Tongue Image Segmentation Based on the Sub-Block Region Growing Algorithm

Segmentation of the tongue area in the tongue image is the basis of the informatization and objectification for the tongue diagnosis. Based on the color features and intensity characteristics of the tongue image, a tongue image segmentation method based on the sub-block region growing algorithm is proposed. The tongue image is transformed from the RGB color space to the HIS color space. According to the seed points selected by the user, the hue and intensity characteristics of the adjacent sub-blocks are analyzed. Then the criteria for sub-block merging in the area are calculated, and all the sub-blocks that meet the criteria are merged. Then the sub-blocks grow and merge again in a smaller size until no more sub-blocks meet the criterion. The experimental results show that the algorithm can effectively segment the tongue image area.