The Study Of Detectine Methodfor Syndrome Elments In Tongue Manifestation

Tongue diagnosis is one of the unique medical diagnostic methods of Traditional Chinese Medicine(TCM), which is a principal evidence for TCM diagnosis. Tongue diagnosis can reflect the pathological changes of the human body objectively. Therefore, using computer technology to achieve the condition records, syndrome elements detection, prescription assistance, and telemedicine is an important part of TCM to realize modernization and informatization. This paper studies the method of detecting syndrome elements in tongue manifestation, the initial development of a computer-aided system tongue. This paper mainly research on the tongue color correction, including tongue body, tongue coating and the top of tongue. We also make a research on the tongue image segmentation and discrimination. The main contents are as follows:Firstly, the tongue image acquisition system designed by ourselves, to avoid shortcoming that the traditional tongue diagnosis is easily affected by illumination. The system consists of a black box, a standard light source, a camera and a computer. The black box shields the most of the external light source to provide a stable acquisition environment. The standard light source provides illumination environment that close to natural light. The camera is connected with a computer to realize of tongue image acquisition, storage and processing.Secondly, The color correction of tongue image is the premise of tongue image analysis. Captured the tongue will be affect by cameras, and environmental, resulting in inaccurate tongue color. Observation of tongue color is an important content of tongue inspection in traditional Chinese medicine. However, the current research on tongue color correction is less, and not comprehensive. This paper presents an optimization method commonly used to improve the color correction, not only more reliable data are provided for detecting tongue manifestation syndrome elements the diagnosis of TCM, but also for detecting tongue manifestation syndrome elements provides a reliable premise.Thirdly, image segmentation is a major step in the tongue image processing. This paper studies the classical image segmentation methods include threshold method, region growing method and watershed method Based on analyzing the advantages and disadvantages of the classical algorithms, a new image segmentation method is proposed that combined with region growing algorithm and the watershed algorithm. And it obtained a better segmentation effect.Finally, the color of tongue is the main factor in detect syndrome elements of tongue manifestation, which include the color of tongue body, tongue coating and the top of tongue. We propose an improved threshold method to realize the separate of tongue coating and tongue body, then color classification. Through tongue symptoms accurately extract to realize syndrome elements detection...