Comparison of Edge Segmentation Methods to Tongue Diagnosis in Traditional Chinese Medicine

In Traditional Chinese Medicine, the human tongue is one of the important organs which contain the information of health status. In order to achieve an automatic tongue diagnostic system, an effective segmentation method for detecting the edge of tongue is very important. We mainly compare the Canny, Snake and threshold (Otsu's thresholding algorithm) methods for edge segmentation. The segmentation using Canny algorithm may produce many false edges after cutting; thus, it is not suitable for use. The Snake segmentation in tongue requires a larger convergence number and spends too much time. The threshold method using Otsu's thresholding algorithm and filtering process can achieve an easy, fast and effective segmentation result in tongue diagnosis. Therefore, it may be useful in clinical automated tongue diagnosis system.

[1]  David Zhang,et al.  The bi-elliptical deformable contour and its application to automated tongue segmentation in Chinese medicine , 2005, IEEE Transactions on Medical Imaging.

[2]  Bing Wang,et al.  An Improved CANNY Edge Detection Algorithm , 2009, 2009 Second International Workshop on Computer Science and Engineering.

[3]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[4]  Issam Dagher,et al.  WaterBalloons: A hybrid watershed Balloon Snake segmentation , 2008, Image Vis. Comput..

[5]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  N. Otsu A threshold selection method from gray level histograms , 1979 .

[7]  David Zhang,et al.  Tongue image analysis for appendicitis diagnosis , 2005, Inf. Sci..