基于傅立叶变换红外光谱分析的3种石首鱼科鱼类品种区分  

Identification of Three Speciesof Sciaenidae Based on Fourier Transform Infrared Spectroscopy

作　　者：王世光 杨梦琼 杨盈悦 梅光明[1] 张小军[1] 黄丽英[1] 顾捷[1] WANG Shiguang;YANG Mengqiong;YANG Yingyue(Zhejiang Marine Fisheries Research Institute,Zhoushan 316021;School of Food and Pharmacy of Zhejiang Ocean University,Zhoushan 316022,China)

机构地区：[1]浙江省海洋水产研究所,浙江舟山316021 [2]浙江海洋大学食品与药学学院,浙江舟山316022

出　　处：《浙江海洋大学学报（自然科学版）》2024年第4期327-334,共8页Journal of Zhejiang Ocean University：Natural Science

基　　金：国家重点研发计划“蓝色粮仓科技创新”重点专项项目(2020YFD0900900);2024年浙江省科技厅科研院所扶持专项项目(HYS-CZ-202406)。

摘　　要：为实现鲈形目石首鱼科中大黄鱼、小黄鱼和黄姑鱼的品种快速准确区分,应用傅立叶变换红外光谱仪对大黄鱼、小黄鱼和黄姑鱼肌肉样品在4000~650 cm-1范围内的近中红外光谱吸收特征进行了测定,基于特征波段下的光谱吸收差异并结合主成分分析(principal component analysis,PCA)、聚类分析(cluster analysis,CA)和线性判别分析(linear discriminant analysis,LDA)、支持向量机(support vector machine,SVM)数据建模对3种石首鱼类样品进行品种区分。结果表明:3种鱼类肌肉样本在特征波段1800~650 cm-1下的红外吸光谱选择一阶导数(first derivative,1st Der)+标准正态变换(standard normal variate transformation,SNV)和1st Der+多元散射校正(multiplicative scatter correction,MSC)2种方式预处理后进行PCA,3种鱼样本间彼此间距远、无重叠;特征波段1800~650 cm-1下的红外光谱数据经1st Der结合SNV预处理后进行聚类分析时各自聚为一类;特征波段1800~650 cm^(-1)下的红外光谱经过1st Der、SNV、MSC预处理建立的LDA模型和经过SNV、MSC预处理建立的SVM模型测试集准确率均为100%。研究结果表明利用近中红外光谱吸收特征差异,结合化学计量学分析手段能够有效对大黄鱼、小黄鱼和黄姑鱼进行品种判别,为水产品真实性鉴别提供技术参考。In order to quickly and accurately distinguish the species of Larimichthys crocea,L.polyactisand Nibea albiflorain Sciaenidae,Fourier transform infrared spectrometer was used to determine the near-mid infrared spectral absorption characteristics of muscle samples of the three species of Sciaenidae in the range of 4000-650 cm-1.Based on the spectral absorption differences under the characteristic bands,combined with principal component analysis(PCA),cluster analysis(CA)and data modeling,the three species of Sciaenidae were distinguished.The results showed as follows:the infrared absorption spectra of the three species of Sciaenidae at the characteristic band of 1800-650 cm-1 were selected.And the PCA results after pretreatment with 1st Der+SNV and 1st Der+MSC showed that the three species of Sciaenidae were far apart from each other and did not overlap;The infrared spectral data of the three species of Sciaenidae at the characteristic band of 1800~650 cm^(-1)were clustered into one group after 1st Der(first-order derivation,1st Der)combined with SNV(standard normal variety,SNV)pretreatment.The LDA(linear discriminant analysis,LDA)model established by 1st Der,SNV or MSC(multiple scattering correction,MSC)pretreatment and the SVM(support vector machines,SVM)model established by SNV or MSC pretreatment were all with the accuracy of 100%.The results showed that the use of near-infrared and mid-infrared spectral absorption characteristics combined with chemometric analysis methods can effectively identify the three species of Sciaenidae and provide technical reference for the authenticity identification of aquatic products.

关 键 词：石首鱼 傅立叶变换红外光谱 主成分分析 聚类分析 品种区分 

分 类 号：O657.33[理学—分析化学] S917.4[理学—化学]