Tongue region detection method and system based on deep learning

Abstract

The invention discloses a tongue region detection method and system based on deep learning. The method comprises the steps: carrying out the marking of an obtained image data set comprising a tongue,and carrying out the preprocessing of the marked image data set, so as to obtain a first image data set; setting proportions of a plurality of fixed reference frames, and performing clustering in a k-means clustering mode to obtain a plurality of clustering center reference frames; performing training based on a DarkNet network structure, determining an output layer dimension according to the plurality of clustering center reference boxes, and performing training according to the first image data set to determine a first detection model; utilizing the first detection model to detect an image data set which does not contain a tongue part so as to obtain a false detection image data set of false detection; and adjusting the network structure of the first detection model, modifying the dimension of an output layer, and retraining by using the first image data set and the false detection image data set to determine a tongue detection model for tongue region detection.

Description

technical field [0001] The present invention relates to the technical field of deep learning algorithms, and more specifically, to a tongue region detection method and system based on deep learning. Background technique [0002] At present, many tongue diagnosis algorithms based on TCM theory require the subject to stick out his tongue at a fixed distance and a fixed area through fixed devices or equipment when performing tongue diagnosis and analysis. Then analyze the pixels in this fixed area in the picture. In the real world, different people have different tongue protruding states and protruding sizes. If a fixed area is used, there may be more background pixels than tongue area pixels, which has a great impact on the actual tongue diagnosis analysis. [0003] Moreover, the existing technical application scenarios have great limitations. The main problems include: 1. There are a lot of tongue image shooting requirements for users, and users need to stick out their tong...

AI Technical Summary

Problems solved by technology

[0005] The present invention proposes a tongue region detection method and system based on deep learning to solve the problem of how to accurately and intelligently determine the tongue region

Images