近红外光谱法结合模式识别技术快速无损鉴别天然牛黄、体外培育牛黄和人工牛黄

Fast and non-destructive identification of Bovis Calculus,Bovis Calculus Sativus and Bovis Calculus Artifactus by near infrared spectroscopy combined with pattern recognition technology

目的:采用近红外光谱技术建立天然牛黄、体外培育牛黄和人工牛黄的快速无损鉴别方法。方法:采集天然牛黄、体外培育牛黄和人工牛黄的近红外漫反射光谱,比较其光谱差异,利用主成分分析法观察不同样品主成分空间分布,进一步优化光谱预处理方式,采用判别分析法建立定性鉴别模型。结果:3类牛黄原始光谱存在差异;其主成分空间分布明显不同,天然牛黄、体外培育牛黄和人工牛黄各自单独聚为一类;判别分析模型预测准确率为100%。结论:近红外光谱结合模式识别技术可有效区别不同种类牛黄,为天然牛黄、体外培育牛黄和人工牛黄的鉴别提供一种新的参考方法。

Objective：To develop a rapid and non-destructive method for identification of Bovis Calculus（natural Calculus Bovis）,Bovis Calculus Sativus（in-vitro cultured Calculus Bovis）and Bovis Calculus Artifactus（artificial Calculus Bovis）using near infrared（NIR）spectroscopy.Methods：Diffuse NIR spectra of Bovis Calculus,Bovis Calculus Sativus and Bovis Calculus Artifactus were collected.Then principal component analysis（PCA）was applied to observe the spatial distribution of different samples.Next discriminant analysis（DA）was adopted to classify samples based on optimization of the spectral pretreatment methods.Results：Difference was observed in original NIR spectra of 3 kinds of samples.Bovis Calculus,Bovis Calculus Sativus and Bovis Calculus Artifactus clustered to their own group in principal component spatial distribution.The classification accuracy of the DA model was 100%.Conclusion：It is demonstrated that near infrared spectroscopy combined with pattern recognition technology could effectively distinguish 3 different kinds of Bovis Calculus samples and thus offer a new identification method for identification of Bovis Calculus,Bovis Calculus Sativus and Bovis Calculus Artifactus.