湖北农业科学 ›› 2020, Vol. 59 ›› Issue (5): 138-141.doi: 10.14088/j.cnki.issn0439-8114.2020.05.027

• 信息工程 • 上一篇    下一篇

基于近红外光谱的普洱茶产地识别研究

李余进a, 李超a, 李家华b, 施伟a, 李哲a, 姚应方a, 张艳诚a   

  1. 云南农业大学,a.机电工程学院;b.龙润普洱茶学院,昆明 650201
  • 修回日期:2019-08-06 出版日期:2020-03-10 发布日期:2020-06-05
  • 通讯作者: 张艳诚,副教授,主要从事农业信息图像处理方面的研究,(电话)18787027521(电子信箱)zhangyancheng@163.com。
  • 作者简介:李余进(1994-),女,安徽铜陵人,在读硕士研究生,研究方向为农业生物环境控制,(电话)15258264195(电子信箱)15258264195@163.com
  • 基金资助:
    国家自然科学基金项目(NNSFC31160173; NNSFC31160196)

Identification of Pu'er tea origin based on near infrared spectroscopy

LI Yu-jina, LI Chaoa, LI Jia-huab, SHI Weia, LI Zhea, YAO Ying-fanga, ZHANG Yan-chenga   

  1. a.College of Mechanical and Electrical Engineering;b.Longrun Pu'er Tea College,Yunnan Agricultural University,Kunming 650201,China
  • Revised:2019-08-06 Online:2020-03-10 Published:2020-06-05

摘要: 以81份来自西双版纳、普洱市、临沧市3个不同产地的普洱茶样本为研究对象,其中校正集54份,验证集27份,利用近红外光谱采集单个普洱茶样本在1 100~2 498 nm的光谱数据,分别采用主成分聚类分析法和判别分析法建立普洱茶产地识别定性分析模型。结果表明,2种分析模型的校正集正确识别率均大于90%,可用于普洱茶产地的识别。其中,基于判别分析方法建立的判别分析模型效果更好,其校正集和验证集的识别正确率达到98.15%和100%,更适用于普洱茶产地的识别。应用近红外光谱技术可快速、无损识别普洱茶产地,为普洱茶产地检测提供参考。

关键词: 近红外光谱, 普洱茶, 聚类分析, 判别分析, 产地识别

Abstract: : 81 Pu'er tea samples from three different producing areas, such as Xishuangbanna, Pu'er City and Linyi City, were studied. Among them, 54 were corrected and 27 were verified. A single Pu'er tea sample was collected by near-infrared spectroscopy at 1100- The spectral data in the range of 2498 nm was established by principal component cluster analysis and discriminant analysis to establish a qualitative analysis model for Pu'er tea production. The test results show that the correct recognition rate of the correction set of the two analytical models is greater than 90%, which can be used for the identification of Pu'er tea origin. Among them, the discriminant analysis model based on the discriminant analysis method is better, and the correctness rate of the calibration set and the verification set is 98.15% and 100%, which is more suitable for the identification of Pu'er tea origin. Therefore, the application of near-infrared spectroscopy technology can quickly and non-destructively identify the origin of Pu'er tea, providing a new method for the detection of Pu'er tea production.

Key words: near-infrared spectroscopy, Pu'er tea, cluster analysis, discriminant analysis, origin identification

中图分类号: