论文部分内容阅读
为研究一种简便的藜麦粗淀粉含量测定方法,在10 000~4 000 cm-1波数范围内,采集100个藜麦样品的近红外光谱,运用一阶导数+矢量归一化光谱方法进行预处理,结合化学方法所得数据建立藜麦粗淀粉近红外定量模型。结果表明,该模型校正和预测效果最佳,所得粗淀粉近红外定量模型的交叉验证决定系数(r_(cv)~2)为0.914 7,外部验证决定系数(r_(val)~2)为0.903 1。由结果可知,基于近红外光谱(NIR)法测定藜麦完整籽粒的淀粉含量是完全可行的。
In order to study a simple method for determination of crude starch content in quinoa, 100 samples of quinoa collected in the range of 10 000-4 000 cm-1 were analyzed by near-infrared spectroscopy using first derivative + vector normalized spectroscopy Pretreatment, combined with the chemical methods of data obtained quinoa crude starch near infrared quantitative model. The results show that the calibration and prediction of the model are the best. The cross validation coefficients (r cv) ~ 2 of the obtained crude starch NIR quantitative model are 0.914 7 and the external validation coefficients (r val 2) are 0.903 1. From the results, it is completely feasible to determine the starch content of the intact grain of Quinoa by near infrared spectroscopy (NIR).