Académique Documents
Professionnel Documents
Culture Documents
Abstract: This work presents different eye dieses like Diabetic processing techniques with artificial neural network. This
Retinopathy, Catract and Glaucona which causes our eye section of the paper describes eye anatomy along with a
damage. Among these three Diabetic Retinopathy is most brief description of “NPDR and PDR” effects on the eye.
perilous due to its different stages “(Mild NPDR – Non
Proliferative Diabetic Retinopathy Moderate NPDR and PDR
– Proliferative Diabetic Retinopathy)” and segregate it from a
normal eye with the study of fundus images. In this work, a
computer based approach have been proposed in which the
features are extracted from these images using image pre-
processing and furnish into the “SVM (Support Vector
Machine) using Sequential minimal optimization algorithm”.
I. INTRODUCTION
233
2015 International Conference on Computer and Computational Sciences (ICCCS)
III. IMAGE PRE-PROCESSING equalization. Through this, intensities are better distributed
on the histogram which is accomplished by effective
Image pre-processing deals with enhancing data images spreading out of most frequent intensity values.
prior to computational processing. It can significantly
increase the reliability and efficiency of proposed system.
Colour image
Image pre-processing
Image pre-processing for
for detection of
detection of blood vessels.
exudates.
1. Sum of on pixels.
Area of Exudates
2. Mean
234
2015 International Conference on Computer and Computational Sciences (ICCCS)
B. Detection of Exudates
For detection of exudates, image processing is carried out
The gradient is high at the borders and low inside the fundus
by first by checking the green channel, gray scale
images thus providing a somewhat distinction between
conversion, and conversion into binary through decided
blood vessels, haemorrhages and the other components of
threshold, morphological closing and finally removing
the image.
largest area (optic nerve) to get exudates.
ROI Processing: ROI (Region of interest) is a portion of an Figure12. (a) Green Channel Selection (b) Gray Scale Conversion
image that we want to filter. We define a ROI by defining a
binary mask i.e. our image which we want to process, with
pixel that define the ROI set to 1 and all the pixels set to 0.
We can declare more than one ROI for a single image.
Region can be geographic in nature.
235
2015 International Conference on Computer and Computational Sciences (ICCCS)
that kept important information about the image. It uses it becomes an important feature for efficient
delicacy in the image data, calculation of the new value is demarcation.
based on averaging of brightness values in some
neighborhood 0. Smallest component of a digital image is IV. SVM (SUPPORT VECTOR MACHINE) FOR
pixels [8]. Each pixels carries some intensity, mean to CLASSIFICATION OF HYPER PLANES
characterize the color of a small rectangular segment.
Debluring is an iterative process, you might need to repeat SVM (Support Vector Machine) is a better alternative for
the debluring process many times, varying the parameters Multilayer Preceptor Network [13]. SVM algorithm is also
you specify to the debluring function with each iteration [8]. used to calculate data and clustering and recognization of
patterns, used for classification and regression analysis. In
Morphological Closing: The optic disc which is bright has common it is used for linear-separation or classification of
similar features when compared to exudates, and thus can one class or category from another.
give false recognition; so it becomes necessary to remove it
[8]. In fundus images, the optic disc is marginally separated Sequential minimal optimization algorithm (SMO)[14]
with presence of blood cells at the extreme corner of the which break the problem down into two sub parts or we can
disc resulting in extremely close fragmented portions as say sub problems. There are so many advantages of using
seen in figure 11. In order to group them together dilation is SMO for SVM implementation.
performed, and then it is followed by erosion to separate the
optic disc from the exudates (close operator). The 1. Quadratic Programming (QP) SVM breaks into sub
structuring element used is ‘disc’ since it results in proper problems which we say as QP problems.
dilation of disjointed components of optic disc in a circular
fashion [9]. 2. Later on these QP problems solved analytically
Elimination of Largest Area and Detection of 3. Total amount of memory required for SMO is liner
Exudates:[1] Finally, using the region props each which permit SMO[15] to handle very huge training
connected component in the image is formulated with set
respect to their areas. The region with the maximum area
(optic disc) is found and then eliminated in order to get the 4. SMO is faster for linear SVM’s and sparse data set.
rest of the binary images just containing the exudates.
1) Feature Extraction
We calculate three different parameters of feature extraction
after all the above steps are followed:
Mean = (Sum of ON pixels) The training, validation and testing was carried out with
(Total no. of ON + Black pixels) following details:
No. of input on the basis of different category - 3
3. Sum of exudates: The total area of exudates present in Total no. of sample images -96
the image will also help us during classification of the No. of images for testing- 54
disease. With exudates prominently present in NPDR, Iteration for boundary testing -36
236
2015 International Conference on Computer and Computational Sciences (ICCCS)
V. RESULTS AND DISCUSSION [3] Eye Fundus Image Analysis for Automatic Detection of Diabetic
Retinopathy, Tomi Kauppi, Thesis for the degree of Doctor of
Science (Technology) presented at Student Union House at
Several features were extracted by image processing Lappeenranta University of Technology, Lappeenranta, Finland on
approach and insert as input to the classifier with analyzed the 12th of December, 2010, pp 25-29.
results. The results of SVM classification are “shown in [4] Review of automated diagnosis of diabetic retinopathy using the
Table” and thus, we can show that our classifier identifies support vector machine: Priya.R, Aruna.P. International Jornal Of
all the normal and PDR tested images whereas in the case of Applied Engineering Research, Dindigul, Volume 1, No 4, 2011.
NPDR it was unable to identify 4 images correctly because [5] R. Gonzalez and R. Woods (1992) Digital Image Processing.
Retrieved from World Wide Web,
of the size of exedutes which is greater than the size of optic
[6] http://users.dcc.uchile.cl/~jsaavedr/libros/dip_gw.pdf.
disk.
[7] Vallabha, D.; Dorairaj, R.; Namuduri, K.; Thompson, H., "Automated
detection and classification of vascular abnormalities in diabetic
TABLE I: Result of SVM Classification retinopathy, " Signals, Systems and Computers, 2004. Conference
Record of the Thirty-Eighth Asilomar Conference on, vol.2, no.,
correctly pp.1625, 1629 Vol.2, 7-10 Nov. 2004
Trained Tested Accuracy
classified [8] Sivakumar, R.; Ravindran, G.; Muthayya, M.; Lakshminarayanan, S.;
Images Images (%)
images Velmurughendran, C.U., "Diabetic retinopathy classification,
96 54 50 92.6% "TENCON 2003. Conference on Convergent Technologies for the
Asia-Pacific Region, vol.1, no., pp.205, 208 Vol.1, 15-17 Oct. 2003
[9] T. Walter, J. C. Klein, and P. Massin, “A Contribution of image
VI. CONCLUSION AND FUTURE WORK processing to the diagnosis of diabetic retinopathy detection of
exudates in colour fundus images of the human retina, ” IEEE
This paper provides a basis of classification of Normal, Trans.Med. Imaging, 21, pp. 1236-43, 2002.
NPDR or PDR affected eye with high accuracy percentage [10] Detection of Exudates in Retinal Images Using a Pure Splitting
of 92.6%. These results strengthen the idea that SVM can be Technique: Hussain F. Jaafar, Asoke K. Nandi, and Waleed Al-
used efficiently and efficiently as a classifier for detecting Nuaimy.32nd Annual International Conference of the IEEE EMBS
Buenos Aires, Argentina, August 31 - September 4, 2010
eye related diseases causes by diabetic. Even with this good
[11] Automated Identification of Diabetic Retinopathy Stages Using
results and progress, our machine would not give desired Digital Fundus Images - Jagadish Nayak & P Subbanna Bhat &
results in case the exudates areas at a particular section in Rajendra Acharya U & C. M. Lim & Manjunath Kagathi. Springer
fundus exceed that of optical disc size. With these Science + Business Media, LLC 2007.
limitations and results, work should be carried on to derive [12] K. Mehrotra, C.K. Mohan & Sanjay Ranka, “Elements of ANN”,
several more options and develop more efficient systems. MIT Press, 1997.
[13] Simon Haykin, ”Neural Networks – A comprehensive foundation”,
Macmillan Publishing Company, New York, 1994.
REFERENCES [14] Text Categorization with Support Vector Machines: Learning with
Many Relevant Features Thorsten Joachims Universit at Dortmund
[1] Kanth, S.; Jaiswal, A.; Kakkar, M., "Identification of different stages Informatik LS8, Baroper Str. 301 44221 Dortmund, Germany
of Diabetic Retinopathy using artificial neural network, "
[15] Drucker, H.; Wu, S.; Vapnik, V.N., "Support vector machines for
Contemporary Computing (IC3), 2013 Sixth International Conference
spam categorization, " Neural Networks, IEEE Transactions on,
on, vol., no., pp.479, 484, 8-10 Aug. 2013
vol.10, no.5, pp.1048, 1054, Sep 1999
[2] Eye Smart - What Is Diabetic Retinopathy?, © 2013 American
[16] Chapelle, O.; Haffner, P.; Vapnik, V.N., "Support vector machines
Academy of Ophthalmology.
for histogram-based image classification, " Neural Networks, IEEE
Transactions on, vol.10, no.5, pp.1055, 1064, Sep 1999
237