基于红外光谱指纹和挥发性组分信息融合模型鉴别大米产地来源  

Identification of Geographical Origin of Rice Based on Fingerprint Information Fusion Model of Infrared Spectrum and Characteristic Volatile Compounds

作　　者：杜梦佳 毛波[2] 沈飞 李彭 裴斐 胡秋辉 方勇 DU Mengjia;MAO Bo;SHEN Fei;LI Peng;PEI Fei;HU Qiuhui;FANG Yong(Key Laboratory of Grains and Oils Quality Control and Processing,Collaborative Innovation Center for Modern Grain Circulation and Safety,College of Food Science and Engineering,Nanjing University of Finance and Economics,Nanjing 210023,China;College of Information Engineering,Nanjing University of Finance and Economics,Nanjing 210023,China)

机构地区：[1]南京财经大学食品科学与工程学院,江苏省现代粮食流通与安全协同创新中心,江苏高校粮油质量安全控制及深加工重点实验室,江苏南京210023 [2]南京财经大学信息工程学院,江苏南京210023

出　　处：《食品科学》2018年第8期243-248,共6页Food Science

基　　金：“十三五”国家重点研发计划重点专项(2016YFD0401203);国家农产品质量安全风险评估项目(GJFP201700102);江苏省优势学科建设工程项目(PDAD)

摘　　要：为建立一种红外光谱指纹信息和挥发性组分信息融合鉴别模型,提高模型对大米产地的鉴别率。通过傅里叶红外光谱和气相色谱-质谱联用分析20份盘锦大米、19份射阳大米和15份五常大米样品中红外光谱吸光度和挥发性组分含量,利用方差分析筛选出特征光谱和挥发性组分,结合偏最小二乘判别分析(partial least squares-discriminant analysis,PLS-DA)法建立融合这2种指纹信息的鉴别方法。结果表明,信息融合模型的大米产地鉴别准确率为97.4%,与单一光谱指纹信息模型(92.9%)和挥发性指纹信息模型(88.9%)相比,分别提高了4.5%和8.5%。因此,信息融合技术提高了该模型鉴别效果,采用PLS-DA法信息融合模型对大米产地进行鉴别是可行有效的。This study aimed to establish an accurate model based on fingerprint information fusion of characteristic volatile compounds and infrared spectrum for identifying the geographical origin of rice.A total of 20,19 and 15 rice samples respectively collected from Panjin,Sheyang and Wuchang were analyzed for their volatile compounds by gas chromatography-mass spectrometry(GC-MS)and Fourier transform infrared spectra of these samples were recorded.Analysis of variance(ANOVA)was employed to screen out the characteristic volatile components and characteristic infrared spectra,which were combined to establish a fingerprint information fusion model by partial least squares discriminant analysis(PLS-DA).The results showed that the identification accuracy of the fingerprint information fusion model was 97.4%,which was increased by 4.5%and 8.5%compared with individual infrared spectrum(92.9%)and volatile fingerprints(88.9%),respectively.Therefore,the PLS-DA information fusion model is feasible to identify the geographical origin of rice with high accuracy.

关 键 词：地理标志大米 产地鉴别 气相色谱-质谱联用 傅里叶红外光谱 偏最小二乘判别分析 

分 类 号：TS201.1[轻工技术与工程—食品科学]