A Case Study : Circle Detection Using Circular Hough Transform

International Journal of Scientific Research and Engineering Development-– Volume 2 Issue 5, Sep – Oct 2019
Available at www.ijsred.com
ISSN : 2581-7175 ©IJSRED: All Rights are Reserved Page 1070
A Case Study: Circle Detection Using Circular Hough
Transform
Utkarsh Biranje, Suraj Suryavanshi, Rahul Khatkar, Asif Khalifa, Shivani Joshi
Department of Computer Science and Engineering,
Dr. J. J. Magdum College Of Engineering ,Jaysingpur, India.
Prof.Ms.A.B.Shikalgar (Project Guide)
Department of Computer Science and Engineering,
Dr. J. J. Magdum College Of Engineering Jaysingpur, India.
Abstract – Circle detection is an important aspect of image processing. The Hough transform is one of the methods to find
circles in an object especially the objects used in various industrial applications. This paper presents a modified method based
on the basic CHT algorithm and using relatively less calculations in order to improve the computational performance for a good
accuracy of circle detection in a binary image. The purpose of the technique is to find circles in imperfect image inputs. As the
results of the experiments, the object detection rate of the proposed system was 95% in accuracy.
Keywords – Circular Hough Transform, edge detection, circular shape recognition, Computation Time.
I. INTRODUCTION
Paul Hough suggested an efficient method for detecting
lines and edges in binary images in 1962 [1]. In everyday
experience circular features are commonly sought in
digital image processing. The problem of detecting circular
features is very important for image analysis, in
particular for industrial applications such as automatic
inspection of manufactured products and components,
aided vectorisation of drawings, target detection,
mechanical parts[2], and particle trajectories[3,4] etc.
Object detection and recognition in noisy and cluttered
images is challenging problem in computer vision. The
goal of this research is to identify objects in an image (here
we consider coins). There are several problems in
detecting and recognizing the coins in the image. First, the
target object is obscured due to the presence of the other
object which can interfere with recognition process.
Second, some of the objects are overlapping between each
other that make the recognition process challenging. Third,
the various object positions and finally, the images itself
contain noise that make the recognition process difficult
without proper preprocessing and segmentation process.
This paper gives basic introduction to circular hough
transform with giving some experimental results. Edge
detection Algorithm is also introduced.
The paper is organized as follows. Section 2 introduced
related works performed by researchers in this domain.
Section 3 includes the concept of the proposed method. In
section 4 proposed the algorithm. Section 5 shown the
results and section 6 conclude the paper.
II. RELATED WORK
The detection of circular and elliptic shapes is a common
task in computer vision and image recognition. Some
methods rely on converting gray-scale images to binary
ones using edge detection techniques and calculating
numerical shape descriptors. Simple shape descriptors are
sorted out in [8]. One of them, known as elliptic variance,
is especially useful for detecting ellipses. Rosin proposed
other descriptors (moment invariants, Euclidean distances)
that can be adapted to measure the ellipticity of shapes [5].
The first to present a direct method for fitting ellipses
to a set of points in the least-squares sense. Their method
is exploited in the segmentation algorithm presented in this
work. Previous methods used a generic conic fitting or an
iterative approach to recover elliptic solutions. A variety of
‘fit-to-data’ functions were discussed [6]. Low level edge
detection operators do not guarantee continuous
boundaries of objects. This makes many image analysis
tasks difficult, especially for noisy images. The aim of
contour grouping algorithms is to connect edges that are
supposed to be sub-parts of the same object. Contour
grouping techniques were concentrated mainly on
detecting salient curves [7].
Their basic idea is to reshape an initial curve provided
by any rough segmentation technique, subject to
constraints from a given image. Initially, the current curve
is set to the initial one. [2] Showed that active contours can
also be employed for tracking moving objects. The energy
function calculates the difference between object features
and background features, so it is useful for stained smears.
RESEARCH ARTICLE OPEN ACCESS

III. CIRCULAR HOUGH TRANSFORM
The construction of real circle detection algorithm is one
of the most challenging tasks because images are
sensitive to noise and other complexities involved in
processing. Various algorithms were proposed in this
domain but Circular Hough Transform (CHT) has long
been recognized as robust techniques for circle
detection. The circle is actually simpler to represent in
parameter space. The circle equation is
r
n
= (x−a)2
2
+ (y−b)2
2
(1)
As it can be seen the circle got three parameters a, b and r.
Centre of the circle in x and y direction is given by a and
b, r is the radius. Circle parametric representation is given
by
x = a + r cos (θ) (2)
y = b + r sin (θ) (3)
Thus the parametric space for a circle will belong
Thus the parametric space for a circle will belong to R. As
the number of parameters needed to describe the
shape increases as well as the dimensions of the parameter
space R increases so do the complexity of the Hough
Transform. So with the Circle Hough Transform, we
expect to find triplets of (x, y, R) that are highly
probably circles in the image. First find out the edges
using Canny edge detection algorithm. For each edge
point, draw a circle assuming that edge point as centre
with the desired radius. The circle is drawn
in the parameter space (fig 1).
The ‘a’ value is the x-axis, ‘b’ value is the y-axis while
the z-axis is the radii in the parameter space of the circle
drawn. At the coordinates which belong to the
perimeter of the drawn circle we increment the value in
our accumulator matrix which essentially has the same
size as the parameter space. In this way for each edge
point in the input image drawing circles with the
desired radii and incrementing the values in our
accumulator. When completed with every edge point and
every desired radius, turn attention to the accumulator. The
numbers in the accumulator will represent the number of
circles passing through the individual coordinates. Thus
the highest numbers are taken into consideration.
Fig. Circular Hough Transform concept
Implementation :
Implementing Circular Hough Transformation on coins.
We get the following results:
Firstly the original image is taken into consideration.
Various coins are placed alongside as shown in Fig2.
The image is now inserted as input for cicular hough
transformation. Circular Hough transformation will use
Sobel or Canny edge algorithm to detect edges in the
image.
The coins in the image are detected by Sobel detecting
Algorithm in Circular Hough Transform as shown in Fig3.
Fig2 .Original image of coins
Fig3.Circles detected using
Edge Detection
Fig4. Parametric Space
Representation of Circle
IV. CIRCLE EDGE DETECTION
Syntax:
circles = houghcircle(im, minR, maxR);
or
circles = houghcircle(im, minR, maxR, ratio);
or
circles = houghcircle(im, minR, maxR, ratio, delta);

Input:
im: input image (gray or color)
minR: minimal radius (unit: pixels)
maxR: maximal radius (unit: pixels)
ratio: minimal number of detected edge pixels of a circle to
the circle perimeter (0<ratio)
Output:
circles:
n-by-4 array of n circles;
each circle is represented by [cx cy R c],
where (cx cy), R, and c are the center coordinate, radius,
and pixel count, respectively.
Steps:
1.First of all we check the input parameter.
2. If the number of input is ‘4’, in that case , delta=12,
because Each element in [cx cy R] may deviate 4 pixels
approx.
3. If the number of input is ‘3’, then ratio=0.3 ,which is the
1/3 of the perimeter and delta=12.
4. If the number of input is not ‘5’,at that time program
will show a message that” Require at least 3 input
arguments”
5. If minR<0 or maxR<0 or minR>maxR or ratio<0 or
ratio>1 or delta<0, the it will show this message
'Required: 0<minR, 0<maxR, minR<=maxR, 0<ratio<1,
and 0<delta'
The shape recognition of an object is possible through
sobel edge detection algorithm that gives accuracy to find
edges of the object in given image.
The computation time is dependent upon the factors such
as the type of Edge Detection Algorithm and the
Dimension of input image.
V. EXPERIMENTAL RESULTS
Fig5.Final result given by Circular
Hough transform
The highest success rate of the proposed system to detect
the object was 95%.
Experiment
number
Image
dimension
True
Circles
Results
of Edge
Detection
CHT
Results
1 250 x 180 10 14 10
From fig.4 The edge detection algorithm detected around
14 circles. Then after passing these parameters the circular
Hough transform gives the result as 10.
VI. CONCLUSION
In this paper we have proposed Circular Hough
Transform and Edge Detection methodology. Circular
Hough Transformation algorithm can be used along
with the existing algorithms or concept can be applied
in developing new algorithm. The experimental results
indicated that our algorithm can be effectively applied
to detect the objects. Future research domain include the
areas such as algorithm optimization, reduction in
complexity of algorithm, accurate circle detection etc.
VII. REFERENCES
[1] P. V. C. Hough, Method and Means for Recognizing
Complex Patterns, US Patent 3,069,654, Ser. No. 17,
7156 Claims, 1962.
[2] U. M. Landau, “Estimation of a circular arc center and its
radius,” Comput. Vis. Graph. Image Process., no. 38,
pp.317–326, Jun. 1986.
[3] J. F. Crawford, “A noniterative method for fitting
curves”.
[4] V. Karimäki, “Effective circle fitting for particle
trajectories,” Nucl.Instrum. Meth. Phys. Res., no. 305,
pp. 187–191, 1991.
[5] Rosin P. (2000). Measuring shape: Ellipticity,
rectangularity, and triangularity, Proceedings of the
15-th International Conference on Pattern
Recognition, Barcelona, Spain, pp. 1952–1955.
[5] Rosin P. (2000). Measuring shape: Ellipticity,
rectangularity, and triangularity, Proceedings of the
15-th International Conference on Pattern
Recognition, Barcelona, Spain, pp. 1952–1955.
[6] Rosin P. (1996). Assessing error of fit functions for
ellipses, Graphical Models and Image Processing
58(5): 494–502.
[7] Shashua A. and Ullman S. (1991). Grouping
contours by iterated pairing network, Neural Info 3:
335–341.
[8] Peura M. and Iivarinen J. (1997). Efficiency of
simple shape descriptors, Aspects of Visual Form,
Proceedings of the 3- rd International Workshop on
Visual Forum, Capri, Italy, pp. 443–451.

A Case Study : Circle Detection Using Circular Hough Transform

Recommended

More Related Content

What's hot (19)

Similar to A Case Study : Circle Detection Using Circular Hough Transform (20)

More from IJSRED (20)

Recently uploaded (20)

A Case Study : Circle Detection Using Circular Hough Transform