Eye fundus photo age identification method and device based on residual network model, and storage medium

The invention relates to a residual network model-based fundus photo age identification method and device and a storage medium, and is applied to the technical field of fundus photo age identification, and the method comprises the steps: inputting fundus photo data into a WGAN network, and obtaining the expanded fundus photo data; an SE attention mechanism module is added into the Resnet neural network model; obtaining an optimal model through training; recognizing the age of the fundus color photo through the optimal model; in the scheme adopted by the invention, aiming at the characteristic that fundus retina image data is not easy to obtain in batches, the image data of the training set is expanded by using the WGAN generative adversarial network, so that the training efficiency of the model is higher, and neural network training is performed in combination with a channel attention mechanism on the basis of the ResNet18 residual network, so that the accuracy of model recognition is improved. 本发明涉及残差网络模型的眼底