Glaucoma is the second major cause of loss of vision in the world. Assessment of Optic Nerve Head (ONH) is very important for diagnosing glaucoma and for patient monitoring after diagnosis. Robust and effective Optic Disc (OD), Optic Cup (OC) detection is a necessary preprocessing step to calculate the Cup-To-Disc Ratio (CDR), Inferior Superior Nasal Temporal (ISNT) ratio and distance between optic disc center and optic nerve head (DOO). OD is detected using Region Matching (RM) followed by medial axis detection in fundus images. OD and OC segmentations are carried out by Improved Super Pixel Classification (ISPC), Adaptive Mathematical Morphology (AMM) and the effectiveness of the algorithm is compared with existing k-means and fuzzy C means (FCM) algorithm. Experiments show that OD detection accuracies of 100%, 98.88%, 98.88% and 100% are obtained for the DRIVE, DIARETDB0, DIARETDB1 and DRISHTI data sets, respectively. In this work, three statistically significant (p<0.0001) features are used for Naive Bayes (NB), k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), Feed Forward Back Propagation Neural Networks (FFBPNN), Distributed Time Delay Neural Network (DTDNN), radial basis function exact fit (RBFEF) and Radial Basis Function Few Neurons (RBFFN) classifier to select the best classifier. It is demonstrated that an average classification accuracy of 100%, sensitivity of 100%, specificity of 100% and precision of 100% have been achieved using FFBPNN, DTDNN, RBFEF and RBFFN. With expert ophthalmologist's validation, Decision Support System (DSS) used to make glaucoma diagnosis faster during the screening of normal/ glaucoma retinal images.