Detection of Eye Strain using Retina Medical Images through CNN

In recent days the number of people who are using Multimedia have increased. Due to long exposure on such streams causes Discomfort in Eyes which leads to Dryness, Headache and blurry vision. In the Proposed system, the detection of Eye Strain through Retina images is done by Transfer learning model using Convolutional Neural Networks. In order to identify the condition of eyes, the training and testing of images are done using Pre-trained models such as Inception V3 and Resnet152V2 architectures. These two architectures are used to classify the stage of the disease by combining the Optimization algorithm such as Adam with each transfer learning model. Finally, the conclusion was done with the model performance by comparing the number of Epochs trained. The ResNet152V2 with the optimization algorithm of Adam have performed better by gaining higher accuracy by minimizing the loss. The inception v3 with Adam optimizer obtained a good accuracy in differentiation with other traditional models.