FT-NIR在油菜籽品质指标快速检测中的应用研究

以203份不同品种的油菜籽为原料,开展了油菜籽整籽粒和粉碎样品的水分、蛋白质和粗脂肪含量的近红外研究。研究结果显示:油菜籽整籽粒和粉碎样品的水分含量近红外检测模型的内部交叉验证决定系数R~2分别为0.967和0.953,油菜籽整籽粒和粉碎样品的蛋白质含量的近红外检测模型的内部交叉验证决定系数R~2分别为0.810和0.947,油菜籽整籽粒和粉碎样品的粗脂肪含量的近红外检测模型的内部交叉验证决定系数R~2分别为0.776和0.896。同时,测定水分时可采用整籽粒扫描,而粉碎试样所建的蛋白质和粗脂肪模型优于整籽粒扫描。油菜籽粉碎样品的外部验证模型评价效果较好,外部验证水分、蛋白质和粗脂肪决定系数R~2分别为0.966、0.937和0.918。结果表明,近红外光谱技术可用于油菜籽品质指标的检测。

Applied Research on Rapid Detection of Rapeseed Quality Indicators by FT-NIR

This study was to explore the moisture, protein and fat content of whole and crushed rapeseed using FT-NIR. For this purpose, 203 samples of different varieties of rapeseed were prepared. Coefficient of determination in interned cross-validation of the moisture, protein and fat content of whole grain were 0.967, 0.810 and 0.776, while the crushed were 0.953, 0.947 and 0.896. Meanwhile, Moisture model can be established by whole grains, However, crushed grains were better for crush rapeseed. Coefficient of determination in external validation were good for crushed rapeseed, its moisture, protein and fat content determination coefficient were 0.966, 0.937 and 0.918, respectively. The results show that near infrared spectroscopy can be used to detect the quality of rapeseed.