Artificial intelligence based tongue image detection and recognition method

Abstract

The invention discloses an artificial intelligence based tongue image detection and recognition method. By adoption of camera of a mobile phone or a tablet computer or a special tongue image acquisition device, limitation of large-scale tongue image acquisition due to a single acquisition mode is solved, and convenience is provided for tongue image acquisition. Real-time automatic detection of tongue images and tongue positions can be realized, manual or automatic shooting, storing and uploading can be realized, and the problem of difficulty in accurate detection of a tongue body due to the fact that other parts of the human face in a traditional acquired tongue image occupy most of areas while the tongue body position is less is solved. Tongue colors, tongue shapes, tongue coating textures and tongue coating texture properties can be automatically recognized in the tongue images, and doctors of traditional Chinese medicine can be assisted in tongue diagnosis. A detection and recognition process avoids a complex man-machine interaction process, and quickness in tongue image data analysis and calculation and accuracy of results are realized.

Description

technical field [0001] In particular, the invention relates to an artificial intelligence-based tongue image detection and recognition method. Background technique [0002] Diagnosis steps in TCM include four aspects: looking, smelling, asking, and feeling. The main method of inspection is tongue diagnosis to diagnose patients' diseases. Tongue diagnosis is a diagnosis made by TCM doctors by observing relevant information of the tongue image and combining their own experience, which is easily influenced by the doctor's subjective consciousness. However, the present invention standardizes the relevant information of the collected tongue image data, detects and recognizes it with the help of artificial intelligence technology, and helps doctors of traditional Chinese medicine quickly conduct tongue diagnosis and make an accurate diagnosis. [0003] Tongue image detection and recognition is an important part of the tongue diagnosis process, and the use of computer-aided detect...

AI Technical Summary

Problems solved by technology

To sum up, the existing tongue image detection method is relatively complicated and the process is cumbersome, and has the following disadvantages: 1: Tongue image acquisition needs to be carried out in a special tongue image acquisition device, the acquisition method is single, and it does not have the ability to collect tongue images on a large scale. narrow range

Disadvantage 2: The above-mentioned tongue image detection method is relatively complicated, the process is cumbersome, and the detection process takes too long

Disadvantage 3: In the collected tongue image, other parts of the face occupy most of the face, and the position of the tongue body is relatively small, so it is relatively difficult to detect the tongue body correctly

Images