Tongue image multi-label classification learning method based on graph convolution network

Abstract

The invention discloses a tongue image multi-label classification learning method based on a graph convolution network, and the method comprises the following steps: S1, carrying out the tongue body detection of an original image, and obtaining a tongue body image through extraction; S2, performing image preprocessing on the tongue body image extracted in the step S1, wherein the preprocessing comprises reflection point removing processing, sharpening processing and straightening processing; S3, for each label, performing semi-automatic labeling on the preprocessed tongue body image to obtaina large-sample multi-label data set; and S4, training and inferring the large-sample multi-label data set obtained in the step S3 by using a graph convolution network to obtain a tongue body multi-label classification model based on the graph convolution network. According to the invention, the plurality of tags of the tongue image are classified and diagnosed at the same time through one graph convolution network, and the dependency relationship among the tags is fully learned, so that the tongue diagnosis process of the machine becomes more efficient and accurate.

Description

technical field [0001] The invention relates to the technical field of detection and classification of machine vision for tongue diagnosis in traditional Chinese medicine, in particular to a new method for tongue detection, tongue preprocessing, tongue semi-automatic labeling process and tongue image multi-label classification based on graph convolutional network. Background technique [0002] Among the four diagnoses "look, smell, ask, and feel" based on the diagnosis of traditional Chinese medicine, "look" is the most important. "Tongue observation" is an important part of "examination", because the internal organs of the human body are connected to the tongue through meridians, and changes in the human body can be reflected in the tongue. Tongue diagnosis in traditional Chinese medicine is observed with the naked eye, which is highly subjective. Therefore, quantitative analysis methods can provide a basis for more accurate tongue diagnosis. [0003] Tongue diagnosis is ...

AI Technical Summary

Problems solved by technology

[0004] In the past, most of the classification studies on tongue images classified each label separately, ignoring the dependencies between labels, and the results output multiple classification models, which means that multiple models need to be loaded during inference. thereby affecting efficiency

A small number of studies using multi-label either did not use deep learning technology, or did not fully mine the dependencies between labels, which affected the accuracy

Images