International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org
Volume 3, Issue 3, March 2014
ISSN 2319 - 4847
Coin recognition system using ANN
Poonam sharma1,Naveen dhillon2,Kuldeep sharma3
1
student ECE deptt RIET,Phagwara,Punjab,INDIA
Assistant prof ECE deptt RIET,Phagwara,Punjab,INDIA
2&3
Abstract
Coins play very important role in our daily life. In grocery store, banks, buses, trains everywhere there is need of coins. So
coins can be sorted and counted automatically. For this it is necessary that coins can be recognized automatically. So ANN
(Artificial Neural Network) based Automated Coin Recognition System for the recognition of Indian Coins of denomination `1,
`2, `5 and `10 with rotation invariance is used. The objective of this paper is to classify recently released Indian coins of
different denomination. The objectiveis to recognize the coins and count the total value of the coin in terms of Indian National
Rupees (INR).
1. INTRODUCTION
Coins have been the integral part of our day to day life. So there is requirement of accurate and efficient automatic coin
recognition system.
There are three types of coin recognition systems available in the market based on different methods:
Mechanical method based systems
Electromagnetic method based systems
Image processing based systems
The mechanical method based systems use parameters like diameter or radius, thickness, weight and magnetism of the
coin to differentiate between the coins. But these parameters can not be used to differentiate between the different
materials of the coins.
The electromagnetic method based systems can differentiate between different materials because in these systems at
certain frequency ,the coins are passed through an oscillating coil and different materials bring different changes in the
amplitude and direction of frequency. So these changes and the other parameters like diameter, thickness, weight and
magnetism can be used to differentiate between coins. The electromagnetic method based coin recognition systems
improve the accuracy of recognition but still they can be fooled by some game coins.[6]
In image processing based systems image of the coin to be recognized is taken either by camera or by some scanning.
Then these images are processed by using various techniques of image processing like FFT , Gabor Wavelets ,DCT, edge
detection, segmentation, image subtraction , decision trees etc and various features are extracted from the images. then
coins are recognized by using these features.
2. REVIEW OF LITREATURE
Several coin recognition approaches are mentioned in the literature. In 1992 for coin recognition,Minoru Fukumi et al.
presented a rotational invariant neural pattern recognition system . 500 yen coin and 500 won coins were used for
experiments. In this work they have created a multilayered neural network and a preprocessor consisting of many slabs of
neurons to provide rotation invariance. In 1993, they further extended their work and tried to achieve 100% accuracy for
coins. In this work they have used BP (Back Propagation) and GA (Genetic Algorithm) to design neural network for coin
recognition. In 2006, Adnan Khashman et al. presented an Intelligent
Coin Identification System (ICIS), which uses neural
network and pattern averaging for recognizing rotated coins at various degrees. It shows 96.3% correct identification i.e.
77 out of 80 variably rotated coin images were correctly identified.
Mohamed Roushdy had used Generalized Hough Transform to detect coins in image. Zhang et al. uses genetic
programming for a number of object classification and detection problems. Typically, low-level pixel statistics are used to
form the terminal set, the four arithmetic operators are used to construct the function set, and the fitness functions are
based on either classification accuracy or error rate for object classification problems, and detection rate and false alarm
rate for object localization and detection problems. Results have been achieved on classification and detection of regular
objects against a relatively uncluttered background. Since the work to be presented in this paper focuses on the use of
genetic programming techniques for object recognition.
3. IMPLEMENTATION DETAILS:
Pattern Recognition :
statistical approach and structural approach are basic approaches in pattern recognition. In statistical approach, the
pattern is represented as a vector in a feature space. Then statistical concept is a decision algorithm, it decides to which
class the pattern belongs. In the structural approach, the pattern is represented by its structure. For example, a string of
Volume 3, Issue 3, March 2014
Page 288
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org
Volume 3, Issue 3, March 2014
ISSN 2319 - 4847
symbols, a graph connecting the primary elements, etc. The statistical method can be broadly classified into classical and
Artificial Neural Networks (ANN) approaches .[6]
The coin classification technique is based on computations like there should be proper lightning on coin.coins should
move on conveyor belt,proper separation should be there in between coins and fed to system for recognition,each coin
should be weighed accurately,coin should be collected from both sides.coin image can be rotated by any degree.coin
images with 256 gray values are to be computed.
Coin recognition system:
There are basically seven steps for coin recognition process:
1. Acquire RGB coin image.
2. Convert RGB image into Gray scale.
3. Remove shadow from image.
4. Crop and trim the image.
5. Generate pattern averaged image .
6. Generate feature vector and pass it as input to trained Neural network.
7. Give appropriate result according to output of neural network.
Pre-Processing :
The Zooming and de-zooming are the important processes by which a coin image is increased or decreased in size.
The zooming helps us to make the size of the coin image bigger, by which recognition rate is increased.[7]
Data Acquisition :
Ordinary Cartesian coordinate system is used
to represent a pixel of an image. In this system, g(x, y) is the gray level at the pixel (x, y). Images can alternatively be
thought of as ordinary matrices in which the gray level of a pixel is represented as g1(i, j).
Figure 1. Pre processing image.
4. CONCLUSION
Due to large intra-class variance, the classification
of ancient coins is still a challenging task, especially if attempted from single 2D images.An ANN based automated coin
recognition system has been
developed using MATLAB. In this system, firstly preprocessing of the images is done and then these preprocessed images
are fed to the trained neural network
References:
[1] Fukumi M. and Omatu S. Designing A Neural NetworkFor Coin Recognition By A Genetic Algorithm. Proceedings
of 1993International Joint Conference on Neural Networks, pages 2109–2112, 1993.
[2] Khashman A., Sekeroglu B. and Dimililer K. Intelligent Coin Identification System. Proceedings of the IEEE
International Symposium on Intelligent Control, pages 1226–1230, 2006.
[3] Roushdy M. Detecting Coins with Different Radii based on Hough Transform in Noisy and Deformed Image. In the
proceedings of GVIP Journal, (1), 2007.
[4] Bischof H., Wildenauer H., and Leonardis A. Illumination insensitive eigenspaces. Proc.of International Conference
on Computer Vision, pages 233–238, 2001.
[5] Thumwarin. P, S.Malila, P.Janthawang, W.Pibulwej, and T.Matsura. A Robust Coin Recognition method with
rotation Invariance. IEEE, pages 520–523.
[6] modi ,s.,bawa., “automated coin recognition system using ANN . International journal of computer applications,
pages. 8887-0975,2011.
[7] kaur, p., “A Survey : coin recognition techniques using ANN”. International Journal For Technological Research In
Engineering Volume 1, Issue 4, pages.209-212,2013.
Volume 3, Issue 3, March 2014
Page 289