食品与发酵工业 >

2023 , Vol. 49 >Issue 8: 258 - 264

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.032253

分析与检测

基于近红外光谱和多层感知机的贻贝中腹泻性贝毒快速无损检测

  • 刘忠艳 ,
  • 刘瑶 ,
  • 乔付 ,
  • 郝博麟 ,
  • 姜微 ,
  • 熊建芳
展开
  • 1(岭南师范学院 计算机与智能教育学院,广东 湛江,524048)
    2(岭南师范学院 电子与电气工程学院,广东 湛江,524048)
第一作者：博士,副教授(通信作者,E-mail:912968544@qq.com)

收稿日期: 2022-05-07

  修回日期: 2022-06-10

  网络出版日期: 2023-05-16

基金资助

广东省自然科学基金项目(2020A1515011368;2021A1515012440);国家自然科学基金青年科学基金项目(62005109);岭南师范学院红树林研究院课题(PYXM04);岭南师范学院人才专项(ZL2054);岭南师范学院自然科学研究项目(ZL1902;ZL2007);广东省哲学社会科学规划学科共建项目(GD22XJY32)

收起

Rapid non-destructive detection of diarrheal shellfish poison in mussels based on near-infrared spectroscopy and multi-layer perceptron

  • LIU Zhongyan ,
  • LIU Yao ,
  • QIAO Fu ,
  • HAO Bolin ,
  • JIANG Wei ,
  • XIONG Jianfang
Expand
  • 1(School of Computer Science and Intelligence Education, Lingnan Normal University, Zhanjiang 524048, China)
    2(School of Electronic and Electrical Engineering, Lingnan Normal University, Zhanjiang 524048, China)

Received date: 2022-05-07

  Revised date: 2022-06-10

  Online published: 2023-05-16

Fold

摘要

以腹泻性贝毒(diarrheal shellfish poison,DSP)污染和未污染良好贻贝为研究对象,利用近红外光谱仪采集950~1 700 nm波长内各120个样本的光谱数据;分析确定适合贻贝近红外光谱(near-infrared spectroscopy,NIS)的最佳预处理方法来消除环境光的影响;构建多层感知机(multi-layer perceptron,MLP)模型作为检测DSP污染贻贝的分类器。将240个样本构成的数据集按照7∶3随机划分为训练集和测试集,将运行50次模型得到的准确率的平均值作为最终评价指标,检测DSP污染贻贝的准确率达到99.94%。该研究所构建的基于NIS的MLP模型对DSP的检出限为35 μg/kg。对于实际应用中存在的数据集中训练集所占比重不同、小样本和类别不均衡等问题,分析了MLP模型的检测性能。实验结果表明,基于一阶导数光谱预处理的MLP模型对以上3种问题不敏感,说明该分类器泛化能力及鲁棒性较强。因此,将NIS与MLP分类器结合是一种可行的贝毒无损鉴别的新方法。

关键词： 近红外光谱; 腹泻性贝毒; 贻贝; 多层感知机; 光谱预处理

本文引用格式

刘忠艳 , 刘瑶 , 乔付 , 郝博麟 , 姜微 , 熊建芳 . 基于近红外光谱和多层感知机的贻贝中腹泻性贝毒快速无损检测[J]. 食品与发酵工业, 2023 , 49(8) : 258 -264 . DOI: 10.13995/j.cnki.11-1802/ts.032253

Abstract

In this study, diarrheal shellfish poison (DSP)-contaminated and non-contaminated mussels were used as the research objects, the near-infrared spectrometer was used to collect the spectral data of 120 mussel samples of each class in the wavelength range of 950-1 700 nm. The best preprocessing method for near-infrared spectroscopy (NIS) of the mussels was determined to eliminate the influence of ambient light. Multi-layer perceptron (MLP) model was constructed as a classifier to detect DSP-contaminated mussels. The dataset composed of 240 samples was randomly divided into training and test datasets according to the ratio of 7∶3, the average accuracy of the model by running 50 times was the final evaluation index, and the accuracy of detecting DSP-contaminated mussels reached 99.94%. The detection limit of the MLP model based on NIS for DSP was 35 μg/kg. The detection performance of the MLP model was analyzed for the problems of different ratios of training sets, small sample datasets, and unbalanced classes in practical application. The experimental results showed that the MLP model based on first derivative spectral preprocessing was insensitive to these three problems, which indicated that the classifier had strong generalization ability and robustness. Therefore, the combination of NIS and the MLP classifier provided a feasible new method for the non-destructive identification of shellfish toxicity.

Key words： near-infrared spectroscopy; diarrheal shellfish poison; mussels; multi-layer perceptron; spectral preprocessing

参考文献

[1] 耿倩倩, 郭萌萌, 李风铃, 等.BDE-47在紫贻贝中的分布、蓄积、消除和毒性效应[J].中国环境科学, 2022, 42(3):1 385-1 393.
GENG Q Q, GUO M M, LI F L, et al.Tissue distribution, accumulation, elimination characteristics and toxicity of 2,2’,4,4’-tetrabromodiphenyl ether in blue mussel[J].China Environmental Science, 2022, 42(3):1 385-1 393.
[2] 王运选, 李春爱, 黎承, 等.贝类食品中腹泻性贝类毒素检测能力验证结果分析[J].现代食品, 2022, 28(1):182-187.
WANG Y X, LI C A, LI C, et al.Analysis of verification results of detection ability of diarrheal shellfish toxins in shellfish food[J].Modern Food, 2022, 28(1):182-187.
[3] 彭志兰, 罗海军, 王维洁, 等.舟山海域麻痹性贝类毒素污染情况及其2种检测方法比较[J].食品安全质量检测学报, 2017, 8(4):1 436-1 440.
PENG Z L, LUO H J, WANG W J, et al.Investigation of paralytic shellfish poison in waters of Zhoushan and comparison of 2 detection methods[J].Journal of Food Safety and Quality, 2017, 8(4):1 436-1 440.
[4] 张晓萌, 秦鸣蔚, 赵新月, 等.鹅膏毒肽类毒素检测方法的研究进展[J].食品与发酵工业, 2020, 46(19):295-300.
ZHANG X M, QIN M W, ZHAO X Y, et al.Research progress on detection methods of amanita cyclopeptide toxins[J].Food and Fermentation Industries, 2020,46(19):295-300.
[5] 姚建华, 谭志军, 周德庆.液相色谱-串联质谱检测贝类组织中5种脂溶性贝毒素[J].分析化学, 2010, 38(12):1 714-1 720.
YAO J H, TAN Z J, ZHOU D Q.Determination of five lipophilic marine toxins in shellfish by liquid chromatography with tandem mass spectrometry[J].Chinese Journal of Analytical Chemistry, 2010, 38(12):1 714-1 720.
[6] 田华. 全球食品领域近红外光谱应用研究文献计量分析[J].食品工业科技, 2021, 42(18):41-47.
TIAN H.Bibliometric analysis of near-infrared spectroscopy in global food areas[J].Science and Technology of Food Industry, 2021, 42(18):41-47.
[7] 李凯, 李雪莹, 栾丽丽, 等.近红外光谱和多分类器融合的葡萄酒品种判别研究[J].光谱学与光谱分析, 2016, 36(11):3 547-3 551.
LI K, LI X Y, LUAN L L, et al.Determination of wine varieties with NIR and fusion of multiple classifiers[J].Spectroscopy and Spectral Analysis, 2016, 36(11):3 547-3 551.
[8] 夏吉安, 杨余旺, 曹宏鑫, 等.云计算的蚕豆虫害可见-近红外光谱分类[J].光谱学与光谱分析, 2018, 38(3):756-760.
XIA J A, YANG Y W, CAO H X, et al.Classification of broad bear pest of visible-near infrared spectroscopy based on cloud computing[J].Spectroscopy and Spectral Analysis, 2018, 38(3):756-760.
[9] 王劭晟, 田绪红, 邱少健, 等.基于近红外光谱结合机器学习的鳕鱼品种二分类方法研究[J].食品安全质量检测学报, 2021, 12(22):8 651-8 659.
WANG S S, TIAN X H, QIU S J, et al.Research on cod species binary classification method based on near infrared spectroscopy and machine learning[J].Journal of Food Safety and Quality, 2021, 12(22):8 651-8 659.
[10] LIU Y, XU L L, WANG R T, et al.Study on the detection of heavy metal lead (Pb) in mussels based on near-infrared spectroscopy technology and a REELM classifier.Microchemical Journal, 2022.DOI:10.1016/j.microc.2022.107394.
[11] 孟庆龙, 冯树南, 尚静, 等.基于可见/近红外光谱预测枇杷糖度及模型优化[J].食品与发酵工业, 2022,48(12):249-254.
MENG Q L, FENG S N, SHANG J, et al.The establishment and optimization of the model for predicting the sugar content of loquat by Vis/NIR spectroscopy and its optimization[J].Food and Fermentation Industries, 2022,48(12):249-254.
[12] 张立欣, 杨翠芳, 陈杰, 等.基于变量优选和近红外光谱技术的红富士苹果产地溯源[J].食品与发酵工业, 2022,48(20):36-43.
ZHANG L X, YANG C F, CHEN J, et al.Tracing the origin of Red Fuji apple based on variable optimization and near-infrared spectroscopy[J].Food and Fermentation Industries, 2022,48(20):36-43.
[13] 袁旭林. 基于高光谱成像技术的苹果糖度无损检测系统研究[D].济南:山东大学, 2021.
YUAN X L.Research on nondestructive detection system of apple sugar content based on hyperspectral imaging technology[D].Jinan:Shandong University, 2021.
[14] 伍臣鹏. 基于高光谱成像技术的猕猴桃品质无损检测方法研究[D].南昌:江西农业大学, 2020.
WU C P.Research on nondestructive detection quality of kiwifruit based on hyperspectral imaging technology[D].Nanchang:Jiangxi Agricultural University, 2020.
[15] MURLIDHAR B R, NGUYEN H, ROSTAMI J, et al.Prediction of flyrock distance induced by mine blasting using a novel Harris Hawks optimization-based multi-layer perceptron neural network[J].Journal of Rock Mechanics and Geotechnical Engineering, 2021, 13(6):1 413-1 427.
[16] 花蕊. 深度学习技术在阿尔茨海默症分类预测中的应用[D].南京:东南大学, 2019.
HUA R.Application of deep learning technology in classification and prediction of Alzheimer's disease[D].Nanjing:Southeast University, 2019.
[17] 徐玲玲, 迟冬祥.面向不平衡数据集的机器学习分类策略[J].计算机工程与应用, 2020, 56(24):12-27.
XU L L, CHI D X.Machine learning classification strategy for imbalanced data sets[J].Computer Engineering and Applications, 2020, 56(24):12-27.
[18] BOUDIBI S, SAKAA B, BENGUEGA Z, et al.Spatial prediction and modeling of soil salinity using simple cokriging, artificial neural networks, and support vector machines in El Outaya plain, Biskra, southeastern Algeria[J].Acta Geochimica, 2021, 40(3):390-408.
[19] COVER T.Estimation by the nearest-neighbor rule[J].IEEE Transactions on Information Theory, 1968, 14(1):50-55.
[20] BREIMAN L.Random forests[J].Machine Learning, 2001, 45(1):5-32.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司