A new pattern recognition algorithm and its application about tongue fur image classification of Traditional Chinese Medicine

In this paper, a new classification method of tongue fur image based on self-organizing map is presented. The method firstly designed a template to help automatically extract the image feature region of tongue fur, then extracted the color characteristics of the feature region using L*a*b* color model. Finally the training of a self-organizing feature map neural network was implemented for the automatic classification of tongue fur images. The new method provides a new channel of thinking on the tongue fur image classification. The application result in make clear that, the new algorithm this article expatiated can be faster and accurate more.