基于近红外光谱的三文鱼无损真伪判别分析  

Analysis of Nondestructive Authenticity Identification of Salmon Based on Near-Infrared Spectroscopy

作　　者：巴浩冉 伍志强 戴金然 季超 赵良娟 郑文杰 BA Haoran;WU Zhiqiang;DAI Jinran;JI Chao;ZHAO Liangjuan;ZHENG Wenjie(Tianjin Key Laboratory of Animal and Plant Resistance,College of Life Sciences,Tianjin Normal University,Tianjin,300387 China)

机构地区：[1]天津师范大学生命科学学院天津市动植物抗性重点实验室,天津300387

出　　处：《科技资讯》2025年第6期236-242,共7页Science & Technology Information

基　　金：天津市科技支撑重点项目(项目编号:23YFZCSN00260)。

摘　　要：三文鱼是一种营养丰富且深受消费者欢迎的食物,目前,国内市场商品化品种主要有大西洋鲑和虹鳟。在商业利益驱使下,鲑鱼之间的掺假、替代事件时有发生,这些欺诈行为不仅损害消费者的经济利益,甚至还危害健康。因此,在日益复杂和全球化的食品供应链中开发一种快速、环保且可现场检测的方法尤为重要。针对收集的6个国家190份鲑鱼样品(其中,大西洋鲑81份,虹鳟鱼109份),按2∶1的比例分配到训练集和测试集。以11种光谱数据预处理,结合偏最小二乘判别分析(Partial Least Squares Discriminant Analysis,PLS-DA)、反向传播人工神经网络(Back Propagation Artificial Neural Network,BP-ANN)和极限学习机(Extreme Learning Machine,ELM)3种有监督模式识别方法进行判别模型构建。确定最佳建模方法为PLS-DA,在最佳潜在变量为23时,模型的预测集正确率为100%。这表明,近红外光谱与PLS-DA相结合是一种潜在的三文鱼物种现场无损快速鉴别的工具。Salmon is a nutritionally rich and popular food among consumers,with the main commercially available varieties in the domestic market being salmo salar and onchorynchus mykiss.Driven by commercial interests,salmon adulteration and substitution incidents have been frequent.These fraudulent practices not only damage consumers'economic interests but may also pose health risks.Therefore,developing a fast,environmentally friendly,and on-site detection method is particularly important in the increasingly complex and globalized food supply chain.In this study,190 salmon samples were collected from six countries(81 salmo salar and 109 onchorynchus mykiss),which were split into training and testing sets at a 2:1 ratio.Eleven spectral data preprocessing methods combined with three supervised pattern recognition techniques:Partial Least Squares Discriminant Analysis(PLSDA),Backpropagation Artificial Neural Network(BP-ANN),and Extreme Learning Machine(ELM),were used to construct discriminant models.The best modeling method was identified as PLS-DA,with the optimal number of latent variables(Lvs)being 23,achieving a prediction accuracy of 100%for the test set.This indicates that the combination of near-infrared spectroscopy and PLS-DA is a potential tool for on-site,nondestructive,and rapid identification of salmon species.

关 键 词：近红外光谱 大西洋鲑 虹鳟 真伪判别 

分 类 号：TS251.7[轻工技术与工程—农产品加工及贮藏工程]