太赫兹光谱在转基因菜籽油鉴别中的应用: 基于改进蜉蝣算法的支持向量机模型

陈涛; 李欣

doi:10.7498/aps.73.20231569

摘要

为实现对转基因和非转基因菜籽油的快速准确鉴别, 结合太赫兹时域光谱技术, 提出了一种基于改进蜉蝣优化算法的支持向量机模型. 以两种转基因和两种非转基因菜籽油为研究对象, 应用太赫兹时域光谱技术获取其光谱信息, 发现相比于非转基因菜籽油, 转基因菜籽油在太赫兹波段具有更强的吸收特性, 同时它们的吸收光谱极为相似, 难以通过观察法进行准确区分. 为此, 提出一种基于改进蜉蝣优化算法的支持向量机模型, 通过采用蜉蝣优化算法对支持向量机参数进行寻优, 并引入自适应惯性权重和Lévy飞行两种策略改进蜉蝣优化算法在寻优过程容易陷入局部最优解的问题, 增强蜉蝣优化算法的全局搜索能力和稳健性. 实验结果表明: 改进后的蜉蝣优化算法能够更有效地寻找到支持向量机的最优参数组合, 提升鉴别模型的整体性能, 该模型对4种菜籽油的识别精度为100%. 因此, 本研究为转基因菜籽油的类型鉴别提供了一种快速有效的新方法, 也为其他转基因物质的鉴别提供了有价值的参考.

关键词:

Abstract

To achieve rapid and accurate identification of genetically modified (GM) and non-GM rapeseed oils, a support vector machine (SVM) model based on an improved mayfly optimization algorithm and coupled with the terahertz time-domain spectroscopy, is proposed. Two types of GM rapeseed oils and two types of non-GM rapeseed oils are selected as research subjects. Their spectral information is acquired by using the terahertz time-domain spectroscopy. The observations show that GM rapeseed oils exhibit stronger terahertz absorption characteristics than non-GM rapeseed oils. However, their absorption spectra are highly similar, making direct differentiation difficult through visual inspection alone. Therefore, SVM is used for spectral recognition. Considering that the classification performance of SVM is significantly affected by its parameters, the mayfly optimization algorithm is combined to optimize these parameters. Furthermore, adaptive inertia weight and Lévy flight strategies are introduced to enhance the global search capability and robustness of the mayfly optimization algorithm, thus addressing the issue of easily becoming trapped in local optima in the optimization process. Moreover, principal component analysis is used to reduce the dimensionality of the absorbance data in a 0.3–1.8 THz range, aiming to extract critical features, thereby enhancing modeling efficiency and reducing redundancy in spectral data. Experimental results demonstrate that the improved mayfly optimization algorithm effectively identifies the optimal parameter combination for SVM, thereby enhancing the overall performance of the identification model. The proposed SVM model, in which the improved mayfly optimization algorithm is used, can achieve a recognition accuracy of 100% for the four types of rapeseed oils, surpassing the 98.15% accuracy achieved by the SVM model with the original mayfly optimization algorithm. Thus, this study presents a rapid and effective new approach for identifying GM rapeseed oils and offers a valuable reference for identifying other genetically modified substances.

Keywords:

作者及机构信息

陈涛,
李欣

桂林电子科技大学电子工程与自动化学院, 桂林　541004

通信作者: 陈涛, tchen@guet.edu.cn

基金项目: 国家自然科学基金(批准号: 62261012, 61841502)资助的课题.

Authors and contacts

Chen Tao,
Li Xin

School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China

Corresponding author: Chen Tao, tchen@guet.edu.cn

Funds: Project supported by the National Natural Science Foundation of China (Grant Nos. 62261012, 61841502).

文章全文 : translate this paragraph

参考文献

[1]	国际农业生物技术应用服务组织 2021 中国生物工程杂志 41 114 ISAAA 2021 China Biotechnol. 41 114
[2]	Kumar K, Gambhir G, Dass A, Tripathi A K, Singh A, Jha A K, Yadava P, Choudhary M, Rakshit S 2020 Planta 251 91 Google Scholar
[3]	Demeke T, Dobnik D 2018 Anal. Bioanal. Chem. 410 4039 Google Scholar
[4]	Gampala S S, Wulfkuhle B, Richey K A 2019 Transgenic Plants 1864 411 Google Scholar
[5]	彭晓昱, 周欢 2021 70 240701 Google Scholar Peng X Y, Zhou H 2021 Acta Phys. Sin. 70 240701 Google Scholar
[6]	Mittleman D M 2017 J. Appl. Phys. 122 230901 Google Scholar
[7]	Sun L, Zhao L, Peng R Y 2021 Mil. Med. Res. 8 28 Google Scholar
[8]	胡颖, 王晓红, 郭澜涛, 张存林, 刘海波, 张希成 2005 54 4124 Google Scholar Hu Y, Wang X H, Guo L T, Zhang C L, Liu H B, Zhang X C 2005 Acta Phys. Sin. 54 4124 Google Scholar
[9]	陈涛 2016 量子电子学报 33 392 Chen T 2016 Chin. J. Quantum Electron. 33 392
[10]	张文涛, 李跃文, 占平平, 熊显名 2017 红外与激光工程 46 1125004 Google Scholar Zhang W T, Li Y W, Zhan P P, Xiong X M 2017 Infrared Laser Eng. 46 1125004 Google Scholar
[11]	Liu J J 2017 Microw. Opt. Technol. Lett. 59 654 Google Scholar
[12]	Liu J J, Fan L L, Liu Y M, Mao L L, Kan J Q 2019 Spectrochim. Acta A Mol. Biomol. Spectrosc. 206 165 Google Scholar
[13]	Gu Q H, Chang Y X, Li X H, Chang Z Z, Feng Z D 2021 Expert Syst. Appl. 165 113713 Google Scholar
[14]	Guo L, Xu C, Yu T H, Tuerxun W 2022 IEEE Access 10 36335 Google Scholar
[15]	Cortes C, Vapnik V 1995 Mach. Learn. 20 273 Google Scholar
[16]	Tuerxun W, Xu C, Guo H Y, Jin Z J, Zhou H J 2021 IEEE Access 9 69307 Google Scholar
[17]	Zervoudakis K, Tsafarakis S 2020 Comput. Ind. Eng. 145 106559 Google Scholar
[18]	Ding Y H, You W B 2020 IEEE Access 8 207089 Google Scholar
[19]	Nickabadi A, Ebadzadeh M M, Safabakhsh R 2011 Appl. Soft Comput. 11 3658 Google Scholar
[20]	Syama S, Ramprabhakar J, Anand R, Guerrero J M 2023 Results Eng. 19 101274 Google Scholar
[21]	Liu N, Luo F, Ding W C 2019 2019 IEEE Symposium Series on Computational Intelligence (SSCI) Xiamen, China, December 6–9, 2019 p3104
[22]	Pan P Y, Xing Y H, Zhang D W, Wang J, Liu C L, Wu D, Wang X Y 2023 J. Food Sci. 88 3189 Google Scholar
[23]	Elahi N, Duncan R W, Stasolla C 2016 Plant Physiol. Biochem. 100 52 Google Scholar

施引文献

图 1 THz-TDS系统原理图

Fig. 1. Schematic diagram of THz-TDS system.

下载: 全尺寸图片幻灯片

图 2 360个菜籽油样本的THz时域光谱

Fig. 2. THz time-domain spectra of 360 rapeseed oil samples.

下载: 全尺寸图片幻灯片

图 3 4种菜籽油及参考信号的THz时域光谱

Fig. 3. THz time-domain spectra of four types of rapeseed oils and reference signal.

下载: 全尺寸图片幻灯片

图 4 4种菜籽油及参考信号的THz频域光谱

Fig. 4. THz frequency-domain spectra of four types of rapeseed oils and reference signal.

下载: 全尺寸图片幻灯片

图 5 360个菜籽油样本在0.3—1.8 THz波段内的吸光度谱

Fig. 5. Absorption spectra of 360 rapeseed oil samples in the 0.3—1.8 THz range.

下载: 全尺寸图片幻灯片

图 6 4种菜籽油在0.3—1.8 THz波段内的平均吸光度谱

Fig. 6. Average absorption spectra of four types of rapeseed oils in the 0.3–1.8 THz range.

下载: 全尺寸图片幻灯片

图 7 吸光度的主成分方差贡献率变化条形图

Fig. 7. Bar chart of variance contribution rates for absorbance’s principal components.

下载: 全尺寸图片幻灯片

图 8 吸光度前3个主成分的3D散点图

Fig. 8. 3D scatter plot of the first three principal components of absorbance.

下载: 全尺寸图片幻灯片

图 9 两种算法下SVM参数寻优过程中的适应度变化曲线　(a) MOA; (b) ALMOA

Fig. 9. Fitness evolution curves during SVM parameter optimization process for two algorithms: (a) MOA; (b) ALMOA.

下载: 全尺寸图片幻灯片

图 10 两种模型的分类结果混淆矩阵　(a) MOA-SVM模型; (b) ALMOA-SVM模型

Fig. 10. Confusion matrices of the classification results for the two models: (a) MOA-SVM model; (b) ALMOA-SVM model.

下载: 全尺寸图片幻灯片

表 1 实验样品信息

Table 1. The information of experimental sample.

标识符	品牌	类型	样本数
标识符	品牌	类型	训练集	测试集
Non-GMO1	道道全	非转基因	63	27
Non-GMO2	鲁花	非转基因	63	27
GMO1	金龙鱼	转基因	63	27
GMO2	鄉佬坎	转基因	63	27

下载: 导出CSV

表 2 两种算法的SVM参数寻优结果

Table 2. Results of SVM parameter optimization under two algorithms.

优化算法	最佳适应度/%	参数
优化算法	最佳适应度/%	c	g
MOA	97.22	12.42	0.79
ALMOA	98.41	84.62	0.12

下载: 导出CSV

表 3 MOA-SVM模型与ALMOA-SVM模型的性能评价

Table 3. Performance evaluation of the MOA-SVM model and ALMOA-SVM model.

模型	样品	查全率/%	查准率/%	精度/%
MOA-SVM	Non-GMO1	100	96.43	98.15
	Non-GMO2	92.59	100
	GMO1	100	96.43
	GMO2	100	100
ALMOA-SVM	Non-GMO1	100	100	100
	Non-GMO2	100	100
	GMO1	100	100
	GMO2	100	100

下载: 导出CSV

map

[1]	国际农业生物技术应用服务组织 2021 中国生物工程杂志 41 114 ISAAA 2021 China Biotechnol. 41 114
[2]	Kumar K, Gambhir G, Dass A, Tripathi A K, Singh A, Jha A K, Yadava P, Choudhary M, Rakshit S 2020 Planta 251 91 Google Scholar
[3]	Demeke T, Dobnik D 2018 Anal. Bioanal. Chem. 410 4039 Google Scholar
[4]	Gampala S S, Wulfkuhle B, Richey K A 2019 Transgenic Plants 1864 411 Google Scholar
[5]	彭晓昱, 周欢 2021 70 240701 Google Scholar Peng X Y, Zhou H 2021 Acta Phys. Sin. 70 240701 Google Scholar
[6]	Mittleman D M 2017 J. Appl. Phys. 122 230901 Google Scholar
[7]	Sun L, Zhao L, Peng R Y 2021 Mil. Med. Res. 8 28 Google Scholar
[8]	胡颖, 王晓红, 郭澜涛, 张存林, 刘海波, 张希成 2005 54 4124 Google Scholar Hu Y, Wang X H, Guo L T, Zhang C L, Liu H B, Zhang X C 2005 Acta Phys. Sin. 54 4124 Google Scholar
[9]	陈涛 2016 量子电子学报 33 392 Chen T 2016 Chin. J. Quantum Electron. 33 392
[10]	张文涛, 李跃文, 占平平, 熊显名 2017 红外与激光工程 46 1125004 Google Scholar Zhang W T, Li Y W, Zhan P P, Xiong X M 2017 Infrared Laser Eng. 46 1125004 Google Scholar
[11]	Liu J J 2017 Microw. Opt. Technol. Lett. 59 654 Google Scholar
[12]	Liu J J, Fan L L, Liu Y M, Mao L L, Kan J Q 2019 Spectrochim. Acta A Mol. Biomol. Spectrosc. 206 165 Google Scholar
[13]	Gu Q H, Chang Y X, Li X H, Chang Z Z, Feng Z D 2021 Expert Syst. Appl. 165 113713 Google Scholar
[14]	Guo L, Xu C, Yu T H, Tuerxun W 2022 IEEE Access 10 36335 Google Scholar
[15]	Cortes C, Vapnik V 1995 Mach. Learn. 20 273 Google Scholar
[16]	Tuerxun W, Xu C, Guo H Y, Jin Z J, Zhou H J 2021 IEEE Access 9 69307 Google Scholar
[17]	Zervoudakis K, Tsafarakis S 2020 Comput. Ind. Eng. 145 106559 Google Scholar
[18]	Ding Y H, You W B 2020 IEEE Access 8 207089 Google Scholar
[19]	Nickabadi A, Ebadzadeh M M, Safabakhsh R 2011 Appl. Soft Comput. 11 3658 Google Scholar
[20]	Syama S, Ramprabhakar J, Anand R, Guerrero J M 2023 Results Eng. 19 101274 Google Scholar
[21]	Liu N, Luo F, Ding W C 2019 2019 IEEE Symposium Series on Computational Intelligence (SSCI) Xiamen, China, December 6–9, 2019 p3104
[22]	Pan P Y, Xing Y H, Zhang D W, Wang J, Liu C L, Wu D, Wang X Y 2023 J. Food Sci. 88 3189 Google Scholar
[23]	Elahi N, Duncan R W, Stasolla C 2016 Plant Physiol. Biochem. 100 52 Google Scholar

[1]	居学尉, 张林烽, 黄峰, 朱国锋, 李淑锦, 陈燕青, 王嘉勋, 钟舜聪, 陈盈, 王向峰. 数字型太赫兹带通滤波器的逆向设计及优化. , 2024, 73(6): 060702. doi: 10.7498/aps.73.20231584
[2]	刘泉澄, 杨富, 张祺, 段勇威, 邓琥, 尚丽平. 太赫兹光谱学研究CL-20/MTNP共晶振动特性. , 2024, 73(19): 193201. doi: 10.7498/aps.73.20240944
[3]	冯龙呈, 杜琛, 杨圣新, 张彩虹, 吴敬波, 范克彬, 金飚兵, 陈健, 吴培亨. 太赫兹实时近场光谱成像研究. , 2022, 71(16): 164201. doi: 10.7498/aps.71.20220131
[4]	王志全, 施卫. 太赫兹时域光谱中脉冲太赫兹波全息探测. , 2022, 71(18): 188704. doi: 10.7498/aps.71.20220983
[5]	王晨, 夏威, 索鹏, 王伟, 林贤, 郭艳峰, 马国宏. 准二维范德瓦耳斯本征铁磁半导体CrGeTe₃的THz光谱. , 2022, 71(23): 237303. doi: 10.7498/aps.71.20221586
[6]	索鹏, 夏威, 张文杰, 朱晓青, 国家嘉, 傅吉波, 林贤, 郭艳峰, 马国宏. 准二维范德瓦耳斯磁性半导体CrSiTe₃的THz光谱. , 2020, 69(20): 207302. doi: 10.7498/aps.69.20200682
[7]	连宇翔, 戴泽林, 许向东, 谷雨, 李欣荣, 王福, 杨春, 成晓梦, 周华新. 有机电光晶体4-(4-二甲基氨基苯乙烯基)甲基吡啶对甲基苯磺酸盐的太赫兹光谱研究. , 2017, 66(24): 244211. doi: 10.7498/aps.66.244211
[8]	闫微, 马淼, 戴泽林, 谷雨, 朱宏钊, 刘禹彤, 许向东, 韩守胜, 彭勇. 全反式-胡萝卜素太赫兹光谱的实验及理论研究. , 2017, 66(3): 037801. doi: 10.7498/aps.66.037801
[9]	宋丹, 樊晓平, 刘钟理. 一种基于非基因信息的免疫记忆优化算法. , 2015, 64(14): 140203. doi: 10.7498/aps.64.140203
[10]	鹿文亮, 娄淑琴, 王鑫, 申艳, 盛新志. 基于太赫兹时域光谱技术的伪色彩太赫兹成像的实验研究. , 2015, 64(11): 114206. doi: 10.7498/aps.64.114206
[11]	孙怡雯, 钟俊兰, 左剑, 张存林, 但果. 血凝素蛋白及抗体相互作用的太赫兹光谱主成分分析. , 2015, 64(16): 168701. doi: 10.7498/aps.64.168701
[12]	高文, 汤洋, 朱明. 复杂背景下目标检测的级联分类器算法研究. , 2014, 63(9): 094204. doi: 10.7498/aps.63.094204
[13]	梁美彦, 张存林. 相位补偿算法对提高太赫兹雷达距离像分辨率的研究. , 2014, 63(14): 148701. doi: 10.7498/aps.63.148701
[14]	张铠云, 杜海伟, 陈民, 盛政明. 基于光场离化电流机制产生强太赫兹辐射的参数优化研究. , 2012, 61(16): 160701. doi: 10.7498/aps.61.160701
[15]	张戎, 郭旭光, 曹俊诚. 太赫兹量子阱光电探测器光栅耦合的模拟与优化. , 2011, 60(5): 050705. doi: 10.7498/aps.60.050705
[16]	陈扬, 贾丽萍, 张太宁, 郭澎, 王湘晖, 常胜江. 用于光谱分类的光谱峰谷沿匹配算法. , 2010, 59(1): 271-280. doi: 10.7498/aps.59.271
[17]	侯碧辉, 菅彦珍, 王雅丽, 张尔攀, 傅佩珍, 汪力, 钟任斌. PbB4O7 晶体的太赫兹光谱和软光学声子. , 2010, 59(7): 4640-4645. doi: 10.7498/aps.59.4640
[18]	王卫宁. 苏氨酸的太赫兹及拉曼光谱研究. , 2009, 58(11): 7640-7645. doi: 10.7498/aps.58.7640
[19]	王卫宁, 李元波, 岳伟伟. 组氨酸和精氨酸的太赫兹光谱研究. , 2007, 56(2): 781-785. doi: 10.7498/aps.56.781
[20]	岳伟伟, 王卫宁, 赵国忠, 张存林, 闫海涛. 芳香族氨基酸的太赫兹光谱研究. , 2005, 54(7): 3094-3099. doi: 10.7498/aps.54.3094

计量

文章访问数: 2419
PDF下载量: 45
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板