[Methodological study on digitalization of tongue image in traditional Chinese medical diagnosis]

This is a research aimed at proposing a computerized tongue analysis method based on computerized image processing for quantizing the tongue properties in traditional Chinese medical diagnosis. The chromatic algorithm and 2-D Gabor wavelet transformation are applied to segmenting tongue from original image. The statistical method is adopted in identifying the colors of each pixel, which are attributed to the tongue substance and coating respectively. Thickness of tongue coating is determined by energy of 2-D Gabor wavelet coefficients (GWTE). The distribution of GWTE and invariant moment algorithm are used to judge the tongue texture. The experiment result shows that all methods proposed in this paper are effective.