Detection of Glaucoma using Convolutional Neural Network (CNN) with Super Resolution Generative Adversarial Network (SRGAN)

Eye Disease is one of the major impacts in human life. Glaucoma is dangerous eye disease because it causes permanent blindness by damaging the optic nerves. According to the data, it is one of the primary causes of blindness and the second most common eye illness. Early detection of this disorder is very important to avoid partial or complete visual loss. A high fluid pressure inside the eye occurs when the circulation of a liquid called aqueous humor in the front region of the eyes is obstructed. Therefore, the trabecular meshwork in the eye is blocked. It stop the flow of the fluid and so raise in the pressure causes a change in the size of optic nerve. The communication between the eyes and the brain is lost, resulting in vision loss. Normally, only the most well-prepared physicians perform a laborious physical review on the fundus images. So, CNN-SRGAN is proposed to detect Glaucoma using eye fundus images. Since high resolution image is needed for classification, SRGAN enhancement is done. UNet is deployed for image segmentation. Finally, the images are classified using CNN. The proposed system provides better accuracy in detection of Glaucoma.