| Traditional Chinese Medicine (TCM) consists of four diagnostic methods suchas look, smell, ask and palpate‘, and tongue diagnosis is the key content ofinspection, the doctor of traditional Chinese medicine clinical will be found. In thetheory of traditional Chinese medicine, the human body is regards as an organicunity of the whole, which means every change of each part is closed related with thewhole. And just so Tongue is considered the most intuitive reflection of humanphysiological symptoms, the important information of internal organs such as Qi,Blood and Body Fluid, and essence Qi can be accessed through the tongue image.With obvious advantages, Chinese tongue diagnosis can acquire the wholeinformation of internal organs easily and quickly via observing the tongue image nomatter how complex the pathological symptoms are. And tongue diagnosis play aguiding role in the prescription drug-used as well as disease prognosis. Thus, thesignificant feature of Chinese tongue diagnosis is painless, non-invasive.Chinese tongue diagnosis, however, also has its traditional drawbacks, it is toodependent on subjective observation of Chinese medicine doctors, and the diagnosticresults usually vary from person to person, it‘s also not repeatable generally, whichextremely blocked it‘s further development. Aiming at this point, it’s even moreimportant to the observation of tongue diagnosis. If we can overcome this problem,it not only caters to the standardization and quantitative requirements that TCMpromote, but also promotes the practical application value of tongue diagnosis.So, In order to promote the Chinese tongue diagnosis better, this paper will aimat the observation of the objectification and systematic of tongue diagnosis whichmainly refer to the DIP, Related content and innovation are as follows:In the first chapter, we mainly introduce the development and research status,point out the advantages and problems while applying DIP to the observation oftraditional tongue diagnosis, and research content and significance of this paper. In the second chapter, the establishment of framework the auxiliary system oftongue diagnosis which based on the Android OS, frequently-used color model, thetheory of variable model, Level Set, image feature extraction and image recognitiontheory are introduced.In the third chapter, we contrast characteristics of H S V components fromtongue image and obtain the initial contour of the tongue through thresholdsegmentation; Aiming at the characteristics of fuzzy boundaries between the tongueand face, we enhance the G component in RGB color space to enhance the weakboundary of tongue significantly; Finally we improve the traditional level setalgorithm to segment the tongue image better. As we can see by a large number ofexperiments, the improved algorithm can improve the effect come from the weakboundary as well as obtain better segmentation effect in the end.In the fourth chapter, we firstly divide the segmented tongue according to thereflection area of internal organs. Then, we take example of the center of tongue(corresponding to the spleen and stomach) to study the TCM five-color recognitionclassification method in the application of tongue diagnosis: to feature the extractedtongue block by the quantified color histogram, and building a tongue colorrecognition model by this features with sLDA method, which will be used torecognize the tongue color.In the fifth chapter, we test and apply the tongue collection and auxiliary systemof tongue diagnosis based on the Android OS tablet. Introduce the operation processand information of test, analyze its application prospect and prefect its function lateron (this version is1.0by default).In the sixth chapter, we make a summarization and prospect. Firstly summarizethe main research contents and innovation of the subject, then prospect theapplication of the objectification of TCM tongue diagnosis to the mobile platform. |
