Extraction of Texture Feature Parameter of Transverse Section in Chinese Herbal Medicine by Gray-Level Co-occurrence Matrix

VernacularTitle:基于灰度共生矩阵的中药饮片横切面图像纹理特征参数的提取＊
Author: Ou TAO ; Yanling ZHANG ; Qian CHEN ; Yun WANG ; Yanjiang QIAO
Publication Type:Journal Article
Keywords: Gray-level Co-occurrence Matrix; textural feature; feature parameter; Chinese herbal medicine
From: World Science and Technology-Modernization of Traditional Chinese Medicine 2014;(12):2531-2537
CountryChina
Language:Chinese
Abstract: Digitalization is an important method for the objectification and quantification of quality control on Chi-nese herbal medicine. To solve the problem of texture online visualization of Chinese herbal medicine and the estab-lishment of automatic identification method based on the texture, 12 Chinese herbal medicines were selected to cap-ture the images of their transverse sections. A total of 11 features were extracted from images using Gray-level Co-occurrence Matrix (GLCM) method. Through analyzing the influence of distances and angles between pixels on identi-fication results, it was concluded that when the distance was d = 3 and the angle was ? = 0o, the features extracted were suitable for expressing the texture of the transverse sections. The results indicated the feasibility of establishing the automatic identification method on Chinese herbal medicine based on image of transverse section. It will provide new technologies for the objectification and quantification of identification on Chinese herbal medicine.