Detection of exudates and feature extraction of retinal images using fuzzy clustering method

• Published in: Proceedings of third International Conference on Computational Intelligence and Information Technology pp. 388 - 394
• Main Authors: Princye, P.H, Vijayakumari, V
• Format: Conference Proceeding
• Language: English
• Published: Stevenage, UK IET 2013; The Institution of Engineering & Technology
• Subjects: Combinatorial mathematics; Computer vision and image processing techniques; Image recognition; Integral transforms; radii method; fuzzy set theory; fuzzy clustering method; optic disc centre; retinal recognition; DR; CHT; feature extraction; exudates detection; retinal images; fundus image; optic disc region; blood vessels; optic disc; retinal exudates detection; CLACHE; blood vessel extraction; diabetic retinopathy; circular Hough transform; FCM clustering method; contrast limited adaptive histogram equalization; pattern clustering; green component; connected component analysis; Hough transforms
• ISBN: 9781849198592; 1849198594
• DOI: 10.1049/cp.2013.2617

Diabetic retinopathy (DR) is the micro vascular changes that cause detectable changes in the optic disc. This paper aims at the detection of retinal exudates and other features such as blood vessels and optic disc from fundus image. The two methods are implemented for the detection of exudates they are morphological method and FCM clustering method. Contrast limited adaptive histogram equalization (CLACHE) is used to extract the green component in the image. In blood vessel extraction, blood vessels are extracted by top hat transformation followed by connected component analysis. The optic disc centre is found using Circular Hough Transform (CHT) and propagation through radii method is employed and the entire optic disc region is blackened and removed. Exudates detection is the important characteristics of diabetic retinopathy and its varies depends upon the severity of the DR. The FCM method used to detect exudates. The overall sensitivity, specificity and accuracy are calculated and 98% accuracy obtained.