Blood Vessel Extraction of Fundus Images Based on Edge Enhancement using Kirsch's Template and other Image Processing Techniques

J. Pradeep Kandhasamy and S. Balamurali

ABSTRACT

Objective - Now a day's diabetes has become a more dangerous and quickly increasing disease all over the world irrespective of age group people. Since untreated diabetic leads to various ophthalmic disorders, among that Diabetic Retinopathy (DR) is a dangerous retinal disease. The main disadvantage of this DR is that it has no prior clear symptoms. So, the people with diabetic should undergo regular retinal monitoring process for the early stage detection of DR. Extracting Blood Vessels from the Fundus images is a challenging task. This paper aims to develop a better algorithm which uses image processing techniques like Kirsch's templates, image pre-processing, edge enhancement and thresholding for the extraction of Blood vessels.

Methodology/Technique - Our proposed algorithm has the following stages. First the RGB images are converted to Gray scale image to reduce the size of the image to uniform area. Then the image is passed to the filter with Kirsch's template for edge enhancement and it is subjected to threshold after that the blood vessels are extracted.

Findings - The results obtained from the proposed method are compared with the publicly available DRIVE database and found that the accuracy is high when compared to other techniques.

Novelty - Our proposed system will help the ophthalmologists for only assessing and analysing the funds images for identifying diabetic retinopathy and it will not replace the ophthalmologists for taking any decision.

Type of Paper: Review

Keywords: Diabetic Retinopathy, Filtering, Image processing, Krisch's template, Vessel Extraction.

Global Journal of Engineering & Technology Review Conferences 2016-17:
Meet Inspiring Speakers and Experts at our 1000+ Global Annual Meetings