JSEG Based Color Separation of Tongue Image in Traditional Chinese Medicine
The process of color separation of tongue image in traditional Chinese medicine (TCM) is decomposed into two steps: region partition and color classification. In the first step, a partition method based on JSEG is proposed to obtain homogenous regions in the tongue. In the second step, a priori template and some standard color patches are designed to assist the classification according to a priori knowledge from the TCM experts, where the nearest neighbour classifier is applied for color classification. The experimental results show that the proposed approach is greatly promising to computerized tongue diagnosis.
- 1.Jang, J.H., Kim, J.E., Park, K.M., et al.: Development of the digital tongue inspection system with image analysis. In: IEEE Proc. of the 2nd Joint EMBS/BMES Conf. Volume I., Houston, TX, USA, pp. 1033–1034 (2002)Google Scholar
- 3.Zhao, Z.X., Wang, A.M., Shen, L.S.: An automatic tongue analyzer of chinese medicine based on color image processing. In: The 4th Int. Conf. on Electronic Measurement and Instruments Conf. Proc., Harbin, China, pp. 830–834 (1999)Google Scholar
- 4.Wang, Y.G.: Research on pragmatizing the tongue image analysis instrument of traditional chinese medicine. Master’s thesis, Beijing Polytechnic University (2001)Google Scholar
- 5.Deng, Y.N., Manjunath, B.S.: Unsupervised segmentation of color-texture regions in images and video. IEEE Trans. on PAMI 23, 800–810 (2001)Google Scholar
- 6.Deng, Y.N., Kenney, C., Moore, M.S., et al.: Peer group filtering and perceptral color image quantization. In: Proc. IEEE Int. Symp. Circuits and Systems, Orlando, FL, vol. 4, pp. 21–24 (1999)Google Scholar