食品与发酵工业 >

2024 , Vol. 50 >Issue 20: 286 - 293

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.037906

分析与检测

基于近红外光谱的婺源绿茶感官品质评价

  • 俞素琴 ,
  • 杨玉璞 ,
  • 张处平 ,
  • 董春旺 ,
  • 祁丹丹 ,
  • 杨崇山
展开
  • 1(江西婺源茶业职业学院 茶学系,江西 上饶,334000)
    2(内蒙古民族大学 农学院,内蒙古 通辽,028000)
    3(江西河红茶业有限公司,江西 上饶,334000)
    4(山东农业科学院茶叶研究所,山东 济南,250000)
第一作者:学士,副教授(祁丹丹博士和杨崇山博士为共同通信作者,E-mail:qidandan07@126.com;1029345485@qq.com)

收稿日期: 2023-11-09

  修回日期: 2024-01-03

  网络出版日期: 2024-11-01

基金资助

江西省科技合作重点项目(20212BDH80025,20212BDH8011);浙江省重点研发计划项目(2022C02010,2023C02043)

收起

Sensory quality evaluation of Wuyuan green tea based on near-infrared spectroscopy

  • YU Suqin ,
  • YANG Yupu ,
  • ZHANG Chuping ,
  • DONG Chunwang ,
  • QI Dandan ,
  • YANG Chongshan
Expand
  • 1(Department of Tea Science, Jiangxi Wuyuan Tea Vocational College, Shangrao 334000, China)
    2(College of Agriculture, Inner Mongolia University for Nationalities, Tongliao 028000, China)
    3(Jiangxi River Red Tea Co.Ltd., Shangrao 334000, China)
    4(Institute of Tea Research, Shandong Academy of Agricultural Sciences, Jinan 250000, China)

Received date: 2023-11-09

  Revised date: 2024-01-03

  Online published: 2024-11-01

Fold

摘要

该文以不同品种和品质等级的婺源绿茶为研究对象,基于近红外光谱无损检测技术,分别建立了多品种婺源绿茶的感官评分和儿茶素含量预测模型,比较了不同预处理算法、变量筛选方法和建模方法对预测精度的影响。首先,将原始光谱预处理后,使用主成分分析进行降维处理,随后通过竞争自适应重加权采样法(competitive adaptive reweighted sampling, CARS)、随机蛙跳跃算法和变量空间迭代收缩法筛选出与感官评分和儿茶素含量有关的特征波段,分别建立了偏最小二乘法和随机森林(random forest algorithm, RF)预测模型。结果表明,感官评分最佳的预处理和变量筛选算法分别为标准正态变量变换和CARS,儿茶素含量的最佳的预处理和变量筛选算法分别为标准化和CARS,非线性RF模型效果最佳,对感官评分和儿茶素含量的预测精度分别达到了0.927和0.939,相对标准偏差值均>2,表明模型预测性能较好,鲁棒性较强。研究表明近红外光谱技术可用于不同品质等级的婺源绿茶感官评分和儿茶素含量的快速预测。

关键词： 婺源绿茶; 近红外光谱; 感官评分; 儿茶素; 预测

本文引用格式

俞素琴 , 杨玉璞 , 张处平 , 董春旺 , 祁丹丹 , 杨崇山 . 基于近红外光谱的婺源绿茶感官品质评价[J]. 食品与发酵工业, 2024 , 50(20) : 286 -293 . DOI: 10.13995/j.cnki.11-1802/ts.037906

Abstract

In this paper, the sensory score and catechin content prediction models of multi-variety Wuyuan green tea were established based on near-infrared (NIR) spectral nondestructive testing technology with different varieties and quality grades of Wuyuan green tea as research objects, and the effects of different pretreatment algorithms, variable screening methods, and modelling methods on the prediction accuracy were compared.First, the raw spectra were pre-processed and principal component analysis (PCA) was used to reduce the dimensionality.The characteristic wavelengths related to sensory scores and catechin content were subsequently screened by competitive adaptive reweighted sampling (CARS), shuffled frog leading algorithm (SFLA), and variable iterative space shrinkage approach (VISSA).The linear partial least squares regression (PLS) and random forest algorithm (RF) prediction models were developed.Results showed that the best preprocessing and variable screening algorithms for sensory scores were standard normal variate (SNV) and CARS, respectively, and the best preprocessing and variable screening algorithms for catechin content were min-max normalization (Min-max) and CARS, respectively.The nonlinear RF model based on variable screening was the best, and the prediction accuracy of sensory scores and catechin content reached 0.927 and 0.939, respectively, and the relative standard deviation (RSD) values of the prediction models were greater than 2, indicating that the model has better prediction performance and better robustness.The study indicates that NIR spectroscopy can be used for sensory scoring and rapid prediction of the catechin content of Wuyuan green tea of different quality grades.

Key words： Wuyuan green tea; near infrared spectroscopy; sensory score; catechins; prediction

参考文献

[1] 方县良,卢新松,程建华,等.婺源县茶产业发展的优势、问题及建议[J].蚕桑茶叶通讯,2022(2):37-39.
FANG X L, LU X S, CHENG J H, et al.Advantages, problems and suggestions of tea industry development in wuyuan county[J].Newsletter of Sericulture and Tea,2022(2):37-39.
[2] 徐春晖,王远兴.基于GC-MS结合化学计量学方法鉴别3种江西名茶[J].食品科学,2020,41(20):141-150.
XU C H, WANG Y X.Discrimination of three famous teas in Jiangxi using gas chromatography-mass spectrometry combined with chemometrics[J].Food Science,2020,41(20):141-150.
[3] YANG C S, ZHAO Y, AN T, et al.Quantitative prediction and visualization of key physical and chemical components in black tea fermentation using hyperspectral imaging[J].LWT, 2021, 141:110975.
[4] RANATUNGA M, UWADAIRA Y, IKEHATA A, et al.NIR spectroscopic determination of polyphenol content in teas and tea extract at 2142 nm[J].Sensors and Materials, 2021, 33(12):4135.
[5] AMSARAJ R, MUTTURI S.Classification and quantification of multiple adulterants simultaneously in black tea using spectral data coupled with chemometric analysis[J].Journal of Food Composition and Analysis, 2024, 125:105715.
[6] TURGUT S S, ENTRENAS J A, TAŞKıN E, et al.Estimation of the sensory properties of black tea samples using non-destructive near-infrared spectroscopy sensors[J].Food Control, 2022, 142:109260.
[7] JOHNSON J B, THANI P R, NAIKER M.Through-container detection of tea tree oil adulteration using near-infrared spectroscopy (NIRS)[J].Chemical Papers, 2023,77(4):2009-2017.
[8] DONG C W, LIU Z Y, YANG C S, et al.Rapid detection of exogenous sucrose in black tea samples based on near-infrared spectroscopy[J].Infrared Physics & Technology, 2021,119:103934.
[9] 王华杰,滑金杰,余勤艳,等.基于IRAE-HS-SPME/GC-MS分析杀青方式对绿茶栗香形成的影响[J].食品科学,2021,42(14):209-217.
WANG H J, HUA J J, YU Q Y, et al.Effects of fixation methods on the formation of the chestnut-like aroma of green tea analyzed by infrared-assisted extraction coupled to headspace solid-phase microextraction combined with gas chromatography-mass spectrometry[J].Food Science,2021,42(14):209-217.
[10] JIA J M, ZHOU X F, LI Y, et al.Establishment of a rapid detection model for the sensory quality and components of Yuezhou Longjing tea using near-infrared spectroscopy[J].LWT, 2022, 164:113625.
[11] DONG C W, AN T, YANG M, et al.Quantitative prediction and visual detection of the moisture content of withering leaves in black tea (Camellia sinensis) with hyperspectral image[J].Infrared Physics & Technology, 2022, 123:104118.
[12] 李艳坤,董汝南,张进,等.光谱数据解析中的变量筛选方法[J].光谱学与光谱分析,2021,41(11):3331-3338.
LI Y K, DONG R N, ZHANG J, et al.Variable selection methods in spectral data analysis[J].Spectroscopy and Spectral Analysis,2021,41(11):3331-3338.
[13] 甄欢仪,马瑞峻,陈瑜,等.基于CARS和K-S的马拉硫磷农药浓度吸收光谱预测模型研究[J].光谱学与光谱分析,2020,40(5):1601-1606.
ZHEN H Y, MA R J, CHEN Y, et al.Study on prediction model of malathion pesticide concentration absorption spectra based on CARS and K-S[J].Spectroscopy and Spectral Analysis,2020,40(5):1601-1606.
[14] 周修理,张萍萍,秦娜,等.基于GA-RF模型土壤坚实度对黑土区大豆产量的影响[J].东北农业大学学报, 2022, 53(10):67-75.
ZHOU X L, ZHANG P P, QIN N, et al.Effects of soil penetration resistance on soybean yield in black soil region based on GA-RF model[J].Journal of Northeast Agricultural University, 2022, 53(10):67-75.
[15] BENSEMMANE, N, BOUZIDI N, DAGHBOUCHE Y, et al.Prediction of total phenolic acids contained in plant extracts by PLS-ATR-FTIR[J].South African Journal of Botany, 2022,151:295-305.
[16] 张贱根,刘均华,刘知远,等.基于相关性和主成分分析的上犹名优绿茶品质评价[J].食品安全质量检测学报,2023,14(3):285-292.
ZHANG J G, LIU J H, LIU Z Y, et al.Quality analysis of Shangyou famous green tea based on correlation and principal component analysis[J].Journal of Food Safety and Quality,2023,14(3):285-292.
[17] 孙晓荣,胡毅然,刘翠玲,等.基于近红外光谱技术的茶叶新旧鉴别及产地溯源研究[J].食品安全质量检测学报,2023,14(5):53-59.
SUN X R, HU Y R, LIU C L, et al.Identification of new and old tea and origin traceability based on near infrared spectroscopy[J].Journal of Food Safety and Quality,2023,14(5):53-59.
[18] 浦宇文.基于近红外光谱的茶鲜叶分级研究与试验[D].合肥:安徽农业大学,2022.
PU Y W.Research and experiment on classification of fresh tea leaves based on near-infrared spectroscopy[D]. Hefei:Anhui Agricultural University,2022.
[19] 林正雨,陈强,邓良基,等.基于MaxEnt和MCR的四川省柑橘生产布局模拟[J].中国农业资源与区划,2019,40(9):64-74.
LIN Z Y, CHEN Q, DENG L J, et al.Simulation of Citrus production layout in Sichuan province based on maxent and mcr[J].Chinese Journal of Agricultural Resources and Regional Planning,2019,40(9):64-74.
[20] 宋真.基于改进CenterNet的植物叶片分割与参数提取[D].桂林:广西师范大学,2021.
SONG Z.Plant leaf segmentation and parameter extraction based on improved CenterNet[D].Guilin:Guangxi Normal University,2021.
[21] 李丰先,罗磊,李亚杰,等.基于PCA和隶属函数法分析的马铃薯创新种质抗旱性鉴定与分类[J].干旱区资源与环境,2022,36(11):141-147.
LI F X, LUO L, LI Y J, et al.Identification and classification of drought resistance of potato germplasm based on PCA and membership function analysis[J].Journal of Arid Land Resources and Environment, 2022,36(11):141-147.
[22] 王文才,赵刘,李绍稳,等.基于特征波长选择和建模的高光谱土壤总氮含量估测方法研究[J].浙江农业学报,2018,30(9):1576-1584.
WANG W C, ZHAO L, LI S W, et al.Prediction of soil total nitrogen content from hyperspectral data based on charateristic wave-length selection and modelling[J].Acta Agriculturae Zhejiangensis,2018,30(9):1576-1584.
[23] 聂志刚,李广,王钧,等.基于APSIM模型旱地小麦叶面积指数相关参数的优化[J].中国农业科学,2019,52(12):2056-2068.
NIE Z G, LI G, WANG J, et al.Parameter optimization for the simulation of leaf area index of dryland wheat with the APSIM model[J].Scientia Agricultura Sinica,2019,52(12):2056-2068.
[24] SUN J, TIAN Y, WU X H, et al.Nondestructive detection for moisture content in green tea based on dielectric properties and VISSA-GWO-SVR algorithm[J].Journal of Food Processing and Preservation, 2020, 44(5).
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司