Preprocessing of color retinal fundus images

Abstract

Preprocessing is a primary step in many systems for retinal image segmentation and analysis. It improves segmentation and localization of anatomical structures of the image. In this paper, a new automatic method for preprocessing of digital color fundus images is proposed. First, a binary mask is generated using Gaussian filter to define the region of interest. Then intensity information is extracted from both red and green channels to obtain a modified intensity channel component. The performance of the proposed method has been evaluated using 241 images from three publicly available datasets. Experimental results show a visual enhancement of the retinal images with emphasis on the optic disc boundary.