Abstract—The retinal image diagnosis is an important methodology for diabetic retinopathy detection and analysis.Retinal images play a vital role in most of the applications like ocular retinal image operations and human recognition. Also, it is used to detect the diabetes in early stages by evaluating all the retinal blood vessels together. In this paper, a novel algorithm called multiresolution Curvelet transform and normalized graph cut segmentation is proposed to detect the blood vessels and optic disc in the retinal images efficiently. In our algorithms, the pre-processing takes place, such as image filtration and colour contrast enhancement, and after that, the combined approach for image segmentation and classification are executed using texture, thresholding, and morphological operation. This construction results in a flexible multi-resolution, local, and directional image expansion using contour segments, and thus it is named the Curvelettransform. Our method takes as first step the extraction of the retina vascular tree using the normalized graph cut technique.

Index Terms—Retinal images, Curvelet transform, Contour segments, Normalized graph cut technique

Cite: M Ranganayaki, Nithya Jeevanand and W Nancy, "SEGMENTATION OF BLOOD VESSELS AND OPTIC DISC IN RETINAL IMAGES USING NORMALIZED GRAPH CUT SEGMENTATION," International Journal of Electrical and Electronic Engineering & Telecommunications, Vol. 1, No. 1, pp. 239-245, March 2015.