摘要
在中草药现代化的进程中,识别不同产地的中草药对于中医药的发展十分重要。利用傅里叶变换红外光谱仪测量不同产地黄芩和防风的红外光谱,选取指纹特征区域400~3500 cm^-1的红外光谱,经过预处理、归一化建立基于主成分分析的识别模型。结果表明,防风中三个主成分能代表99.99%的指纹区变量信息,黄芩中三个主成分能代表99.70%的指纹区变量信息。样本在主成分及主成分得分值空间中的分布情况,不仅实现了对不同产地防风、黄芩的识别,还可以通过样本的分布疏密度反映样本的亲缘关系。基于主成分分析原理建立的数学模型可以高效地识别不同产地防风和黄芩,为中草药鉴别提供了有效的参考。
In the modernization process of Chinese herbal medicine,it is very important to recognize the Chinese herbal medicine from different habitats.The infrared spectra of Divaricate Saposhnikovia Root and Scutellaria Baicalensis Georgi in different habitats were measured by Fourier Transform Infrared spectrometer,and the infrared spectra of 400~3500 cm^-1 in the fingerprint characteristic region were selected.The recognition model based on PCA was established through pretreatment and normalization.The results showed that three main components in Divaricate Saposhnikovia Root could represent 99.99% fingerprint region variable information,and three main components in Scutellaria Baicalensis Georgi could represent 99.7% fingerprint region variable information.The samples were aggregated into different categories in principal component and principal component score space,and the identification of Divaricate Saposhnikovia Root and Scutellaria Baicalensis Georgi in different habitats was realized.The sample sparsity and density of distribution could reflect their relationship.Mathematic model established by PCA could quickly and non -destructively identify Divaricate Saposhnikovia Root and Scutellaria Baicalensis Georgi in different habitats,which would provide an effective reference for the identification of Chinese herbal medicine.
作者
岳星宇
王乐新
赵新月
孙太凡
张欣艳
Yue Xingyu;Wang Lexin;Zhao Xinyue;Sun Taifan;Zhang Xinyan(College of Science,Heilongjiang Bayi Agricultural University,Daqing 163319)
出处
《黑龙江八一农垦大学学报》
2019年第3期45-50,71,共7页
journal of heilongjiang bayi agricultural university
基金
黑龙江省自然基金资助项目(F201427)
黑龙江省大学生创新创业项目(201810223082)
关键词
主成分分析法
中草药
红外光谱
产地识别
principal component analysis
Chinese herbal medicine
IR spectrum
origin identification
