近红外漫反射光谱法快速检测谷子蛋白质和淀粉含量

建立近红外漫反射光谱法测定谷子中的蛋白质和淀粉含量,提供一种快速、简便、无损的分析方法进行谷子品种资源鉴定和筛选。以191份山西核心谷子种质为材料,采用近红外漫反射光谱法建立谷子蛋白质和淀粉含量的快速检测模型。结果表明,采用一阶导数+矢量归一化光谱预处理,分别建立谷子蛋白质和淀粉含量的校正模型,模型的校正决定系数(R~2_(cal))分别为0.977 0和0.907 3,交叉验证均方根误差分别为0.203%和0.466%,外部验证决定系数(R~_(2val))分别为0.989 6和0.977 2,预测均方根误差分别为0.225%和0.368%。对于谷子蛋白质和淀粉的预测,化学法和近红外仪器法测定间无显著差异,近红外测定结果是准确可靠的。说明采用近红外漫反射光谱分析技术能够满足对谷子蛋白质和淀粉含量的检测。

Application of Near Infrared Diffuse Reflectance Spectroscopy in Rapid Detection of Crude Protein and Starch in Foxtail Millet

This study aimed to establish a near infrared (NIR) diffuse reflectance spectrometry method for measuring the contents of crude protein and starch in foxtail millet for providing a fast, simple and non-destructive analytical method for the identification and screening of foxtail millet germplasm resources. For this purpose, a total of 191 samples of a foxtail millet core collection in Shanxi province were measured using NIR diffuse reflectance spectroscopy. The results showed that with the spectral preprocessing method of first derivative + vector normalization, calibration models were established to predict the contents of crude protein and starch in foxtail millet. The corresponding coefficients of determination for calibration (R2cal) were 0.977 0 and 0.907 3, and the root mean square errors of cross-validation (RMSECV) were 0.203% and 0.466%, the coefficients of determination for external validation (R2val) were 0.989 6 and 0.977 2, and the root mean square errors of prediction (RMSEP) were 0.225% and 0.368%, respectively. For the detection of crude protein and starch foxtail millet, no significant differences were seen between chemical analysis and NIR measurement, but the results obtained with NIR measurement were more accurate and reliable. NIR diffuse reflectance spectrometry technology can be useful for the detection of crude protein and starch in foxtail millet.