Computer Aided Tongue Diagnosis System In Traditional Chinese Medicine

Tongue diagnosis, one of the jewels of traditional Chinese medicine (TCM), is a profoundly subtle instrument for the early diagnosis and prevention of disease. Yet far too often, in the haste of modern education and practice, it has become a neglected art. The application of tongue inspection of TCM on the quantificational diagnosis of the tongue body and tongue coating can provide objective references for clinic of TCM.The inspection of the tongue is a vital diagnostic procedure in the practice of TCM. The color, coating, shape and texture of various parts of the tongue yield information about the state of the organs. The tongue is the only internal organ visible from the outside of the body and so to TCM healers, it is no less accurate than an X-ray of what is happening internally. To work out more precisely what is going on inside the body, TCM doctors use image process methods to objectify and quantify the tongue diagnosis research.Image recognition is an active research area in image processing. It is a specific type of pattern matching that identifies patterns in a set of pixels. The basic idea of image recognition is to analyze the images, extract features of lower levels, compare and classify these features, finally identify the contents of images. Nowadays with the rapid development of image processing technology, TCM doctors hope to build up an automatic digital tongue diagnosis system to guide the clinical and educational treatment. When integrate with the image processing methods, traditional tongue diagnosis can be more precise, more objective and more efficient.The content and innovation points of this thesis are as the following: first, some background information in chapter 1 and 2. A novel algorithm to detect and segment a tongue from the background is put forward in chapter 3. We focus on how to separate tongue body from tongue coating in chapter 4 and come up with a dynamic threshold detect algorithm. In chapter 5, all aspects of tongue feature are discussed and quantificational methods are proposed. Finally, in chapter 6 we talk over determine the category of a tongue image using the image database and the disease database.