小麦近红外特征波长提取及蛋白质含量测定

Determination of Protein Content of Wheat and Wavelength Selection of Near-Infrared Spectral Information

采用多元线性回归的逐步分析算法,对小麦粉的近红外光谱波段(Ⅰ:1000nm～1400nm;Ⅱ:1400nm～1860nm;Ⅲ:1860nm～22500nm;)进行了回归特征提取,确定了最佳回归波长,给出了各段回归的数学模型,对小麦粉蛋白质的近红外谱成分进行了比较分析,讨论了不同波段模型的蛋白质的近红外分析结果,在应用中有一定的参考价值。

Based on Stepwise linear Regression, protein content of ground wheat was analyzed and determined. Spectrum was divided into three ranges,which were Ⅰ (1000nm - 1400nm) and Ⅱ (1400nm - 1860nm) and Ⅲ (1860nm -2500nm). Regression wavelengths of different ranges were selected by stepwise. Regression coefficients were calculated by Matlab5. 3 Program. Regression Models were built up in different ranges. Determination results of protein content of ground wheat were displayed with scatter plot. Different results were discussed and compared.