A Fuzzy-based Image Segmentation on Diabetic Retinopathy Model

Abstract

Clinical reports done suggested that more than ten percent patients with diabetes have a high risk of eye issues. The retinal fundus images are commonly used for detection and analysis in diabetic retinopathy disease. This work presents several state to extract the anatomic components and lesions in colored fundus photographs and some decision support methods to help early clinical diagnosis detection. It also introduces a model of detection in fundus images by automated segmentation of region of interest (ROI). Automatic segmentation of retinal blood vessels from retinal images is applied to make landmarks detection more efficient. The proposed model integrates adaptive Otsu's Threshold and Segmentation using FUZZY C-MEANS clustering for automated detection of hard yellow spots. The proposed hybrid fuzzy-based ROI extraction scheme integrates the effect of the local neighborhood and allow it to influence the membership value of each pixel. A new Hybrid FCM (H-FCM) algorithm is proposed, which integrates spatial information with a 2D adaptive noise removal SS-FCM model. Experiments have been conducted to verify the proposed model. Experiments showed that this proposed model produced high performance in ROI detection under different effects and on different types of retina images. Moreover, the results showed high sensitivity compared with recent researches.