Detection of Glaucoma and Diabetic Retinopathy from Fundus Images by Bloodvessel Segmentation

Abstract

Blood vessel segmentation of fundus images has obtained considerable importance during the past few years since it facilitates the early detection of eye diseases. A method based on high pass filtering and morphological operation is introduced in the proposed method for vessel segmentation. This method can be utilized to detect diseases effecting eyes like glaucoma and diabetic retinopathy. Glaucoma is detected by feature extraction and classification. The local binary pattern of the optic disc is extracted to classify the images on the basis of texture. Sparse representation classifier is utilized to classify the glaucomatous eye. Diabetic retinopathy is a disease caused by the complexity of diabetes. It damages the small blood vessels in the retina resulting in loss of vision. The blood vessel segmentation is an important task in Diabetic Retinopathy detection. Optic disc in the fundus image is detected by Hough transform. After the segmentation the vessels and optic disc are removed from the original image. Diabetic Retinopathy is characterized by the presence of exudates. The exudates are detected by means of imtool operator in the matlab. The simulations are performed on matlab 2011 and the data are collected from DIARETDB1 and HRF databases.