Klasifikasi Citra Glaukoma dengan ANN Berdasarkan Pembuluh Darah Pada Citra Fundus Retina Mengguna

Classification of Glaukoma Image with ANN Based on Blood Vessels on Retina Fundus Image Using Comparison of Otsu-Thresholding Method and Canny Edge Detection

Abstract

Glaucoma is an eye disease that can lead to permanent blindness caused by increased Intraocular Pressure (IOP). There are several methods to detect glaucoma, namely Optical Nerve Hypoplasia Stereo Photographs (ONHSPs), Optical Coherence Tomography (OCT), Scanning Laser Polarimetry (SLP), and Confocal Scanning Laser Ophthalmoscopy (CSLO). However, these methods require a lot of money and expert supervision. In this paper, glaucoma classification will be carried out using the ANN method with a comparison of the otsu-thresholding segmentation method and canny edge detection with the aim of knowing which method gives better results in diagnosing glaucoma images based on the parameters of accuracy, sensitivity, and specificity. The dataset used is RIM-ONE r2 and r3 which will be extracted using 5 features of GLCM and 6 statistical features, and obtained an accuracy of 76% for the otsu-thresholding method, and 79% for canny edge detection.