A progressive loss of vision known as glaucoma is brought on by injury to the optic nerve, specifically the cumulative loss of retinal cells. One condition that impairs vision in the human eye is glaucoma. This illness is thought to be permanent and to induce blindness. They have no early warning signs for this glaucoma. The effects are so subtle that you may not notice any changes in your vision. So far, many models of deep learning (DL) have been created to accurately identify glaucoma. Therefore, we present design based on deep learning for accurate glaucoma detection using convolutional neural networks (CNNs). The distinction between both glaucoma and non-glaucoma patterns can be determined using CNN. To differentiate, CNN offers a picture system with levels. Current methods detect disease. Whether a patient has glaucoma depends on the relationship between the eye socket and the optic disc. Diagnosis is improved by integrating image data generation methods for data enrichment. The results show that the recommended model output forms 98% accuracy and several other existing methods.