摘要
目的基于仿生识别系统,建立金银花"颜色"与"气"数字化分析方法,并建立金银花的电子鼻判别模型。方法针对"颜色""、气"两方面,采用传统性状鉴别法对金银花样品进行感官评价,再分别利用仿生识别系统——色度仪和电子鼻进行测定;采用判别因子分析(DFA)、正交偏最小二乘判别分析法(OPLS-DA)等进行数据分析并建立判别模型。结果不同金银花样品"颜色与气"差异显著,性状鉴别法和仿生识别系统的鉴别结果基本一致。基于电子鼻MOS传感器对金银花的响应值所建立的DFA判别模型正判率较高,能给出正确的识别。结论仿生识别系统(色度仪和电子鼻技术)能将传统性状鉴别中的经验性、主观性感官评价具体化、数字化,且电子鼻能用于区分不同产地及处理方式、储藏期的金银花;所建立的判别模型可为指导中药材鉴别实践工作提供参考。
Objective To explore a digital analysis method of Lonicera japonica Thunb.based on the technologies of bionic recognition system,taking color and odor as the targeted indicators.And establish a discriminant model based on response values of MOS sensor array in electronic nose.Methods Two characteristics,namely color and odor,of Lonicera japonica Thunb.were screened out and described via macroscopic identification method on human sensory panel.Accordingly,bionic recognition system,including colorimeter and electronic nose,were introduced and utilized to determine the color and odor of the samples,respectively.Discriminant factor analysis(DFA)and orthogonal partial least square discriminant analysis(OPLS-DA)were applied for data analyses and discriminant model formations.Results Classification results of Lonicera japonica Thunb.samples of sensory evaluation were generally in agreement with that of bionic recognition systems.The positive judgment rate of DFA discriminant model was higher,which could offer better identification.Conclusion Bionic recognition system could help us digitalize the vague linguistic descriptions of subjective experience,which could be effectively applied in practical identification of Chinese Material Medica.
作者
姚月保
杨晓芸
彭莲
俞越童
邵远洋
邹慧琴
闫永红
Yao Yuebao;Yang Xiaoyun;Peng Lian;Yu Yuetong;Shao Yuanyang;Zou Huiqin;Yan Yonghong(School of Chinese Materia Medica,Beijing University of Chinese Medicine,Beijing 100029,China;Institute of Research,KPC Pharmaceuticals,Inc,Kunming 650105,China;Department of Pharmacy,The Third Affiliated hospital of Sun Yat-sen University,Guangzhou 510630,China)
出处
《世界科学技术-中医药现代化》
CSCD
北大核心
2020年第4期1270-1277,共8页
Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology
基金
国家自然科学基金委员会面上基金项目(81573542):基于形态特征、药效物质及药理作用变化规律采用电子鼻建立苦杏仁“走油”预警机制及质量实时监测系统,负责人:闫永红
国家自然科学基金委员会青年基金项目(81403054):电子鼻技术快速无损评鉴阳春砂“真伪优劣”的基础研究,负责人:邹慧琴
关键词
仿生识别系统
色度仪
电子鼻
金银花
判别因子分析
正交偏最小二乘判别分析法
Bionic recognition system
colorimeter
electronic nose
Lonicera japonica Thunb
discriminant factor analysis
orthogonal partial least square discriminant analysis
