近红外漫反射光谱法快速鉴别不同连翘饮片

目的:研究利用近红外光谱技术结合判别分析法对不同连翘饮片进行定性鉴别分析。方法:采集不同连翘饮片的近红外光谱图,选择多元散射校正法(MSC)加SG平滑方法对光谱进行预处理,选择最佳建模波段为4 065.57~8 350.90 cm-1,最佳主成分数为10,建立定性分析模型,并用未知样品验证该模型。结果:表明所建定性模型的分类结果与实际所属一致,验证结果良好。结论:该方法操作简便,测定快速,结果准确,无污染,对中药材、烟草等多种行业质量检测过程中的能源节约与资源的高效利用均有重要意义。

Fast Qualitative Identification of Different Forsythia Fructus by Near Infrared Diffuse Reflectance Spectrometry

Objective: To conduct qualitative identification analysis of different Forsythia Fructus with use of near infrared spectroscopy combined with discriminant analysis method. Method: Near infrared spectrograms of different Forsythia Fructus were collected, then multiplicative scatter correction method (MSC) and Savitzky-Golay (SG) smoothing method were used for spectral preprocessing. Qualitative analysis model was established under the optimal modeling wave band of 4065.57-8350.90 cm^-1 and optimal number (10) for principal components, and then this model was validated with unknown samples. Result: The classification results of the qualitative model were consistent with the actual conditions, and the validation results were also satisfactory. Conclusion: This method is simple, rapid, accurate, and non-polluting, with great significance for resources saving and efficient use of energy during the quality inspection in Chinese herbs, tobacco and other industries.