Hey can you send me the code to koksandeep1[at]gmail.com.
Thanks in advance.

Abstract

This study proposes a semi automated method for glaucoma detection using CDR and ISNT ratio of a fundus image. CDR (Cup to Disc Ratio) is ratio of area of Optic Cup to area of Optic Disc. For a patient with glaucoma Optic Cup size increases while the Optic Disc size remains same and hence CDR will be high
for glaucoma patient than normal fundus image. The ROI of green plane is taken and K-Means clustering technique is recursively applied and Optic Disc and Optic Cup is segmented. Through elliptic fiiting, area of Optic Disc and Cup is determined and hence CDR is calculated. ISNT is another parameter used for the diagnosis of glaucoma which is determined through the ratio of area of blood vessels in Inferior Superior to Nasal Temporal side. Blood vessels will shift to Nasal side for glaucoma patients, hence value will be less for glaucoma patient than normal fundus image. Matched filter and Local entropy thresholding is applied to extract blood vessels. The code is programmed in C++ using OpenCV library functions. OpenCV (Open Source Computer Vision Library) is a library of programming functions developed by Intel. Core, highgui, imgproc,ml are the main libraries used from OpenCV. The optimized functions in OpenCV increase the speed of operation and is very much suitable for real time mass screening purpose. A batch of 50 retinal images (25 normal set and 25 abnormal set) obtained from the Aravind Eye Hospital, is used to assess the performance of the proposed system.