傅里叶变换近红外技术在乳制品微生物鉴别中的应用研究

发布时间：2018-02-20 01:06

本文关键词： FT-NIR PLS-DA 阪崎肠杆菌金葡萄球菌大肠杆菌牛奶　出处：《光谱学与光谱分析》2016年S1期 　论文类型：期刊论文

【摘要】：乳制品的质量安全问题受到广泛的关注。为快速、准确判定乳制品污染源,利用傅里叶变换近红外光谱技术采集被阪崎肠杆菌、金葡萄球菌、大肠杆菌三种致病菌污染的牛奶样品的近红外透射光谱(NP),使用一阶求导(FD),标准正态变量变换(SNV),多元散射校正(MSC)对光谱进行预处理,结合偏最小二乘判别分析(partial least squares-discriminant analysis,PLS-DA)对三种细菌判别的可行性进行探究。研究表明,利用FD与SNV预处理方法得到PLS-DA模型判别结果均比NP差,而利用MSC预处理后准确率与NP一致达到了100%,且预测集相关系数(Rp)比NP高,预测均方根误差(RMSEP)更小,表明模型在经过MSC预处理后预测性能更理想。阐明利用FT-NIR技术结合化学计量学方法经过合适预处理方法后能有效用于乳制品中微生物类别的鉴别。
[Abstract]:In order to determine the pollution sources of dairy products quickly and accurately, Enterobacter sakazakii and Staphylococcus aureus were collected by Fourier transform near infrared spectroscopy (FT-NIR). The near infrared transmission spectrum (NIR) of milk samples contaminated by three kinds of pathogenic bacteria of E. coli was pretreated by first order derivation, standard normal variable transform (SNV) and multivariate scattering correction (MSC). Combining partial least squares-discriminant analysis with partial least squares discriminant analysis (PLS-DA), the feasibility of three kinds of bacteria discrimination was explored. The results showed that the results of PLS-DA model were worse than that of NP by using FD and SNV pretreatment methods. The accuracy of preprocessing with MSC is 100% consistent with NP, and the correlation coefficient of prediction set (RP) is higher than that of NP, and the root-mean-square error (RMSEP) of prediction is lower than that of NP. It is shown that the model can be used to predict the performance of dairy products after pretreatment with MSC, and it is indicated that FT-NIR combined with chemometrics can be used to identify the microorganism in dairy products effectively.
【作者单位】：福州大学电气工程与自动化学院;福州大学生物科学与工程学院;福建医科大学医学技术与工程学院;福建省出入境检验检疫局检验检疫技术中心;
【基金】：福建省自然科学基金项目(2016J01771)资助
【分类号】：O657.33;TS252.7

【相似文献】