Tongue picture classification model training method, terminal device and storage medium

Abstract

The invention relates to the technical field of tongue picture classification, and discloses a training method of a tongue picture classification model, terminal equipment and a storage medium, and the training method of the tongue picture classification model comprises the steps: inputting an unlabeled tongue picture image into a neural network model for training; preprocessing the label-free tongue picture image to obtain a plurality of groups of image blocks; performing linear projection and feature extraction on each group of image blocks to obtain multiple groups of feature blocks; mask reconstruction and contrast mapping are carried out based on the feature blocks so as to carry out mask loss learning and contrast loss learning, and a pre-trained neural network model is obtained; inputting the labeled tongue picture image into a pre-trained neural network model for supervised training; and performing classification loss learning and detection loss learning on the neural network model through the classification network and the detection network, and taking the learned neural network model as a tongue picture classification model. According to the invention, challenges in few-sample tongue picture classification are solved, and the accuracy and efficiency of tongue picture classification are improved.

Images