基于近红外光谱技术与化学计量学的绿茶无损鉴别方法研究

该文利用近红外光谱技术结合化学计量学方法开发了不同品种绿茶的无损鉴别方法。通过近红外光谱技术得到了8个品种绿茶样品的近红外光谱,比较了单一以及优化组合光谱预处理方法对光谱的影响,利用无监督的主成分分析(PCA)与有监督的线性判别分析方法(LDA)分别构建了茶叶品种鉴别模型。结果表明:对比单一预处理方法,优化组合预处理具有更优的鉴别准确性。标准正态变量变换预处理消除了茶叶样品大小不均造成的光谱散射影响,一阶导数预处理实现了变动背景的消除,减少了基线漂移的影响,突出了图谱中的有效信息,采用二者相结合的预处理方式并结合无监督的主成分分析法可实现较为准确的绿茶样品种类鉴别分析,准确率达75.0%。此外,采用有监督的线性判别分析方法处理原始光谱数据,可达到100%的鉴别准确率,但该方法需提供类别的先验知识。因此,采用近红外光谱技术和化学计量学相结合的手段可实现不同品种绿茶的快速无损鉴别。

A Nondestructive Method Identifying Varieties of Green Tea Based on Near Infrared Spectroscopy and Chemometrics

In this paper,a nondestructive method was developed for the identification of different varieties of green tea by near infrared spectroscopy with chemometrics.The near infrared spectra of eight varieties of green tea samples were collected.Effects of single and optimized combination preprocessing methods on spectra were compared.The identification models were constructed by unsupervised principal component analysis(PCA) and supervised linear discriminant analysis(LDA).Results showed that the optimized combination preprocessing method achieved higher accuracy.The standard normal variable transformation preprocessing method was used to eliminate the spectral scattering effect caused by the uneven size of tea samples.Meanwhile,the first derivative preprocessing was used to eliminate the changing background,reduce the interference of baseline drift and underline the useful information from spectra.By combining the pretreatment methods with the principal component analysis,the tea samples could be identified more accurately,with an accuracy of 75.00%.In addition,a supervised linear discriminant analysis method was used to process the original spectral data,with an identification accuracy reaching up to 100%.However,a priori knowledge of categories was needed in the method.Therefore,the combination of near infrared spectroscopy and chemometrics could realize the rapid and non destructive identification of different green tea varieties.