Volume : III, Issue : VII, July - 2013

Detection of Haemorrhages in Retinal Images

V. Vijayakumari

Abstract :

Now a days in the field of ophthalmology, haemorrhage is the term used more often because of increasing diabetic patients. Its a challenge amidst the ophthalmologist to distinguish the haemorrhage from the blood vessels, this lands in various problems. In the past various techniques were employed for the detection of the haemorrhage but they were not so accurate and often encountered misclassification between haemorrhage and blood vessels. Precise detection and classification of haemorrhage and blood vessel is very important in the diagnosis of many problems. Two algorithms are implemented for the detection of haemorrhages from pathological retinal images. The first algorithm has three main steps namely color image enhancement, image subtraction to extract blood vessels and haemorrhages and use of set of optimally adjusted morphological operators to suppress blood vessels and to highlight only haemorrhages which reduces the misclassification effectively. The second algorithm is based on the concept of Cellular Neural Network which encompasses 2D or 3D arrays of mainly locally connected nonlinear dynamical systems called cells, whose dynamics are functionally determined by a small set of parameters which determine the connection pattern, and are collected into the so–called cloning templates. CNN operates very fast due to its parallel architecture which performs both linear and nonlinear image processing tasks, such as filtering, thresholding and various mathematical morphology operations. Here both the algorithms are compared in terms of certain parameters based on specificity, sensitivity to come up with better results.

Keywords :

Article: Download PDF   DOI : 10.36106/ijar  

Cite This Article:

V. VIJAYAKUMARI Detection of Haemorrhages in Retinal Images Indian Journal of Applied Research, Vol.III, Issue.VII July 2013

Number of Downloads : 577

References :