基于近红外光谱无损快速检测面粉品质的研究

提出一种基于近红外光谱技术结合偏最小二乘法对面粉品质进行无损快速检测的方法.配制含滑石粉的面粉样品30个,采集样品在12 500~4 000 cm-1范围内的近红外漫反射光谱,选择最优的光谱预处理方法和光谱范围,采用偏最小二乘法(PLS)建立定量分析模型.结果表明所建定量分析模型的相关性能比较高,预测相关系数和预测均方根误差均符合要求.研究发现,近红外光谱技术用于快速无损检测面粉掺假是可行的.

Research of Rapid and Undamaged Assessment of Flour Quality Using Near Infrared Spectroscopy

A rapid non-destructive method for detecting flour quality was presented using near-infrared spectroscopy combined with partial least squares method. Thirty flour samples containing talc were prepared, and near-infrared diffuse reflectance spectrograms of the samples were collected in the 12 500 -4 000 cm- l range. A quantitative analysis model was established by selecting the optimal spectral pretreatment methods and spectral range, and using partial least squares （PLS）. The quantitative analysis model showed a relatively high correlation, and the correlation coefficients and predicted root mean square error could met the requirements. The results indicated that the near-infrared spectroscopy was feasible for rapid non-destructive testing adulteration of flour.