黄柏类药材颜色数量分类探讨

Numerical taxonomy of corlor in Phellodendron Cortex

通过收集市场上黄柏类药材的商品28批样品,利用分光测色仪测量其外表面、内表面、断面的颜色Lab色彩空间值,采用Ward联接法和平方Euclidean距离对28组数值进行聚类分析。同时邀请"国医大师"金世元教授依据传统经验对28批黄柏样品进行鉴定、质量评价和分级。将传统经验结果与分光测色仪测量分析结果进行比较。结果显示,黄柏类药材可以通过颜色数值聚类成关黄柏和川黄柏,并按照质量进行分类,且与传统经验鉴别结果具有高度一致性。

Through the investigation of Phellodendron Cortex on the market, and 28 batches of samples were collected. By using spectrophotometer the color values of outer surface, inner surface and cross - section of these samples were measured. These measured color data was translated into 3D structure diagram by using the Lab color space tool. The level difference value, the mean value and the threshold value were calculated based the measured color data of these different batches of samples. All 28 groups measured data was analyzed using the methods of Ward linkage and average Euclidean distance. At the same time, we invited Professor Jin Shiyuan, the ＂Chinese medicine master＂, to identify, quality-evaluate and grade these 28 batches of Phellodendron Cortex samples base on the traditional experience, then compared the traditional empirical results with the spectrophotometer measurement results. The result showed that, the Phellodendron Cortex couhl be divided into Phellodendri Amurensis Cortex and Phellodendri Chinensis Cortex by color numerical clustering, and classified according to quality. The classification result has a high degree of consistency with the traditional experience.