Eye Hypertension Disorder Through Ensemble Frame Networks: A Novel Approach for Early Identification and Segmentation

The most common cause of blindness in humans today is glaucoma, which can potentially cause total blindness in a normal individual. The development of this neurodegenerative condition is caused by high intraocular pressure in the retina. Developing an expert system to recognise the physiological alterations occurring within the human eye is a challenging task. The current technique finds it extremely difficult and time-consuming to adequately identify these infractions since they are categorised manually. Due to these pressing problems, machine learning and deep learning approaches are receiving more attention in the development of glaucoma early diagnostic systems. This paper proposes a novel ensemble of Frame networks with the UNET learning model to achieve effective segmentation and early diagnosis of Glaucoma. The proposed research introduces the Frame networks combined with severe feedforward along self-attention levels. Several datasets, such as DRISHTI-GS, and the Optic Nerve Segmentation Database (DRIONS-DB) have been thoroughly explored with using the predicted Model; the perfection of the anticipated Model is shown by computing and comparing its enactment metrics to additional deep learning facsimiles that are currently in use. The detection accuracy for the DRISHTI-GS, DRIONS-DB, is 98.5%, 99%, respectively.