Comparison of edge segmentation methods to tongue diagnosis in traditional chinese medicine

2Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

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. © 2010 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Wang, C. H., Wei, C. C., & Li, C. H. (2010). Comparison of edge segmentation methods to tongue diagnosis in traditional chinese medicine. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6422 LNAI, pp. 232–238). https://doi.org/10.1007/978-3-642-16732-4_25

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free