基于傅里叶变换红外光谱的姬松茸存储时间检测模型研究

Study on Storage Time Detection Model of Agaricus Blazei Murrill Mushroom Based on Fourier Transform Infrared Spectroscopy

利用傅里叶变换红外光谱技术建立姬松茸存储时间的检测模型,并利用BiPLS和SiPLS两种特征区间提取算法改进模型。结果显示,相较于传统的全光谱数据,使用特征区间提取算法能够提高模型的预测性能,使用BiPLS改进的傅里叶变换红外光谱PLS模型在训练集和测试集中的结果最优。工作表明,基于傅里叶变换红外光谱技术建立的PLS模型是检测姬松茸存储时间的一种有效方法。

The storage time detection model of Agaricus blazei Murrill mushroom was established by Fourier transform infrared spectroscopy,and the model was improved by BiPLS and SiPLS.The results showed that the PLS model based on Fourier transform infrared spectroscopy was an effective method to detect the storage time of Agaricus blazei Murrill mushroom.Compared with the traditional whole spectra,the feature interval extraction algorithms could improve the performance of the model,and the PLS model with Fourier transform infrared spectroscopy improved by BiPLS had the best results in the training set and test set.