Tongue coating greasyness identification method based on texture and distribution characteristics

Abstract

The invention discloses a tongue coating greasyness identification method based on texture and distribution characteristics. The method comprises steps that, (1), a training sample set is established by utilizing acquired tongue surface image samples in combination with Gabor textures, Tamura roughness and tongue coating distribution characteristics; (2), the training sample set is trained by utilizing a support vector machine to establish a tongue coating greasyness classifier; and (3), the established tongue coating greasyness classifier is utilized for identification to determine whether a to-be-identified tongue surface image belongs to a greasy tongue phenomenon type. Through the method, tongue surface image textures, roughness and tongue coating distribution characteristic vectors are acquired, the support vector machine is employed as the classifier to simulate tongue coating image identification of Chinese medical specialists, and repeatability is realized; in an identification stage through the method, artificial guidance and parameter setting are not necessary, the greasy tongue phenomenon determination mechanism is more visual and fairer; Chinese medical identification on the greasy tongue phenomenon is simulated, and problems of strong subjectivity and poor repeatability existing in traditional Chinese medicine visual observation about the greasy tongue phenomenon are solved.

Description

technical field [0001] The invention relates to a method for identifying greasy tongue fur based on texture and distribution features, and belongs to the fields of pattern recognition and image processing. Background technique [0002] There are four basic methods of TCM diagnosis, namely: looking, smelling, asking, and feeling. The main contents of tongue inspection include tongue quality and tongue coating. Tongue coating is a layer of fur attached to the tongue surface. The color, thickness and degree of decay of the tongue coating can all reflect the health of the internal organs. Tongue fur can be divided into rotten fur and greasy fur according to the degree of greasy tongue fur. Greasy fur is characterized by fine and dense moss particles, which are fused into sheets, thick in the middle and thin around, and cling to the tongue surface, which cannot be wiped off, and is not easy to fall off when scraped. On the other hand, rotten fur is characterized by coarse part...

AI Technical Summary

Problems solved by technology

Use computer to simulate Chinese medicine experts to identify tongue coating, and then overcome the disadvantages of strong subjectivity and poor repeatability of traditional identification of rotten tongue coating

Images