Title :
A study on tongue image color description based on label distribution learning
Author :
Mingjun Gui;Xinfeng Zhang;Guangqin Hu;Chunmei Zhang;Zhao Zhang
Author_Institution :
College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing, China
Abstract :
In the process of auxiliary tongue diagnosis, the color of a tongue is always labeled by a single-label. In fact, the tongue color is changeable. It is difficult to give an accurate description to the diversified tongue color for single-label learning. So in this paper, we intend to give a new group of labels to a tongue image by the Label Distribution Leaning (LDL). The value of each label is its description degree of the tongue image color. Experimental results show that a more accurate and complete description to a tongue image is given by LDL, which is beneficial to the further development of auxiliary diagnosis based on tongue color in Traditional Chinese Medicine.
Keywords :
"Tongue","Image color analysis","Coatings","Feature extraction","Training","Euclidean distance"
Conference_Titel :
Biomedical Engineering and Informatics (BMEI), 2015 8th International Conference on
DOI :
10.1109/BMEI.2015.7401490