A condition known as glaucoma, which affects the optic nerve, can cause visual loss that is either partial or total. Therefore, there is a critical need for glaucoma screening at an early age. The symptoms related to glaucoma are not noticeable until severe stages of this illness and they appear when the patient has already lost a significant part of his eyesight. The majority of earlier diagnosis strategies rely on handmade feature engineering. The use of fundus images for the early detection of eye disorders is of tremendous clinical value. Deep learning is becoming more and more common in related applications, such as lesion segmentation, biomarkers segmentation, disease diagnosis, and image synthesis, due to its potent performance. Convolutional neural networks (CNNs) have recently been utilized in the field of ophthalmology to identify specific eye ailments, including glaucoma. They have demonstrated good performance in the early detection of several diseases. The method described in this paper uses densely connected neural networks with numerous layers that were initially trained on ImageNet using the fundus dataset that is readily available. We can be cautiously optimistic about the efficacy of our classification model for the early identification of glaucoma given the accuracy of almost 97 that was achieved.

No abstract