基于SMOTE和Inception-CNN的种植和组培金线莲鉴别

doi:10.3964/j.issn.1000-0593(2024)01-0158-06

摘要
参考文献
相关文章 (10)

全文: PDF (4506 KB)
输出: BibTeX | EndNote (RIS)

摘要：金线莲是一种珍贵中药材，其治疗、保健作用十分显著。金线莲培育方式主要有种植、组培等，不同培育方式的金线莲，在性状上仅表现出细微差异，但药用、市场价值差异显著，培育方式鉴别能有效保证药用疗效、维护良好市场秩序，然而由于不同品系、产地、培育时间等复合差异的影响，增加了培育方式鉴别难度与复杂度。提出一种基于改进1D-Inception-CNN模型的金线莲培育方式鉴别方法。采用近红外光谱仪采集种植、组培金线莲的光谱，首先使用合成少数类过采样技术（SMOTE）进行过采样以解决种植品、组培品样本比例不平衡问题，其次构建基于改进Inception结构的一维卷积神经网络对来自不同品系、产地、培育时间的金线莲进行种植品、组培品鉴别，最后采用贝叶斯优化方法对构建的卷积神经网络模型超参数进行优化；最终五折交叉验证平均鉴别准确率、精确率、召回率、综合评价指标高达97.95%、96.16%、100%、98.02%。研究表明，实验提出的鉴别模型为快速鉴别金线莲种植品、组培品提供一种有效方法。

关键词：金线莲；少数类过采样技术；Inception模块；一维卷积神经网络；贝叶斯优化

Abstract：Anoectochilus roxburghii (Wall.) Lindl. (Orchidaceae) is one of the most precious Chinese medicine with extraordinary effects in medical treatment and health protection. Planting and tissue-cultured are two main cultivated methods of A. roxburghii. There are slight characteristic differences between Planting and tissue-cultured A. roxburghii, but they show significant differences in medicinal and market value. Therefore, the identification of cultivated methods plays an important role in effectively securing the medicinal efficacy of A. roxburghii and maintaining a good market order. However, due to the influence of composite differences such as different cultivars, different geographical origins and different times of cultivation, the difficulty and complexity of identification in cultivated methods increase heavily. This paper proposes an effective model to discriminative different cultivated methods of A. roxburghii based on improved 1D-inception-CNN. The experiments were conducted on two kinds of A. roxburghii, and their NIRS data were collected by a Fourier transform near-infrared spectrometer. Considering the unbalanced proportion of planting and tissue-cultured samples，the NIRS data was over sampled by using SMOTE first. Secondly, a one-dimensional convolutional neural network based on improved Inception was constructed to identify planting and tissue-cultured A. roxburghii though both include different varieties, different geographical origins and different cultivating times. Finally, Bayesian optimization was used to optimize the hyperparameters of the model. The final average identification accuracy, precision, recall, and F1-score of five-fold crossvalidation reached 97.95%, 96.16%, 100%, and 98.02%. The identification model proposed in this experiment provides a useful method to identify planting and tissue-cultured A. roxburghii effectively and rapidly and provides an idea for the identification of cultivation methods of other Chinese herbal medicines.

Key words：Anoectochilus roxburghii;SMOTE;Inception module;One-dimensional convolutional neural network;Bayesian optimization

收稿日期: 2022-07-02 修订日期: 2022-10-15

中图分类号:

O657.33

基金资助: 国家自然科学基金项目（61773124）和福建省自然科学基金项目（2021J01636）资助

通讯作者: 柴琴琴 E-mail: qq.chai@fzu.edu.cn

作者简介: 蓝艳，女，1998年生，福州大学电气工程与自动化学院硕士研究生 e-mail: 2031530637@qq.com

引用本文:

蓝艳，王武，许文，柴琴琴，李玉榕，张勋. 基于SMOTE和Inception-CNN的种植和组培金线莲鉴别[J]. 光谱学与光谱分析, 2024, 44(01): 158-163.
LAN Yan，WANG Wu，XU Wen，CHAI Qin-qin，LI Yu-rong，ZHANG Xun. Discrimination of Planting and Tissue-Cultured Anoectochilus Roxburghii Based on SMOTE and Inception-CNN. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 158-163.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2024)01-0158-06 或 https://www.gpxygpfx.com/CN/Y2024/V44/I01/158

[1] Ye S Y, Shao Q S, Zhang A L. Journal of Ethnopharmacology, 2017, 209: 184.
[2] SHAO Qing-song，YE Shen-yi，ZHOU Ai-cun，et al（邵清松, 叶申怡, 周爱存, 等）. China Journal of Chinese Materia Medica（中国中药杂志）, 2016, 41(2): 160.
[3] WU Shui-hua，CHEN Wei-qing（吴水华, 程伟青）. Research and Practice on Chinese Medicines（现代中药研究与实践）, 2016, 30(6): 8.
[4] CHENG Wei-qing，WU Shui-hua，DENG Xiao-zhi（程伟青, 吴水华, 邓笑治）. Journal of Anhui Agricultural Sciences（安徽农业科学）, 2017, 45(20): 123.
[5] WANG Hai-ge，XU Wen，ZHANG Xun，et al（王海阁, 许文, 张勋, 等）. Journal of Chinese Medicinal Materials（中药材）, 2020, 43(2): 303.
[6] CHEN Ying，REN Li，YAN Gui-jie，et al（陈莹, 任丽, 严桂杰, 等）. Journal of Shenyang Pharmaceutical University（沈阳药科大学学报）, 2019, 36(9): 794.
[7] Qiu T, Yang Y Z, Sun H J, et al. Vibrational Spectroscopy, 2021, 116: 103289.
[8] Li S L, Xing B C, Lin D, et al. Industrial Crops and Products, 2020, 152: 112539.
[9] Lin W Q, Chai Q Q, Wang W, et al. Analytical Methods, 2018, 10(25): 2980.
[10] Chen X Y, Chai Q Q, Lin N, et al. Analytical Methods, 2019, 11(40): 5118.
[11] CHEN Qi，PAN Tian-hong，LI Yu-qiang，et al（陈琦, 潘天红, 李鱼强, 等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析）, 2021, 41(9): 2776.
[12] Chai Q Q, Zeng J, Lin D H, et al. Journal of Pharmaceutical and Biomedical Analysis, 2021, 199: 114035.
[13] ZHOU Zhi-hua（周志华）. Machine Learning（机器学习）. Beijing：Tsinghua University Press（北京：清华大学出版社），2016: 66.
[14] Chawla N V, Bowyer K W, Hall L O, et al. Journal of Artificial Intelligence Research, 2002, 16: 321.
[15] TAO Xin-min，HAO Si-yuan，ZHANG Dong-xue，et al（陶新民, 郝思媛, 张冬雪, 等）. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)[重庆邮电大学学报(自然科学版)], 2013, 25(1): 101.
[16] Yao L, Fang Z P, Xiao Y Q, et al. Energy, 2021, 214: 118866.
[17] Bergstra J, Bengio Y. Journal of Machine Learning Research, 2012, 13: 281.
[18] CHAI Qin-qin，ZENG Jian，ZHANG Xun（柴琴琴, 曾建, 张勋）. Acta Agriculturae Zhejiangensis（浙江农业学报）, 2022, 34(2): 391.
[19] CUI Jia-xu，YANG Bo（崔佳旭, 杨博）. Journal of Software（软件学报）, 2018, 29(10): 3068.