Article: Tongue coating thickness level classification using local binary pattern features in TCM image diagnosis Journal: International Journal of Reasoning-based Intelligent Systems (IJRIS) 2026 Vol.18 No.18 pp.53 - 67 Abstract: In traditional Chinese medicine tongue diagnosis, the thickness of the tongue coating is a key indicator for assessing the severity of a pathological condition. However, traditional methods rely on the clinician's visual observation, which is highly subjective and difficult to quantify. This paper proposes an improved local binary pattern operator - the mean local binary pattern. This operator replaces the central pixel with the mean grey value of the neighbourhood for encoding, thereby reducing sensitivity to image noise and local fluctuations. Experimental results show that, when combined with a random forest classifier, this method achieves a classification accuracy of 91.56%, representing a 5.45 percentage point improvement over the standard method. This approach provides a lightweight and interpretable textural analysis method for the objectification of tongue diagnosis in traditional Chinese medicine. Inderscience Publishers - linking academia, business and industry through research

Title: Tongue coating thickness level classification using local binary pattern features in TCM image diagnosis

Authors: Siqi Fang

Addresses: College of Health Sciences, Shandong University of Traditional Chinese Medicine, Jinan, 250000, China

Abstract: In traditional Chinese medicine tongue diagnosis, the thickness of the tongue coating is a key indicator for assessing the severity of a pathological condition. However, traditional methods rely on the clinician's visual observation, which is highly subjective and difficult to quantify. This paper proposes an improved local binary pattern operator - the mean local binary pattern. This operator replaces the central pixel with the mean grey value of the neighbourhood for encoding, thereby reducing sensitivity to image noise and local fluctuations. Experimental results show that, when combined with a random forest classifier, this method achieves a classification accuracy of 91.56%, representing a 5.45 percentage point improvement over the standard method. This approach provides a lightweight and interpretable textural analysis method for the objectification of tongue diagnosis in traditional Chinese medicine.

Keywords: thickness of the tongue coating; local binary pattern; LBP; traditional Chinese medicine; TCM; tongue diagnosis.

DOI: 10.1504/IJRIS.2026.155470

International Journal of Reasoning-based Intelligent Systems, 2026 Vol.18 No.18, pp.53 - 67

Received: 27 Apr 2026
Accepted: 22 May 2026
Published online: 03 Aug 2026 *

Title: Tongue coating thickness level classification using local binary pattern features in TCM image diagnosis

Keep up-to-date