基于近红外光谱技术的茶叶新旧鉴别及产地溯源研究

目的 建立基于近红外光谱的定性分析模型,实现对茶叶的新旧分类和产地溯源。方法 首先采用傅立叶近红外光谱仪采集茶叶样品的漫反射光谱数据,然后使用卷积(Savitzky-Golay,S-G)平滑算法和数据标准化(Normalization)对光谱数据进行预处理,最后基于遗传优化算法(genetic algorithem,GA)和粒子群优化算法(particle swarm optimization, PSO)分别建立了优化向量机模型(support vector machine, SVM),从而实现新旧茶叶的分类以及产地溯源。结果 与GA-SVM模型相比,PSO-SVM模型的建模效果较好,且分类时间更短,在新旧鉴别和产地溯源实验中都达到了100%的预测精度。结论 基于近红外光谱建立的PSO-SVM模型可以实现茶叶新旧的判别以及产地溯源,为鉴别茶叶年份和追踪茶叶产地提供了理论支撑和技术指导。

Identification of new and old tea and origin traceability based on near infrared spectroscopy

Objective To establish the qualitative analysis model by near infrared spectroscopy, realize the new and old classification and the origin traceability of the tea. Methods Firstly, Fourier near-infrared spectrometer as used to acquire diffuse reflectance spectral data of tea samples, and then the spectral data was pretreated using Savitzky-Golay (S-G) smoothing and data Normalization. Finally, based on genetic optimization algorithm (GA) and particle swarm optimization (PSO), support vector machine (SVM) models were established, respectively. Thus to realize the classification of old and new tea and origin traceability. Results Compared to GA-SVM model, PSO-SVM model had better modeling effects and shorter classification times, which displayed as 100% prediction accuracy in both the old and new identification and origin traceability experiments. Conclusion The PSO-SVM model set by the near infrared spectroscopy can realize the new and old classification and the origin traceability of the tea, which provides theoretical support and technical guidance for the identification of tea years and traceability of tea origin.