The human tongue is a unique organ that can be stuck out of the body for physical examination, and tongue diagnosis is very important in traditional Chinese medicine. Automated tongue area detection is crucial and indispensable for computer‐aided tongue diagnosis, but it is difficult to implement because of the physiological properties of the tongue. For example, as a non‐rigid organ, the tongue has a high degree of variability in size, shape, color, and texture. The purpose of this study is to address this problem.
This paper presents a hybrid framework for tongue area detection based on active shape model and genetic algorithm with the prior knowledge of tongue shape deformation.
A set of 612 tongue images was collected from both healthy and sick subjects. Using these images, the proposed method was compared with state‐of‐the‐art methods. The proposed method achieved an improvement of about 10 percent, 36 percent, and 6 percent over the existing methods in terms of mean Hausdorff distance, mean closest point distance, and Williams Index, respectively. The results demonstrate the efficacy of our proposed method in terms of both robustness and accuracy.
The proposed method gives a new approach for computer‐aided tongue diagnosis in medicine.
Liu, Z., Wang, H. and Jiang, W. (2012), "Automated tongue area detection for computer‐aided diagnosis based on ASM and GA", Sensor Review, Vol. 32 No. 1, pp. 39-46. https://doi.org/10.1108/02602281211197134Download as .RIS
Emerald Group Publishing Limited
Copyright © 2012, Emerald Group Publishing Limited