Glaucoma Detection Using Morphological Filters and GLCM Features

Glaucoma, a neuropathic disorder occurs due to change in shape of optic disk, cup and increase in eye pressure, etc. If Glaucoma is not diagnosed in its initial stages, then it leads to permanent vision loss. Diagnosis of Glaucoma involves various steps including pre-processing or automated processing, features extraction, and classification. Results obtained after automated processing are further used for identification of various properties that has to be analyzed during features extraction. Features or characteristics extraction is used for reducing the image to the various features used for further analysis. Further, classification has been used for differentiating between Glaucomatous and non-Glaucomatous eye. Combination of statistical features and gray level co-occurrence matrix (GLCM) is used for features extraction. Classifier used for differentiating the two classes is support vector machine (SVM), K-nearest neighbor (KNN), and artificial neural network (ANN). Accuracy of classification for KNN, SVM, and ANN is 82.5%, 85%, and 93.7%, respectively.