Retina optic disc detection and segmentation with fuzzy clustering mean

The segmentation of optic disk helps for doctors to find the problems in eye. All nerves pass their direction from this disc to the brain and heart. In this study, we used the fuzzy clustering method based on the Otsu binarization for optic disk segmentation. Fuzzy clustering is a computer vision task to estimate pixel tags corresponding to the region to which it belongs or to the region of the surrounding region. Fuzzy clustering aims to understand the class of special objects in the scene. Fuzzy clustering based on Otsu binarization method has shown outstanding performance for optic disk image segmentation and detection. For testing the method public retinal images from DRIVE, STARE and MESSODIR dataset are used. In this study,for accuracy evaluation we used the confusion matrix that calculate the true negative (TN), true positive (TP), False positive (FP) and false negative (FN). We have compare the output result with the ground truth images. For precision, recall and accuracy we have get high accuracy. KEYWORDS: Image segmentation, Optic Disc Segmentation, Fuzzy clustering