Glaucoma illness is growing increasingly common as a result of strain on the eye cells. Several image processing-based approaches have previously been employed to diagnose glaucoma, however their classification accuracy was lacking. Mobile phones and video games are increasingly being utilised in daily life, putting additional strain on the eye cells. In this study, the Bin Rushed database is utilised to illustrate three distinct ways for detecting glaucoma illness utilising image processing, machine learning, and a deep learning convolutional neural network model. Image processing techniques are used to extract features such as CDR and RDR, which are then classified using a neural network, support vector machine, and other approaches. The VVG-16 deep learning model has an accuracy of 99.6%, with the K nearest neighbour strategy having the highest accuracy of 98%