IRJET- Survey based on Detection of Optic Disc in Retinal Images using Segmentation based Techniques

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 06 | June 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 795
Survey Based on Detection of Optic Disc in Retinal Images Using
Segmentation Based Techniques
D Binny Jeba Durai1, Dr. T. jaya2
1Udhya Polytechnic College, Vellamadam.
2CSI Institute of Technology, Thovali.
----------------------------------------------------------------------***---------------------------------------------------------------------
Abstract— Automated localization and detection of the
optic disc (OD) is an essential step in the analysis of digital
diabetic retinopathy systems. Accurate localization and
detection of optic disc boundary is very useful in
proliferative diabetic retinopathy where fragile vessels
develop in the retina. Detection of optic disc area is
complex because it is located in an area that is considered
as pathological blood vessels. The optic disc appears as a
round region usually brighter than the surrounding. The
results achieved by different algorithms can be compared
when algorithms are applied on the same standard
databases. In this paper we compare the eccentricity,
accuracy and brightness using different segmentation
techniques.
Keywords — optic disc, segmentation, fundus image,
detection
I. INTRODUCTION
Diabetic Retinopathy (DR) is a result of long term
diabetes mellitus and is a significant growing public
health problem. It is one of the predominant causes of
blindness. It causes pathological changes of the retina
such as microaneurysms, intraretinal microvascular
abnormalities, venous bleeding and neovascularities as
well as haemorrhages, exudates and retinal oedema.
Regular screening of Diabetic Retinopathy is
indispensable so that appropriate and timely treatment
can be given which thereby reduces the incidence of
impaired vision and blindness from this condition.
Current methods of detection and assessment of diabetic
retinopathy are manual, expensive and require highly
trained personnel to read large number of fundus
images. The efficiency can be improved by automating
the initial task of analyzing the huge amount of retinal
fundus images.
The optic disc is the brightest part in fundus images
that can be seen as a pale, round or slightly oval disk. It is
the entrance region of blood vessels and also acts as a
landmark and reference for the other features in the
fundus image. There are several methods for optic disc
detection.
Fig 1. Original Retinal Fundus Image
II. LITERATURE REVIEW
Frank ter Haar [6] applied illumination equalization
to the green-band of the image, and then a resolution
pyramid using a simple Haar-based discrete wavelet
transform was created. Finally, the brightest pixel at the
fifth level of the resolution pyramid was chosen to
correspond to the OD area. They also proposed an
alternative to the later method based on the pyramidal
decomposition of both the vasculature and the green-
band, where the fifth level of the resolution pyramid for
both the illumination equalized green-band and the
binary vessel segmentation are summed, and the
highest value corresponds to the OD center.
Hough transform [4] is a technique capable of finding
geometric shapes within an image and was employed to
detect the OD. The Hough transform uses a couple of
methods. In the first one, Hough transform was applied
only to pixels on or close to the retinal vasculature in a
binary image of the vasculature. The binary vasculature
was dilated in order to increase the possible OD
candidates. Alternatively, in the second method Hough
transform was applied once again but only to the
brightest 0.35% of the fuzzy convergence image
obtained. Once more, dilation was applied to the
convergence image to overcome the gaps created by
small vessels. The center and boundary of the optic disc
are found by applying the Hough transform to the
gradient image. The Hough transform is a method for
finding shapes in an image. The basic idea behind the
Hough transform is to transform the image into a
parameter space that is constructed specifically to
describe the desired shape analytically. Maxima in this
parameter space then correspond to the presence of the
desired shape in image space.
The circular hough transform (CHT) is used to detect
the OD which has a roughly circular shape. The retinal
vasculature in the green-band image was suppressed
using the closing morphological operator. The Sobel
operator and a simple threshold were then used to
extract the edges in the image. CHT [8] was finally
applied to the edge points, and the largest circle was
found consistently to correspond to the OD. This is
suitable for normal healthy fundus images where in
optic disc is alone the brightest region of the image. But
our

images contain exudates and lesions. The exudates’
intensity value is similar to that of opticdisc. If it is small
in size then it can be characterized by the intense pixels
density. But in some images the exudates are as large
and dense as exudates. In such cases it is difficult to find
which of the candidate contains the optic disc.
Fuzzy convergence [2] is a novel voting type algorithm
developed by Hoover and Goldbaum in order to determine
the origination of the retinal vasculature (convergence
point), and thus, localize the OD which is considered the
only consistently visible property of the OD. The inputs to
the fuzzy convergence algorithm were six binary vessel
segmentations (each at a different scale) obtained from the
green-band image. Each vessel was modeled by a fuzzy
segment, which contributes to a cu- mulative voting image
(a convergence image) where each pixel equals the amount
of fuzzy segments on which the pixel lied. Finally, the
convergence image was smoothed and thresholded to
determine the strongest point of convergence. If the
final result was inconclusive, the green-image was
illumination equalized, and Fisher’s linear discriminant
was applied to regions containing the brightest pixels to
detect the OD.
K-means clustering [9] is a vector quantization
method, which is initially from signal processing, that is
common for cluster analysis in data mining. The main
aim of K-means clustering is to divide n observations
into k clusters in which each observation fits to the
cluster with the nearest mean. Once the image is resized
and adaptive histogram is applied, clustering algorithm
is applied on image. Cluster analysis is a key technique
for reducing a mountain of rare data to a pile of
meaningful information. It shrinks data which creates
subgroups that are more manageable than individual
datum. Once K-mean clustering algorithm is applied on
the histogram image optic disc is detected in segmented
image. After detecting optic disc in segmented we apply
Mathematical Morphology (MM) on the segmented
image by using disc shape structuring element.
III. METHODOLOGY
The various methods adopted in optic disc detection
is discussed below.
A . Pre-processing
The original (RGB) image is transformed into
appropriate colour space for further processes. And
then, filtering technique is used to reduce the effect of
noise. After using the filtering technique, the noise such
as salt and pepper noise are removed from the image.
Then contrast-limited adaptive histogram equalization
(CLAHE) is used for image enhancement. Unlike
histogram, it operates on small data regions rather than
the entire image. This function uses a contrast-
enhancement method that work significantly better than
regular histogram equalization for most images.
B. Converting Colors from RGB to HSI
In digital image, the input image can be the RGB (Red,
Green, and Blue) images or other. The RGB image can be
described as M x N x 3 array of colour pixels. In this
paper, the RGB input image is transformed into HSI
colour space for further processes.
C. Optic Disc Segmentation
We perform the segmentation using the grey level
image Fodcr, which contains all the information
necessary. However, the high contrast of the blood vessel
inside the optic disc misguides the segmentation energy
functional and breaks the continuity of the optic disc
boundary.
Fig. 2. Retinal images with lesions and distorted shape
of optic disc
Therefore, we apply the morphological closing operation
to remove the blood vessels, and the vessel removed
image (Fvr) is obtained.
D. Threshold estimation
In normal, healthy retinal images, OD is mostly the
brightest region. However, in retinal images affected due
to diseases such as diabetic retinopathy, there may exist
other bright regions in addition to OD. So first we
detected all bright regions within retinal images. In
paper by Li and Chutatape [22], they used the highest
1% gray levels in intensity image to obtain threshold
value to detect candidate bright regions. However, there
is possibility of not detecting OD as candidate region if
highest 1% gray levels are occupied by other bright
regions within image.

Fundus Image
Pre- processing stage
Optic Disc Detection
Blood Vessels Detection
Fig 3. Flow Chart for Optic Detection
E. Image Enhancement
The result image of the median filter is enhanced by
using the histogram equalization technique. The
histogram equalization technique is used to overcome
the uneven-illumination case. There are two methods to
enhance the image: Histogram equalisation and
Adaptive histogram equalisation.
IV. BASIC ALGORITHM FOR DETECTION OF CANDIDATE
REGION FOR OD
Step 1: Estimate Threshold.
Step 2: Apply Threshold and identify bright regions.
Step 3: Select candidate regions which satisfy area
criterion.
Step 4: Select candidate region which satisfies density
criterion.
Step 5: If no candidate region is selected, reduce
threshold
Step 6: If threshold is greater than zero, apply steps 2
through 5.
Step 7: Stop.
However, it is probably for binary segmentation that
these operations are most useful. Therefore, in the
second stage, a global image threshold is used to convert
an intensity image to a binary image, separating brighter
regions from dark background. A global threshold
(level=graythresh (image_close)) that can be used to
convert an intensity image to a binary image with
im2bw(image_close, level) is a normalized intensity
value that lies in the range [0, 1]. The graythresh
function chooses the threshold to minimize the
intraclass variance of the black and white pixels.
RGB to HSI
Median Filter
Contrast limited
Histogram Equalization
Closing
Thresholding
Morphological
Operation
Binarization
Fig 4. Flow Chart for Detection of Candidate Region
V. RESULTS AND DISCUSSION
Fig 5. Detection Of Optic Disc Using K-Means Clustering,
Hough Transform And Fuzzy Convergence
We recorded the optic disc region in each of the
retinal images in this system by determining it
manually in order to evaluate the performance of our
study. We measured the accuracy of different
segmentation techniques with the information of
measured algorithms.

Fig 6: Comparison of Different Segmentation Techniques
Table 1: Comparison for Accuracy
Method Used OD Detected % Accuracy
K-means
clustering 126 96.92
Hough
transform 86 93.2
Fuzzy
convergence 40 95.2
VI. CONCLUSION
. During the research and the development of this
work, it was possible to better understand that the
input image is in RGB color space and for the further
processes the image is converted into appropriate color
space. The median filter is used for the noise reduction
without blurring the edge. And then various other
transform method were used. The results of our work
can play a major role in the screening of eye diseases.
ACKNOWLEDGMENT
We are thankful to God almighty and the supporting
contributors for the well being of our work.
REFERENCES
[1] Siddalingaswamy C , Gopalakrishna rabhu K
,“Automatic localization and boundary detection of
optic disc using implicit active contours”, Internation
Journal of Computer Applications, vol. 1, no. 7, pp. 1-
5, 2010.
[2] Michael D. Abramoff, Meindert Niemeijer, “The
automatic detection of the optic disc location in
retinal images using optic disc location regression”,
Conf roc IEEE Eng Med Biol Soc , 1: pp 44 2- 4435,
2006.
[3] Jaspreet Kaur, Dr. Sinha, “Automated localization of
optic disc and macula from fundus images”,
International Journal of Advanced Research in
Computer Science and Software Engineering, vol. 2,
Issue 4, pp. 242-249, April 2012.
[4] Healey PR, Mitchell P, Smith W, Wang JJ,
“Relationship between cup- disc ratio and optic disc
diameter”, the Blue Mountains Eye Study, Aust N Z J
Ophthalmol.; 25 Suppl 1:S99-101, May 1997.
[5] Rafael C. Gonzalez, Richard E. Woods, Steven
L.Eddins, Digital Image Processing Using MATLAB,
2nd Edition, Prentice-Hall, 2002 (Size 8) S. M.
Metev and V. P. Veiko, Laser Assisted
Microtechnology, 2nd ed., R. M. Osgood, Jr., Ed.
Berlin, Germany: Springer-Verlag, 1998.
[6] A. Dehghani, H. Moghaddam, and S. Moin, “Optic
disc localization in retinal images using histogram
matching,”EURASIP Journal on Image and Video
Processing, vol. 2, no.19, pp. 1687--5281, 2012.
[7] S. Sekhar, W. Nuaimy, and A. Nandi, “Automated
localization of retinal optic disc using Hough
transform,” IEEE International Symposium on
Biomedical Imaging: From Nano to Macro,2008.
[8] D. Godse and D. Bormane, “Automated localization
of optic disc in retinal images,” International
Journal of Advanced Computer Science and
Applications, vol. 4, no. 2, 2013.
[9] P. Healey, P. Mitchell, W. Smith, and J. Wang,
“Relationship between cup-disc ratio and optic disc
diameter,” Australian and New Zeland Journal of
Ophthalmology, 1997.
[10] G. Joshi, J. Sivaswamy, K. Karan, and S.
Krishnadas, “Optic disc and cup boundary detection
using regional information,” IEEE International
Symposium on Biomedical Imaging: From Nano to
Macro, 2010.
[11] S. Lavanya, “Detection of anatomical structures in
optical fundus images,” International Journal of
Application or Innovation in Engineering &
Managemen, vol. 2, no. 5, pp. 430--435, 2013.
[12] S. Sekhar, W. Al-Nuaimy, and A. K. Nandi,
"Automated localisation of retinal optic disk using
hough transform," in Proc.5th IEEE International
Symposium on Biomedical Imaging: From Nano to
Macro, 2008.
[13] A. Osareh, M. Mirmehdi, B. Thomas, and R.
Markham, "Colour morphology and snakes for optic
disc localisation," in Proc. 6th Medical Image
Understanding and Analysis Conference, 2002.
6
5
eccentricity
4 accuracy
3 brightness
2
1
0
fuzzy hough transform k means pyramidal
decomposition clustering decomposition

IRJET- Survey based on Detection of Optic Disc in Retinal Images using Segmentation based Techniques

More Related Content

What's hot (19)

Similar to IRJET- Survey based on Detection of Optic Disc in Retinal Images using Segmentation based Techniques (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Survey based on Detection of Optic Disc in Retinal Images using Segmentation based Techniques