基于拉曼光谱结合CNN-LSTM深度学习方法的铁皮石斛总黄酮含量快速检测研究

doi:10.3964/j.issn.1000-0593(2024)04-1018-07

摘要
参考文献
相关文章 (15)

全文: PDF (4851 KB)
输出: BibTeX | EndNote (RIS)

摘要：铁皮石斛具有很高的商业价值和营养价值，将云南文山、广西金秀、安徽霍山、浙江台州四个产地共130个样品作为研究样本，在785 nm激光下利用便携式拉曼光谱仪获得了铁皮石斛拉曼光谱，采用NaNO₂-Al(NO₃)₃-NaOH比色法测定铁皮石斛总黄酮含量。以每条经过归一化后的拉曼光谱数据作为输入，利用Savitzky-Golay卷积平滑（SG平滑）、标准正态变量变换（SNV）、多元散射校正（MSC）等不同预处理方法对光谱数据进行处理，以偏最小二乘（PLS）、支持向量机（SVM）和卷积神经网络-长短期记忆神经网络（CNN-LSTM）模型作为比较，竞争自适应重加权采样（CARS）作为波长选择方法，对不同的机器学习模型进行比较研究。采用以下预测质量指标：校正集、测试集相关系数（R_c、R_p），校正集、测试集均方根误差（RMSEC、RMSEP），评价铁皮石斛总黄酮含量预测模型的性能。结果表明：光谱在经过SNV预处理之后，CNN-LSTM方法预测铁皮石斛总黄酮含量准确率最高，R_c、R_p分别为0.983和0.964，RMSEC、RMSEP分别为0.032和0.047 mg·g^-1。结合拉曼光谱建立的SNV-CNN-LSTM深度学习模型准确可靠，具有很强的鲁棒性，优于传统的机器学习模型（PLS、SVM）。利用拉曼光谱结合CNN-LSTM模型对铁皮石斛总黄酮含量进行预测，克服了传统的理化鉴别法的缺陷，具有快速无损的特点。该方法能对铁皮石斛的品质进行区分，并加快药食同源植物市场铁皮石斛产业化，构建自主品牌并增加其影响力，同时此项技术也可应用于消费者和市场监管部门。

关键词：铁皮石斛；拉曼光谱；卷积神经网络-长短期记忆神经网络；总黄酮含量；快速检测

Abstract：Dendrobium officinale has high commercial value and nutritional value. In this study, 130 samples were taken as research samples from Wenshan in Yunnan, Jinxiu in Guangxi, Huoshan in Anhui and Taizhou in Zhejiang. The Raman spectrawere obtained by a portable Raman spectrometer under a 785 nm laser. Then, the total flavonoid content of Dendrobium officinale was determined by NaNO₂-Al (NO₃)₃-NaOH colorimetry. With each normalized Raman spectral data as input, different preprocessing methods included SG, SNV and MSC are used to preprocess the spectral data. Partial least squares (PLS), support vector machine (SVM) and convolution neural network short and long-term memory neural network (CNN-LSTM) models are used as a comparison, and competitive adaptive reweighting sampling (CARS) is used as wavelength selection method to compare different machine learning models. In addition, the following prediction quality indicators were used: correction set and correlation coefficient of the test set (R_c，R_p), root mean square error of correction set and root mean square error of prediction set (RMSEC, RMSEP) to evaluate the performance of the prediction model of total flavone content in Dendrobium officinale. The results showed that the prediction accuracy of the CNN-LSTM method was the highest, with R_c and R_p of 0.983 and 0.964, RMSEC and RMSEP of 0.032 and 0.047 mg·g^-1, respectively. The SNV-CNN-LSTM deep learning model based on Raman spectroscopy is accurate, reliable, and robust, superior to traditional machine learning models (PLS, SVM). In this study, Raman spectroscopy combined with the CNN-LSTM model was used to predict the content of total flavonoids in Dendrobium officinale with the characteristics of fast and non-destructive, which overcame the shortcomings of traditional physical and chemical identification methods. This method can distinguish the quality of Dendrobium officinale, accelerate the industrialization of Dendrobium officinale in the market of medicinal and edible homologous plants, build its brand and increase its influence. At the same time, this technology can also be applied to consumers and market supervision departments, with broad prospects.

Key words：Dendrobium officinale; Raman spectroscopy; CNN-LSTM; Total flavonoids; Rapid detection

收稿日期: 2022-10-02 修订日期: 2022-11-26

中图分类号:

O657.3

基金资助: 浙江省重点研发计划项目（2022C04002），国家重点研究与发展计划项目(2017YFF0211302)，浙江省自然科学基金青年基金项目(LQ18F050003)，浙江省教育厅科研资助项目(Y202148007)资助

通讯作者: 张雷蕾，朱诚 E-mail: zhangleilei@cjlu.edu.cn；pzhch@cjlu.edu.cn

作者简介: 刘宗溢，1998年生，中国计量大学生命科学学院硕士研究生 e-mail: lzyzgjldx@163.com

引用本文:

刘宗溢，张彩虹，蒋健康，沈斌国，丁艳菲，张雷蕾，朱诚. 基于拉曼光谱结合CNN-LSTM深度学习方法的铁皮石斛总黄酮含量快速检测研究[J]. 光谱学与光谱分析, 2024, 44(04): 1018-1024.
LIU Zong-yi, ZHANG Cai-hong, JIANG Jian-kang, SHEN Bin-guo, DING Yan-fei, ZHANG Lei-lei, ZHU Cheng. Rapid Determination of Total Flavonoids in Dendrobium Officinale Based on Raman Spectroscopy and CNN-LSTM Deep Learning Method. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(04): 1018-1024.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2024)04-1018-07 或 https://www.gpxygpfx.com/CN/Y2024/V44/I04/1018

[1] Zhao M, Fan J K, Liu Q T, et al. Food Science & Nutrition, 2021, 9(12): 6575.
[2] WEI Ming, ZHANG Qian, QIAN Sen-he, et al(魏明, 张倩, 钱森和, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2022, 38(8): 281.
[3] Wan J Q, Gong X H, Wang F X, et al. Biomedical Chromatography: BMC, 2021, 36(1): 5250.
[4] Wu M, Chen L J, Huang X M, et al. Journal of Luminescence, 2018, 202: 239.
[5] Li Y Z, Zhao C Y, Lu C, et al. Food Chemistry, 2020, 338: 128115.
[6] Govindammal M, Prasath M, Kamaraj S, et al. Biocatalysis and Agricultural Biotechnology, 2019, 18: 101086.
[7] Pompeu D, Larondelle Y, Rogez H, et al. Biotechnologie, Agronomie, Société et Environnement, 2018, 22(1): 13.
[8] Chen X, Li J Q, Liu H G, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2022, 274: 121137.
[9] YU Guo-wei, MA Ben-xue, CHEN Jin-cheng, et al(喻国威, 马本学, 陈金成, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2021, 41(12): 3701.
[10] Zhang C, Wu W Y, Zhou L, et al. Food Chemistry, 2020, 319: 126536.
[11] Chinese Pharmacopoeia Commission(国家药典委员会). Pharmacopoeia of the People's Republic of China(中华人民共和国药典), Part One(第一部). Beijing: China Medical Science Press(北京: 中国医药科学出版社), 2020. 370.
[12] Xie J, Huang J H, Ren G X, et al. Foods, 2022, 11(6): 892.
[13] Zhan B S, Xiao X, Pan F, et al. International Journal of Wireless and Mobile Computing, 2021, 21(1): 41.
[14] Ye A S, Zhou X B, Miao F. Electronics, 2022, 11(5): 775.
[15] Roy D K, Sarkar T K, Kamar S S, et al. Agronomy, 2022, 12(3): 594.
[16] Garbin C, Zhu X Q, Marques O. Multimedia Tools and Applications, 2020, 79: 1.
[17] Ewelina W, Ewelina S, Krzysztof C, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2017, 185: 317.
[18] Hartwig S, Malgorzata B. Vibrational Spectroscopy，2006, 43: 13.
[19] Sylwester M, Izabela F, Magdalena W, et al. Talanta，2018, 189: 308.