Retinal vessel segmentation method based on mixed attention mechanism and asymmetric convolution

The invention discloses a retina vessel segmentation method based on a mixed attention mechanism and asymmetric convolution, which comprises the following steps: firstly, acquiring data, dividing a training set and a test set, carrying out image preprocessing on the training set and the test set, carrying out data enhancement on images of the training set, extracting patches from the training set and the test set, and carrying out image segmentation on the training set and the test set; and constructing a neural network model integrated with a mixed attention mechanism and asymmetric convolution, training the neural network model by using the training set, verifying a model effect by using the test set, inputting the fundus image to be segmented into the trained neural network model, and outputting a retinal vessel segmentation effect by a neural network. According to the method, the asymmetric convolution kernel is used for reducing a large number of training parameters, the calculation complexity is reduced