Tongue Image Segmentation And Detection And Repair Of Reflection

Tongue diagnosis of traditional Chinese medicine(TCM)is an extremely effective and non-invasive auxiliary diagnostic technique,and it analyzes the physiological and psychopathological changes of human body by observing the characteristics of tongue body.It can assist diagnose sub-health which is the "disease of the century" that is now threatening global health.But traditional tongue diagnosis mainly relies on the subjective judgment of doctors to diagnose illnesses,lacking of objectivity,and hinder the widespread application of tongue diagnosis.With the development of computer information technology,the intelligent tongue diagnosis system is born and gradually leads tongue diagnosis towards objectification and modernization.This paper studies the low accuracy of tongue segmentation and the problems such as false recognition of tongue and loss of texture information in the process of tongue surface reflector processing in the current tongue diagnosis system.And propose the following solutions:To solve the problem of tongue image segmentation,in this paper,an improved distance preserving level set evolution(DPLSE)model is proposed by revising the edge indicator function and the area parameter.Before segmentation,the grey-level integral projection algorithm,Otus method as well as morphological processing will be utilized to handle the tongue images,which to obtain the initial contour curve of this method.The improved algorithm has been experimented on 20 tongue images.Compared with other methods,it shows the competitive performances on tongue image segmentation.For the reflection on tongue surface,a new data fusion method and ameliorated CDD model are proposed.The data fusion method makes use of the gray histogram characteristics of tongue image and combines the data fusion algorithm to detect the reflection on tongue surface.And the improved CDD model introduces a new curvature function based on the traditional CDD model in order to repair the detected reflection points.The experimental results show that the method proposed in this paper can reduce the misjudgment of the reflection and better restore the reflection spot on tongue surface.