Research On Color Recognition For Tongue Image In Traditional Chinese Medicine

Tongue diagnosis is one of unique exam methods in Traditional Chinese Medicine (TCM). It is a necessary component in clinical diagnosis. Chinese clinicians have been making tongue diagnoses by means of subjective visual methods which is subject to large inter and intra-observer variation and lacks objective criterion. Therefore, the diagnostic results may be unreliable and inconsistent. With the development and application of Tongue Diagnosis to combine the computer technology such as digital image process and pattern recognition, which prove that traditional tongue diagnosis should be computerized.The segmentation of the body of tongue is a premise to establish a system of automatic diagnosis by the feature of tongue in TCM, whose qualities affect on the performance of tongue diagnosis. Because of similar color between tongue body and background, it is difficult to segment tongue. A tongue segmentation method based on improved Active Contours is presented. In order to overcome two key difficulties with the initialization and boundary concavities (important to the analysis of tooth's marks), a novel method for tongue contour extraction based on improved level set curve evolution is proposed. We present an automatic initialization of contour by the feature of tongue in the HSV color space. Improved level set method takes not only color information but also tongue contour shape constraint represented by energy function between the evolving curve and parametric shape model. Applying our method to the large database of tongue images, we achieve promising experimental results.The separation of tongue coating and tongue proper is a premise to establish a system of automatic diagnosis by the feature of tongue image in TCM, whose qualities affect on the performance of tongue diagnosis. In this paper, we present a novel separation framework of Tongue Coating and Tongue Proper: an improved level set algorithm is designed to assist the separation of tongue coating and tongue proper, in which we explicitly define an external energy that can move the zero level curve toward the object boundaries.The color analyses is the observation of the tongue aspects, however, continuously the Chinese medicine theory quite has been fuzzy to these aspect definition and the classification. Therefore, the method of the color classification of tongue was proposed, which adopted the PCA which reducing the number of dimensions without much loss of information and AdaBoost that will converge to the global minimum in HSV space. Experiments on real tongue image show the method is valid to the color classification of tongue of TCM, which has the very high practical value.When applying the proposed method to the tongue images, we achieved desirable convergence result to detect the edges for various tongue. The result of separation of tongue coating and tongue proper by the improved level set method got satisfying result. The correct rate of color analyses reaches 95% according to the judgments obtained by the experts in Traditional Chinese Medicine. The content of this thesis will promote the process of setting up standard and regulation of tongue exam.