Glaucoma Detection Using UNET Model

Abstract

Most of the algorithms for automated glaucoma assessment using fundus images supported segmentation, which are suffering from the performance of the chosen segmentation method and therefore the feature extraction process. Among other characteristics, convolutional neural networks (CNNs) are known because of their ability to learn highly discriminative features from raw pixel intensities. Recently, the attention mechanism has been successfully applied in convolutional neural networks (CNNs), for boosting the performance of many computer vision tasks. Unfortunately, some of the medical images are incorporate in CNNs. In particular, there exists high redundancy in fundus images for glaucoma detection, such the eye mechanism has potential in improving the performance of glaucoma detection using CNN. This paper proposes a U-net CNN for glaucoma detection. It is use to detect the exact place where the glaucoma occur in our eye, using the U-net model. U-net is the advance version of the CNN model the algorithm is simple and it is 92.5% efficient than CNN model .The image are separate by using preprocessing, segmentation method, feature extraction and classification this are the step which was followed in this U-net.