradient magnitude of the image.
Once the gradient magnitude of the image has been
computed, a process called non maximum suppression is
performed; in which pixels are suppressed if they do not
constitute a local maximum.



Figure 2. Sample image (Coins.jpg)



Figure 3. Result of Sobel Filter




Figure 4. Result of Canny Edge detector
The final step in the canny edge detector is to use hysteresis
operator, in which pixels are marked as either edges, non edges
and in-between, this is based on threshold values. The next step
is to consider each of the pixels that are in-between, if they are
connected to edge pixels these are marked as edge pixels as
well.
The result of this edge detector is a binary image in which
the white pixels closely approximate the true edges of the
original image as shown in Fig. 4.
B. Circular Hough Transfors
The Hough transform and several modified versions have
been recognized as robust techniques for curve detection.
Hough transform for detecting circles is based on the same
principle introduced by the Hough transform for line segment.
The CHT was sketched by Duda et.al [11]. The CHT aims to
find circular patterns within an image. The CHT is used to
transform feature points in the image space into accumulated
votes in the parameter space, or the Hough space. Then, for
International Journal of Electronics Communication and Computer Technology (IJECCT)
Volume 2 Issue 3 ( May 2012)


ISSN:2249-7838 IJ ECCT | www.ijecct.org 103

each feature point, votes are accumulated in an accumulator
array for all parameter combinations. The array elements that
contain the highest number of votes are used to indicate the
presence of the pattern.
A circle pattern is described by equation given as
(x-a)
2
+ (y-b)
2
= r
2
(1)

where a and b are the coordinates of the centre in the x and
y direction respectively and r is the radius of the circle. The
parametric representation of the circle is given as

x = a + rcos (2)

y = b + rsin (3)

As a circle is defined by the three parameters:-centre
coordinates (a, b) and radius (r), the Hough space is a three
dimensional space, with the Z-axis representing the radius.
In order to perform a Hough transform, all possible values
for the radius can be considered. However, in the proposed
system, CHT is used for detecting the coins of different
denominations, so a suitable range for radius of the coins can
be defined.
An example of circular Hough transform is shown in Fig. 5


(a) (b)

Figure 5. (a) The contribution of the edge points to the accumulated space
(b) Edge point contribution to a single accumulator point.

A set of edge points, within the original image, are
indicated by the black circle. Each edge point contributes a
circle of radius R to an output accumulator space indicated by
grey circles. The output accumulator space has a peak where
these contributed circles overlap at the centre of the original
circle. The location of each peak gives the parameters of each
detected circle.
III. EXPERIMENTAL RESULT
Various images have been used to evaluate the performance
of the proposed system. The success rate of the proposed
system to recognize the coins of different denomination is
94%. Table1 shows the radius of 3 coins of different
denomination.
TABLE I. COINS WITH THEIR RADII
Denomination
(Rs)
Radius[pixels]
2 Rs Coin
1 Re Coin
5 Rs Coin
26
24
22




Figure 6. Image after computing radius of coins
IV. CONCLUSION
The coin of different denomination can be recognized based
on the radius of the coin. When coins are placed on a plain
surface, the edges of the coins are easily traced and the HT has
a much higher chance of success. By using the canny edge
detector in conjunction with CHT, the coordinates of the coin
are evaluated.
The proposed system can be used in coin operated
payphones, vending machines, weighing machines and in many
other machines that are based on coin recognition.
The problem arises if the coin image is captured from a
distance and the image tends to be small. Besides that, some of
the coins are overlapped. These restrictions make the detection
process difficult.
Future work may include detection of several coins that are
overlapped with each other.
REFERENCES
[1] Zhang Mingzhu and Cao Huanrong, A New Method of Circles Center
and Radius detection in Image Processing, Proceeding of the IEEE
International Conference on Automation and Logistics, Qingdao, China,
pp:2239-2242, 2008.
International Journal of Electronics Communication and Computer Technology (IJECCT)
Volume 2 Issue 3 ( May 2012)


ISSN:2249-7838 IJ ECCT | www.ijecct.org 104

[2] Zhang Xiao, Peng Weij, Detection of circle based on Hough transform.
Transducer and Micro System Technologies, Vol. 8 pp: 25-34, 2006.
[3] Lam, W.C.Y., Yuen, S.Y, Efficient technique for circle detection using
hypothesis filtering and Hough transform, IEEE Proceedings of
Vision, Image and Signal Processing, Vol. 143, Oct.1996.
[4] Hough P.V.C, Method and means of recognizing complex patterns,
US patent 3069654, Dec 1962.
[5] Moukhtar E. et.al. Rejection Analysis of an Iris Recognition System,
pp: 19-21, Dec. 2005.
[6] Yu Tong Hui Wang, Daoying Pi and Qili Zhang, Fast Algorithm of
Hough transform-Based Approaches for fingerprint Matching, The
Sixth World Congree on Intelligent Contral and Automation, Vol. 2,
pp: 10425-10429, June 2006.
[7] Xu.L, Oja.E, and Kultanan.P, A new curve detection method
randomized Hough transform (RHT), Pattern Recognition Letter,
Vol.11, pp: 331338, 1990.
[8] Kimme, C. D. Ballard and J. Sklansky, Finding circles by an array of
accumulators, Proc. ACM, Vol. 18, pp: 120-122, 1975.
[9] Nick Kanopoulos, Nagesh Vasanthavada and Robert L. Baker Design
of an Image Edge Detection Filter Using the Sobel Operator, IEEE
Journal of solid state circuits, Vol. 23, No. 2, April 1988.
[10] J.Canny, A Computational Approach to EdgeDetection, IEEE
Transactions on Pattern Analysis and Machine Intelligence, Vol.8, No.6,
Nov.1986.
[11] Duda, R.O. and P.E Hart, Use of the Hough transformation to detect
lines and curves in picture, Commum. ACM, pp: 11-15, 1972.