Traditional Chinese medicine tongue quality tenderness identification method based on image restoration and convolutional neural network

Abstract

The invention discloses a traditional Chinese medicine tongue quality tenderness identification method based on image restoration and a convolutional neural network, and the method comprises the steps: obtaining an original tongue image, carrying out the tongue segmentation of the tongue image through a tongue semantic segmentation model, and obtaining a tongue segmentation image; performing tongue coating and tongue quality separation on the tongue body segmentation image by adopting a Gaussian mixture model; obtaining a tongue quality image; establishing a tongue quality image restoration model based on the generative image restoration network, and restoring the tongue quality image by using the tongue quality image restoration model to obtain a tongue quality restoration image with continuous texture features and color changes; carrying out feature extraction and classification on a data set of the tongue quality restoration image obtained after restoration by adopting an improved residual network, and establishing a tongue quality tenderness identification model; and identifying the tenderness of the tongue by using the tenderness identification model of the tongue. According to the method, the influence of the tongue coating on tongue quality tenderness recognition can be avoided, the characteristics obtained through self-learning can reflect richer tongue quality color and texture characteristics, and a good tongue quality tenderness recognition effect is achieved.

Description

technical field [0001] The invention relates to the technical field of tongue image analysis and diagnosis, in particular to a TCM tongue texture identification method based on image restoration and convolutional neural network. Background technique [0002] Tongue diagnosis is one of the important diagnostic methods in traditional Chinese medicine and has high clinical value [35]. Among them, old and tender tongue is an important judgment index of tongue diagnosis. Old tongue with rough tongue texture, firm and old is the main syndrome; young tongue with fine texture, fat and delicate tongue is the main syndrome of deficiency. However, the judgment of tongue quality in clinical diagnosis mainly depends on the doctor's naked eye observation and subjective judgment, and lacks quantitative and objective judgment standards. Therefore, it is very necessary to apply modern computer technology to carry out the research on the identification method of the characteristics of tender...

AI Technical Summary

Problems solved by technology

Although the third method uses a convolutional neural network for tongue image feature recognition, it uses the entire tongue image to identify two types of tongue texture, old and tender, and is also susceptible to interference from the tongue coating.

Images