PICTURE CLASSIFICATION MODEL TRAINING METHOD AND SYSTEM, AND COMPUTER DEVICE

A picture classification model training method based on a small amount of data, said method comprising: constructing a pre-training model according to an ImageNet model, the ImageNet model being a network model which has been obtained according to preset pre-sample pictures (S100); acquiring a plurality of sample pictures, and obtaining a sample picture training set according to the plurality of sample pictures, the plurality of sample pictures comprising pictures of a plurality of target picture types (S102); and training the pre-training model by means of the sample picture training set, so as to obtain a picture classification model (S104). In said method, model training is performed on a small amount of data, so that the image training speed, the image recognition speed and the image recognition accuracy can be improved, and the types of distinguishable images can also be increased, thereby greatly improving the efficiency of image classification. Procédé d'entraînement de modèle de classification d'image