Eye fundus image classification method based on convolutional neural network

The invention relates to a fundus image classification method based on a convolutional neural network, and belongs to the technical field of medical image processing. The method comprises the following steps: S1, taking an eye fundus image as input, and obtaining a tissue image corresponding to each pixel; s2, performing feature extraction on the tissue image through a pre-trained convolutional neural network; wherein the pre-trained convolutional neural network, namely the feature extraction network, is a convolutional neural network obtained through transfer learning by using a public texture image set and a small amount of tissue image sets; s3, after feature extraction, a decoder-encoder network in the twin network based on the GAN is adopted to match the distribution of a source domain and a target domain in the reconstructed regularized shared space, and a discriminator is utilized to distinguish the target domain from the source domain; and S4, training network parameters by adopting adversarial loss a