多叶棘豆产地溯源的一种无损快速检测方法研究

Study on a Non-Destructive Rapid Detection Method for Origin Torcing of Oxytropis Myriophylla

多叶棘豆中含有的黄酮类是其发挥治疗效果的关键活性物质,多叶棘豆没有亚种,因此其成分主要与地理产地有关。为了追溯多叶棘豆的地理起源,利用一种无损快速的检测手段——激光诱导击穿光谱(LIBS),采集多叶棘豆枝干的光谱,然而不同产地枝干的光谱特征几乎重叠,难以鉴别。借助主成分分析(PCA)和支持向量机(SVM)2种多元统计方法对光谱进行分析,进而实现对未知类别数据的归类。选取内蒙古4个不同地区的多叶棘豆枝干共76个样品,实验利用1064 nm激光烧蚀多叶棘豆枝干,采集220~576 nm波段的76组光谱数据,先用PCA对光谱数据进行预处理,由主成分的得分图得出不同地区样本重叠严重。然后采用SVM对数据进一步处理,将提取的30条多叶棘豆枝干的LIBS特征谱线作为SVM的输入变量,SVM模型的识别效果良好。SVM的产地识别准确率可达83.33%。结果表明,利用LIBS技术结合多元统计方法可实现多叶棘豆来源鉴定。

The flavonoids contained in Oxytropis multifolia are the key active substances for its therapeutic effect.Oxytropis myriophylla has no subspecies,so its composition is mainly related to its geographical origin.In order to trace the geographical origin of Oxytropis myriophylla,a non-destructive and rapid detection method,laser induced breakdown spectroscopy(LIBS),was used to collect the spectrum of the branches of Oxytropis myriophylla.However,the spectral characteristics of branches from different places almost overlap and are difficult to identify.In this paper,two multivariate statistical methods,principal component analysis(PCA)and support vector machine(SVM),are used to analyze the spectrum,and then the unknown category data is classified.A total of 76 samples of P.multiphyllum branches were collected from 4 different regions in Inner Mongolia.The experiment used 1064 nm laser ablation to collect 76 sets of spectral data in the band of 220~576 nm.The spectral data were preprocessed by PCA first,and the scores of principal components showed that samples in different regions overlapped seriously.Then SVM was used to further process the data,and the LIBS characteristic spectral lines extracted from 30 branches of Oxytropis myriophylla were used as input variables of support vector machine.The SVM model had a good recognition effect.The accuracy of origin identification by SVM can reach 83.33%respectively.The results showed that the LIBS technique combined with multivariate statistical method could be used to identify the source of Oxytropis myriophylla.