RECOGNITION MODEL TRAINING METHOD AND APPARATUS, FUNDUS FEATURE RECOGNITION METHOD AND APPARATUS, DEVICE AND MEDIUM

摘要

A recognition model training method and apparatus, a fundus feature recognition method and apparatus, a computer device and a storage medium. The method comprises: acquiring a fundus color image sample associated with a tag value, and inputting the fundus color image sample into a preset recognition model, wherein the preset recognition model comprises an input layer unit, a first convolutional neural network and a second convolutional neural network; extracting a red channel image in the fundus color image sample and inputting same into the first convolutional neural network, and acquiring a first recognition result, and a feature map of the red channel image; combining the fundus color image sample with the feature map to generate a combined image, inputting the combined image into the second convolutional neural network, and acquiring a second recognition resu acquiring a total loss value by means of a preset loss function; and when the total loss value is less than or equal to a preset loss threshold value, completing training. A tessellated fundus feature of a fundus color image is automatically recognized, the accuracy of a recognition model is improved, and the efficiency and reliability of the recognition model are improved.