近红外光谱和支持向量机用于凌霄花产地鉴别  

Geographical origin discrimination of Campsis grandiflora by near-infrared spectroscopy coupled with support vector machine

作　　者：王燕[1] 李颖 叶桦珍[1] 李泳宁[1] 徐杰 林振宇[4] WANG Yan;LI Ying;YE Huazhen;LI Yongning;XU Jie;LIN Zhenyu(College of Pharmacy,Fujian Health College,Fuzhou,Fujian 350101,China;College of Marine Biology,Xiamen Ocean Vocational College,Xiamen,Fujian 361102,China;Jiangsu Institute for Food and Drug Control,Nanjing,Jiangsu 210019,China;Key Laboratory Technology for Analytical Science of Food Safety and Biology,Ministry of Education,Fujian Provincial Key Laboratory of Analysis and Detection Technology for Food Safety,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]福建卫生职业技术学院药学院,福建福州350101 [2]厦门海洋职业技术学院海洋生物学院,福建厦门361102 [3]江苏省食品药品监督检验研究院,江苏南京210019 [4]福州大学食品安全与生物分析教育部重点实验室,福建省食品安全分析与检测技术重点实验室,福建福州350108

出　　处：《福州大学学报（自然科学版）》2022年第4期568-573,共6页Journal of Fuzhou University(Natural Science Edition)

基　　金：国家重点研发计划资助项目(2019YFC1604701);福建省教育厅中青年教师教育科研项目(JAT191298);福建省教育厅2018年国内访问学者资助项目。

摘　　要：收集6个产地凌霄花样品的近红外光谱,构建支持向量机(SVM)模型进行产地鉴别.运用竞争自适应重加权采样(CARS)算法提取特征波长变量,在此基础上建立CARS-SVM产地判别模型.将该判别模型与线性判别分析、偏最小二乘判别分析和簇类独立软模式法3种模型进行比较.结果表明,SVM模型对不同产地凌霄花样品的鉴别结果良好,经CARS提取特征波长后,波长变量数从1557减小至52,所构建的CARS-SVM模型对6个产地样品的判别准确率较高,明显优于上述3种模型.因此,近红外光谱技术可快速准确判别凌霄花的产地,为凌霄花的产地鉴别与质量评价提供一种新的方法.The near-infrared(NIR)spectra of 271 samples from six geographical origins were collected and analyzed in this study.Support vector machine(SVM),was used to establish a model.Competitive adaptive reweighted sampling(CARS)was applied to extract characteristic wavelength variables.The CARS-SVM model was established and compared with the linear discriminant analysis,partial least squares discriminant analysis and soft independent modelling of class analogy.The results showed that SVM model had a good discriminant effect on Campsis grandiflora from different origins.When the characteristic wavelength variables were sharply reduced from 1557 to 52 by CARS,the constructed CARS-SVM model can achieve an optimal classification accuracy which was obviously superior to that of other three models.Therefore,NIR spectroscopy can quickly and accurately identify the origin of Campsis grandiflora,and provide a certain reference for the geographical origin discrimination and quality assessment of Campsis grandiflora.

关 键 词：凌霄花 产地鉴别 CARS-SVM模型 近红外光谱 支持向量机 

分 类 号：Q949.777.9[生物学—植物学] O657.33[理学—分析化学]