摘要
利用傅里叶近红外光谱仪（FT-NIR）对金沙蓝钻酒样品进行了光谱采集,采用一阶求导与Savitzky-Golay卷积平滑相结合的预处理方法结合仪器自动优化谱区（4119.20～9881.46 cm-1）建立了金沙蓝钻酒的相似度匹配分析模型,同时采用该模型对已知样品进行相似度计算。结果显示：蓝钻酒样品的相似度均达到95%以上,而其他酒样的相似度均在60%以下,其中赖茅相似度最高,达到了59.24%;在所有乙醇水溶液样品中,体积分数为50%的乙醇水溶液相似度最高,为18.23%。实验中,验证样品全部判别准确,准确率达到100%,表明经过优化的模型预测精度和可靠性较高,可用于实际检测。
In this paper, the spectrum of Sands blue diamond wine was collected by Fourier near-infrared spectrometer（FT-NIR）. A similarity matching model of the wine was established by using the pretreatment of a first order derivative and Savitzky-Golay smoothing convolution combing with automatical optimized spectra region（4119.20～9881.46 cm^-1）. The results of the known samples calculating from the model showed that the similarity of Sands blue diamond wine samples were above 95%, while other wine samples were below 60%, among which Lai Mao is the highest which reached 59.24%; the similarity of ethanol aqueous whose volume fraction is 50 reached 18.23%, which is the highest among all the ethanol aqueous samples. In this experiment, all the samples were verified to correct and the accuracy reached 100%, the high prediction and reliability implied the model was suitable for actual testing.
出处
《食品科技》
CAS
北大核心
2014年第11期315-318,共4页
Food Science and Technology
基金
国家重大科学仪器设备开发专项项目(2011YQ12003506)
贵州省科技计划项目(黔科合NY字[2008]3050号)
关键词
近红外光谱
金沙蓝钻酒
相似度
模型
near-infrared spectrometer
Sands blue diamond wine
similarity
model
