基于近红外漫反射光谱的西红花总灰分无损检测研究

Rapid Determination of Total Ash in Saffron by Near Infrared Diffuse Reflectance Spectroscopy

关键词:  近红外漫反射光谱  西红花  总灰分  光谱预处理  偏最小二乘回归

Key words:  near-infrared diffuse reflectance spectroscopy  saffron  total ash  spectral preprocessing  partial least squares regression

摘要:

目的 采用近红外漫反射光谱法建立西红花中总灰分近红外模型,为实现西红花总灰分快速检测奠定技术基础。方法 以220批西红花样品为研究对象,测定其总灰分含量,并采集近红外光谱数据。采用多种光谱预处理方法包括标准正态变量变换(SNV)、多元散射校正(MSC)、去趋势处理(DT)、Savitzky-Golay平滑与一阶导数及均值中心化对原始光谱进行处理,结合偏最小二乘法(PLS)建立总灰分的定量校正模型。通过比较不同预处理方法组合模型的性能,筛选最优预处理,并利用测试集验证模型的预测准确性。结果 标准正态变量变换(SNV)结合去趋势校正(DT)、Savitzky-Golay平滑、Savitzky-Golay导数和均值中心化模型与多元散射校正(MSC)结合Savitzky-Golay平滑、Savitzky-Golay导数和均值中心模型均为最佳光谱预处理方法,在此基础上构建的西红花总灰分近红外模型表现出良好的预测能力。结论 所建近红外模型可实现西红花总灰分的快速、无损与准确测定,适用于该药材质量的实时监控与分析。

Abstract:

OBJECTIVE This study aimed to establish a near-infrared (NIR) model for the rapid determination of total ash in saffron using near-infrared diffuse reflectance spectroscopy (NIRS), providing a technical foundation for fast quality monitoring. METHODS A total of 220 batches of saffron samples were used to determine total ash content and acquire NIR spectra. Multiple spectral preprocessing methods, including standard normal variate (SNV), multiplicative scatter correction (MSC), detrending (DT), Savitzky–Golay smoothing, first derivative, and mean centering, were applied to the raw spectra. A quantitative calibration model for total ash was developed using partial least squares (PLS) regression. The performance of models under different preprocessing combinations was compared to identify the optimal strategy. The predictive accuracy of the model was further validated using test set. RESULTS The results indicated that both the combination of Standard Normal Variate (SNV) with Detrending (DT), Savitzky-Golay smoothing, Savitzky-Golay derivatives, and mean centering, as well as the combination of Multiplicative Scatter Correction (MSC) with Savitzky-Golay smoothing, Savitzky-Golay derivatives, and mean centering, were identified as the optimal spectral preprocessing methods. Based on these approaches, the near-infrared calibration model developed for total ash content in saffron demonstrated strong predictive performance. CONCLUSION The developed NIRS model allows for rapid, non-destructive, and accurate determination of total ash in saffron, making it suitable for real-time quality control and analysis of saffron.