摘要
该文采用近红外光谱技术与化学计量学方法结合实现贝类毒素无损鉴别。该研究以新鲜翡翠贻贝为研究对象,使用近红外光谱仪采集健康贻贝和感染腹泻性毒素贻贝的反射光谱数据,利用Savitzky-Golay卷积平滑求导结合标准正态变量变换光谱预处理方式消除光谱中的干扰因素,采用间隔影响分析(margin influence analysis,MIA)结合连续投影算法(successive projections algorithm,SPA)对数据进行降维处理,应用偏最小二乘线性判别分析(partial least squares linear discriminant analysis,PLS-LDA)方法构建贝类毒素鉴别模型,并与支持向量机和随机森林分析模型进行比较。结果表明,采用MIA-SPA-PLS-LDA方法,可实现贝类毒素的100%鉴别。为此,利用MIA-SPA-PLS-LDA方法可建立准确的贝类毒素鉴别模型,为贝类毒素的快速鉴别提供了新途径,也为后续各种贝类水产品的毒素鉴别分析提供了参考。
The consumption of mussels contaminated with diarrheal shellfish poisons can cause adverse effects on human health.It is necessary to identify mussels contaminated with poisons.To explore a new method for the identification of shellfish poisons,in this paper,the nondestructive identification of shellfish poisons was realized by combining near-infrared spectroscopy and chemometrics.Taking the fresh mussels as the research object,the near-infrared spectrometer was used to collect the reflection spectrum data of healthy mussels and mussels contaminated with diarrhea shellfish poisons.The spectral preprocessing method combining Savitzky-Golay convolution smooth derivation and standard normal variable transformation was used to eliminate the interference factors in the spectrum.Margin influence analysis(MIA)combined with succession projections algorithm(SPA)was used to reduce the dimension of the data.The identification model of diarrhea shellfish poisons was constructed by the partial least squares linear discriminant analysis(PLS-LDA)method and compared with the identification models by support vector machine and random forest.Results showed that using the MIA-SPA-PLS-LDA method,100%identification of shellfish toxins could be achieved.Results showed that the MIA-SPA-PLS-LDA method could make the identification accuracy of diarrhea shellfish poisons reach 100%.Therefore,the MIA-SPA-PLS-LDA method can be used to establish an accurate identification model of diarrhea shellfish poisons,which provides a new way for the rapid identification of diarrhea shellfish poisons,and also provides a reference for the subsequent identification and analysis of toxins in various shellfish products.
作者
姜微
刘瑶
刘忠艳
曾绍庚
熊建芳
乔付
JIANG Wei;LIU Yao;LIU Zhongyan;ZENG Shaogeng;XIONG Jianfang;QIAO Fu(School of Computer Science and Intelligence Education,Lingnan Normal University,Zhanjiang 524048,China;School of Electronic and Electrical Engineering,Lingnan Normal University,Zhanjiang 524048,China)
出处
《食品与发酵工业》
CAS
CSCD
北大核心
2023年第2期271-279,共9页
Food and Fermentation Industries
基金
国家自然科学基金(62005109)
广东省基础与应用基础研究基金(2020A1515011368,2021A1515012440)
岭南师范学院人才专项(ZL2049)
广东省教育科学规划课题(2022GXJK256)
2022年度岭南师范学院科研创新团队项目(XJ2022000501)
岭南师范学院红树林研究院开放项目(PYXM04)。
关键词
近红外光谱
翡翠贻贝
间隔影响分析
腹泻性贝类毒素
偏最小二乘线性判别分析
near-infrared spectroscopy
mussels
margin influence analysis
diarrheal shellfish poisons
partial least squares linear discriminant analysis
