基于高光谱成像技术的白茶储藏年份判别

陈书媛 张友超 杨杰 蔡梅生 张青碧 何普明 屠幼英

陈书媛,张友超,杨杰,等. 基于高光谱成像技术的白茶储藏年份判别[J]. 食品工业科技,2021,42(18):276−283. doi:  10.13386/j.issn1002-0306.2020110299
引用本文: 陈书媛,张友超,杨杰,等. 基于高光谱成像技术的白茶储藏年份判别[J]. 食品工业科技,2021,42(18):276−283. doi:  10.13386/j.issn1002-0306.2020110299
CHEN Shuyuan, ZHANG Youchao, YANG Jie, et al. Discrimination of Storage Time of White Tea Using Hyperspectral Imaging[J]. Science and Technology of Food Industry, 2021, 42(18): 276−283. (in Chinese with English abstract). doi:  10.13386/j.issn1002-0306.2020110299
Citation: CHEN Shuyuan, ZHANG Youchao, YANG Jie, et al. Discrimination of Storage Time of White Tea Using Hyperspectral Imaging[J]. Science and Technology of Food Industry, 2021, 42(18): 276−283. (in Chinese with English abstract). doi:  10.13386/j.issn1002-0306.2020110299

基于高光谱成像技术的白茶储藏年份判别

doi: 10.13386/j.issn1002-0306.2020110299
详细信息
    作者简介:

    陈书媛(1995−),女,硕士研究生,研究方向:茶叶品质检测,E-mail:584533826@qq.com

    通讯作者:

    何普明(1965−),男,博士,教授,研究方向:茶叶生物化学,E-mail:pmhe@zju.edu.cn

  • 中图分类号: TS272.7

Discrimination of Storage Time of White Tea Using Hyperspectral Imaging

  • 摘要: 储藏年份是决定白茶经济价值的一大因素。为了实现快速便捷地判别白茶储藏年份,本文提出了基于高光谱成像技术判别分析白茶储藏年份的无损检测方法。通过对3、6、10年寿眉高光谱图像感兴趣区域光谱数据的提取,采用最小二乘平滑滤波、标准正态变换、归一化、多元散射校正预处理算法,并用支持向量机、偏最小二乘联合线性判定法、逻辑回归建模对不同预处理后的光谱数据进行判别分析。最后,通过分析混淆矩阵、精确率、召回率来评估模型性能。分析结果表明,经过标准正态变换预处理结合支持向量机所建立的模型判别效果最佳,训练集和测试集的精确率分别为90.83%和86.02%。由此可见,利用高光谱成像技术对白茶储藏年份进行快速无损的判别具有一定的可行性。
  • 图  1  寿眉样本高光谱图像数据采集示意图

    Figure  1.  Sketch of data acquisition

    注:(a)高光谱图像去噪后的原图;(b)在样本区域内随机选取100个ROI示意图;(c)随机提取后得到的100个ROI。

    图  2  各储藏年份寿眉样本原始光谱数据图

    Figure  2.  Diagram of raw spectra of different Shoumei samples

    图  3  预处理效果图

    Figure  3.  Spectra of all samples after being preprocessed

    注:(a)原始光谱;经MinMax(b),MSC(c),SNV(d),SGF(e)预处理后的光谱。

    表  1  SVM建模分析结果

    Table  1.   Results of SVM modeling analysis

    预处理算法训练集精确率(%)测试集精确率(%)
    83.4777.58
    Minmax81.6777.34
    SGF87.3183.29
    SNV90.8386.02
    MSC84.4380.67
    下载: 导出CSV

    表  2  SNV-SVM处理的混淆矩阵

    Table  2.   The confusion matrix in SNV-SVM model

    训练集测试集
    年份(年)3610精确率(%)召回率(%)3610精确率(%)召回率(%)
    31195892.9790.15282187.5090.32
    66113590.4091.13327487.1079.41
    103710589.0091.30122583.3389.29
    总值90.8390.8386.0286.02
    注:寿眉的真实储藏年份(行的数字),模型预测的储藏年份(列的数字),表4表6同。
    下载: 导出CSV

    表  3  PLS-LDA建模分析结果

    Table  3.   Results of PLS-LDA modeling analysis

    预处理算法训练集精确率(%)测试集精确率(%)
    77.1270.73
    Minmax77.1770.73
    SGF75.0369.11
    SNV80.0574.19
    MSC75.2963.41
    下载: 导出CSV

    表  4  SNV-PLS-LDA处理的混淆矩阵

    Table  4.   The confusion matrix in SNV-PLS-LDA model

    训练集测试集
    年份(年)3610精确率(%)召回率(%)3610精确率(%)召回率(%)
    3919971.0983.49202362.5080.00
    617101480.8082.79523174.1979.31
    10201510588.9875.00762686.6766.67
    总值80.0580.0574.1974.19
    下载: 导出CSV

    表  5  LR建模分析结果

    Table  5.   Results of LR modeling analysis

    预处理算法训练集精确率(%)测试集精确率(%)
    77.0673.45
    Minmax84.6478.50
    SGF77.0671.22
    SNV83.6971.39
    MSC77.0674.09
    下载: 导出CSV

    表  6  Minmax-LR处理的混淆矩阵

    Table  6.   The confusion matrix in Minmax-LR model

    训练集测试集
    年份(年)3610精确率(%)召回率(%)3610精确率(%)召回率(%)
    31138488.2890.40274284.3881.82
    651031682.4083.06024677.4280.00
    1010149883.0580.33532273.3373.33
    总值84.6484.6478.5078.50
    下载: 导出CSV

    表  7  三种判别模型的最优处理汇总

    Table  7.   Results of three discriminant models with their best preprocessing algorithm

    处理组合训练集精确率(%)测试集精确率(%)
    SNV-SVM90.8386.02
    SNV-PLS-LDA80.0574.19
    Minmax-LR84.6478.50
    下载: 导出CSV
  • [1] Liu L, Liu B, Li J, et al. Responses of different cancer cells to white tea aqueous extract: Response to white tea aqueous extract[J]. Journal of Food Science,2018,83(10-12):2593−2601.
    [2] Orner G A. Suppression of tumorigenesis in the Apcmin mouse: Down-regulation of β-catenin signaling by a combination of tea plus sulindac[J]. Carcinogenesis,2003,24(2):263−267. doi:  10.1093/carcin/24.2.263
    [3] Rasolifard M H, Nozohor Y. The effect of aqueous extract of white tea on the levels of antioxidant enzymes of rats' liver tissue exposed to arsenic[J]. Journal of Ilam University of Medical Sciences,2018,25(5):146−153. doi:  10.29252/sjimu.25.5.146
    [4] Sangeeta Rangi, Sunil Kumar Dhatwalia, Priti Bhardwaj, et al. Evidence of similar protective effects afforded by white tea and its active component ‘EGCG’ on oxidative-stress mediated hepatic dysfunction during benzo(a)pyrene induced toxicity[J]. Food and Chemical Toxicology,2018,116:281−291. doi:  10.1016/j.fct.2018.04.044
    [5] Camouse Melissa M, Domingo Diana Santo, Swain Freddie R, et al. Topical application of green and white tea extracts provides protection from solar-simulated ultraviolet light in human skin[J]. 2009, 18(6): 522−526.
    [6] 刘琳燕, 周子维, 邓慧莉, 等. 不同年份白茶的香气成分[J]. 福建农林大学学报(自然科学版),2015,44(1):27−33. [Liu L Y, Zhou Z W, Deng H L, et al. Analysis of the aromatic components in white tea produced in different years[J]. Journal of Fujian Agriculture and Forestry University (Natural Science Edition),2015,44(1):27−33.
    [7] Ning J M, Ding D, Song Y S, et al. Chemical constituents analysis of white tea of different qualities and different storage times[J]. European Food Research and Technology,2016,242(6):2093−2104.
    [8] Dai W, Xie D, Lu M, et al. Characterization of white tea metabolome: Comparison against green and black tea by a nontargeted metabolomics approach[J]. Food Research International,2017,96(6):40−45.
    [9] Xie D, Dai W, Lu M, et al. Nontargeted metabolomics predicts the storage duration of white teas with 8-C N-ethyl-2-pyrrolidinone-substituted flavan-3-ols as marker compounds[J]. Food Research International,2019,125(11):7209−7218.
    [10] 刘鹏, 艾施荣, 杨普香, 等. 非线性流形降维方法结合近红外光谱技术快速鉴别不同海拔的茶叶[J]. 茶叶科学,2019,39(6):715−722. [Liu P, Ai S R, Yang P X, et al. Nonlinear manifold dimensionality reduction methods for quick discrimination of tea at different altitude by near infrared spectroscopy[J]. Tea Science,2019,39(6):715−722. doi:  10.3969/j.issn.1000-369X.2019.06.011
    [11] Wang Y J, Li T H, Li L Q, et al. Micro-NIR spectrometer for quality assessment of tea: Comparison of local and global models[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy,2020,237:118403. doi:  10.1016/j.saa.2020.118403
    [12] 刘洪林. 基于近红外光谱技术(NIRS)对工夫红茶审评品质客观评价研究[J]. 食品工业科技,2016,37(5):311−315. [Liu H L. Research of evaluation the quality of Congou black tea by near infrared spectroscopy[J]. Science and Technology of Food Industry,2016,37(5):311−315.
    [13] 李春霖. 基于化学计量学和近红外光谱技术的龙井茶感官及化学品质评价研究[D]. 杭州: 浙江大学, 2019.

    Li C L. Sensory and chemical quality evaluation of Longjing tea using chemoetrics and near-infrared spectroscopy technique[D]. Hangzhou: Zhejiang University, 2019.
    [14] 于英杰, 王琼琼, 王冰玉, 等. 基于高光谱技术的铁观音茶叶等级判别[J]. 食品科学,2014,35(22):159−163. [Yu Y J, Wang Q Q, Wang B Y, et al. Identification of Tieguanyin tea grades based on hyperspectral technology[J]. Food Science,2014,35(22):159−163. doi:  10.7506/spkx1002-6630-201422030
    [15] 李晓丽, 魏玉震, 徐劼, 等. 基于高光谱成像的茶叶中EGCG分布可视化[J]. 农业工程学报,2018,34(7):180−186. [Li X L, Wei Y Z, Xu J, et al. EGCG distribution visualization in tea leaves based on hyperspectral imaging technology[J]. Transactions of the Chinese Society of Agricultural Engineering,2018,34(7):180−186. doi:  10.11975/j.issn.1002-6819.2018.07.023
    [16] 李瑶. 基于高光谱成像的茶叶品质判别方法研究[D]. 成都: 四川农业大学, 2018.

    Li Y. A method of distinguishing tea quality based on hyperspectral imaging[D]. Chengdu: Sichuan Agricultural University, 2018.
    [17] Ren G, Wang Y, Ning J, et al. Using near-infrared hyperspectral imaging with multiple decision tree methods to delineate black tea quality[J]. Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy,2020,237:118407. doi:  10.1016/j.saa.2020.118407
    [18] Hong Z, He Y. Rapid and nondestructive discrimination of geographical origins of Longjing tea using hyperspectral imaging at two spectral ranges coupled with machine learning methods[J]. Applied Science,2020, 10(3):1173−1184.
    [19] 魏玉震. 基于光谱和光谱成像技术的茶叶含水率检测机理和方法研究[D]. 杭州: 浙江大学, 2019.

    Wei Y Z. Moisture content detection of tea leaves based on spectral and spectral imaging technologies[D]. Hangzhou: Zhejiang University, 2019.
    [20] 芦兵, 孙俊, 杨宁, 等. 基于荧光透射谱和高光谱图像纹理的茶叶病害预测研究[J]. 光谱学与光谱分析,2019,39(8):2515−2521. [Lu B, Sun J, Yang N, et al. Prediction of tea disease based on fluorescence transmission spectrum and texture of hyperspectral image[J]. Spectroscopy and Spectral Analysis,2019,39(8):2515−2521.
    [21] 田欣. 高光谱图像的去噪模型与算法[D]. 长沙: 湖南大学, 2019.

    Tian X. Models and algorithms for hyperspectral image restoration[D]. Changsha: Hunan University, 2019.
    [22] H Martens, S Å Jensen, P Geladi. Multivariate linearity transformation for near-infrared reflectance spectrometry[C]// Proc Nordic Symp on Applied Statistics(O. H. J. Christie, ed.), Stokkand Forlag Publ, Stavanger, Norway, 1983: 208-234.
    [23] 郭辉, 彭彦昆, 江发潮, 等. 手持式牛肉大理石花纹检测系统[J]. 农业机械学报,2012,43:207−210. [Guo H, Peng Y K, Jiang F C, et al. Development of conveyable beef-marbling detection system[J]. Transactions of the Chinese Society of Agricultural Machinery,2012,43:207−210. doi:  10.6041/j.issn.1000-1298.2012.03.037
    [24] Shaheen H, Agarwal S, Ranjan P. MinMaxScaler binary PSO for feature selection[M]. First International Conference on Sustainable Technologies for Computational Intelligence, 2020: 705-716.
    [25] M Dhanoa, S Lister, R Sanderson, et al. The link between multiplicative scatter correction(MSC) and standard normal variate(SNV) transformations of NIR spectra[J]. Journal of Near Infrared Spectroscopy,1995,2(1):43−47.
    [26] Saunders C, Stitson M O, Weston J, et al. Support vector machine[J]. Computer Ence,2002,1(4):1−28.
    [27] MATLAB中文论坛. MATLAB神经网络30个案例分析[M]. 北京: 北京航空航天大学出版社, 2010.

    Matlab Chinese Forum. Matlab neural network analysis of 30 cases[M]. Beijing: Beihang University Press, 2010.
    [28] Brereton R G, Lloyd G R. Partial least squares discriminant analysis: Taking the magic away[J]. Journal of Chemometrics,2014,28:214−225.
    [29] Miguel Pérez-Enciso, Tenenhaus M. Prediction of clinical outcome with microarray data: A partial least squares discriminant analysis(PLS-DA) approach[J]. Human Genetics,2003,112(5-6):581−592.
    [30] Sabouri S, Hajrasouliha A, Song Y, et al. Logistic regression[M]. Basic Quantitative Research Methods for Urban Planners, 2020: 15-24.
    [31] Allison P D. Logistic regression using the SAS system: Theory and application[M]. SAS Publishing, 1999: 81-90.
  • 加载中
图(3) / 表(7)
计量
  • 文章访问数:  39
  • HTML全文浏览量:  16
  • PDF下载量:  5
  • 被引次数: 0
出版历程
  • 收稿日期:  2020-12-02
  • 网络出版日期:  2021-08-09
  • 刊出日期:  2021-09-14

目录

    /

    返回文章
    返回

    重要通知

    《食品工业科技》编辑部携手万方数据开通学术不端专属检测通道,具体信息参见本刊动态。