1.南京农业大学 工学院,江苏 南京 210031
2.南京农业大学 人工智能学院,江苏 南京 210031
3.南京市产品质量监督检验院 国家市场监管重点实验室(生物毒素分析与评价),江苏 南京 210019
4.南京农业大学 江苏省智能化农业装备重点实验室,江苏 南京 210031
梁 琨,博士,副教授,研究方向:农产品检测技术,E-mail:kliang@njau.edu.cn
扫 描 看 全 文
宋金鹏,梁琨,张驰等.基于深度学习与可见-近红外光谱的患腥黑穗病小麦籽粒分类研究[J].分析测试学报,2023,42(07):784-793.
SONG Jin-peng,LIANG Kun,ZHANG Chi,et al.Research on Classification of Common Bunt of Wheat Kernels Based on Visible-Near Infrared Spectroscopy Combined with Deep Learning Algorithms[J].Journal of Instrumental Analysis,2023,42(07):784-793.
宋金鹏,梁琨,张驰等.基于深度学习与可见-近红外光谱的患腥黑穗病小麦籽粒分类研究[J].分析测试学报,2023,42(07):784-793. DOI: 10.19969/j.fxcsxb.23030501.
SONG Jin-peng,LIANG Kun,ZHANG Chi,et al.Research on Classification of Common Bunt of Wheat Kernels Based on Visible-Near Infrared Spectroscopy Combined with Deep Learning Algorithms[J].Journal of Instrumental Analysis,2023,42(07):784-793. DOI: 10.19969/j.fxcsxb.23030501.
针对小麦腥黑穗病快速无损的检测需求,该文将可见-近红外光谱与深度学习算法结合建立了小麦腥黑穗病籽粒的分类模型。采用多元散射校正算法(MSC)和标准正态变换算法(SNV)对光谱进行预处理,消除光谱噪音的影响,分别使用竞争性自适应重加权算法(CARS)和随机蛙跳算法(RF)对预处理后的光谱进行特征波长提取。结果显示,特征提取算法可去除大量冗余信息,波段减少比率为93.7% ~ 94.2%,有效降低了模型运行成本,并可防止模型过拟合。结果显示:MSC + CARS + VGG16模型训练集的准确率为96.39%,测试集准确率为91.67%,取得了较好的分类结果。最终建立的VGG16深度学习模型实现了健康、轻度患病和重度患病3类小麦籽粒的分类。对比传统机器学习模型,VGG16模型能够充分提取光谱特征信息,更好地区分健康与轻度患病籽粒。该研究表明深度学习结合可见-近红外光谱方法,能够实现对不同患病程度腥黑穗病小麦籽粒的有效分类,为腥黑穗病小麦籽粒的快速无损检测提供了一种新方法。
In view of the need for rapid and non-destructive detection of common bunt,a visible-near infrared spectroscopy combined with deep learning algorithms was proposed to construct a classification model for identifying common bunt of wheat kernels in this paper.The spectral data were preprocessed by multiple scattering correction(MSC) and standard normal variate(SNV) algorithms to eliminate noise.Meanwhile,feature wavelength extraction was performed on the preprocessed spectra using the competitive adaptive reweighted sampling(CARS) and random frog(RF) algorithms.The results indicated that the feature extraction algorithms effectively removed redundant information,resulting in a reduction of 93.7% to 94.2% in the number of spectral bands. This significantly reduced the model’s computational cost and prevented overfitting.The achieved results demonstrated the effectiveness of the MSC + CARS + VGG16 model with an accuracy of 96.39% for the training set and 91.67% for the test set,providing satisfactory classification outcomes.The final VGG16 deep learning model successfully categorized wheat kernels into three classes:healthy,light suscept,and heavy suscept.Compared to traditional machine learning models,the VGG16 model demonstrates a superior capability in extracting spectral features,and could more accurately distinguish healthy wheats and light suscept wheats.Results showed that visible-near infrared spectroscopy combined with deep learning could realize the effective classification on the wheat kernels with different degrees of common bunt,providing a novel approach for the rapid and non-destructive detection on common bunt wheat kernels.
可见-近红外光谱深度学习腥黑穗病小麦
visible-near infrared spectroscopydeep learningcommon buntwheat
Muhae-Ud-Din G,Chen D L,Liu T G,Chen W Q,Gao L.Sci. Rep.,2020,10(1):9209.
Lin S K,Wang N W,Weng R Q,Zhou W C,Guo Q X.Plant Quar. (林授锴,王念武,翁瑞泉,周卫川,郭琼霞.植物检疫),2008,30(5):318-320.
He T,Xu T S,Muhae-Ud-Din G,Guo Q Y,Liu T G,Chen W Q,Gao L.Biology (Basel),2022,11(6):865.
Shen T,Xiao Y,Su S Q,Jia Y F,Guo Q L,Liu Q,Xiao Y G.Mol. Plant Breed. (沈潼,肖越,粟神强,贾玉凤,郭巧玲,刘琦,肖永贵.分子植物育种),2023,21(8):1-17.
Singh A,Knox R E,De Pauw R M,Singh A K,Cuthbert R D,Kumar S,Campbell H L.Theor. Appl. Genet.,2016,129(2):243-256.
Qiu Z L.Study on Synergistic Toxicity and Metabonomics of Human Respiratory Tract Cells Induced by Trimethylamine and Its Metabolites. Guangzhou:Guangdong University of Technology(仇智霖.三甲胺及其代谢物诱导人呼吸道细胞的协同毒性效应及代谢组学研究. 广州:广东工业大学),2021.
Liu X Z,Li J S,Cheng J,Yu Y Q.J. Tianjin Inst. Light Ind. (刘显曾,李金生,成坚,于艳琴.天津轻工业学院学报),1992,7(1):39-44.
Bishnoi S K,He X Y,Phuke R M,Kashyap P L,Alakonya A,Chhokar V.Front. Plant Sci.,2020,11:569057.
Fang Y,Ramasamy R P.Biosensors (Basel),2015,5(3):537-561.
Shen Z H.Mod. Agric. Sci. Technol. (沈志河.现代农业科技),2008,491(21):148-151.
Zhang Q Q,Men X Y,Hui C,Ge F,Ouyang F.Agric. Ecosyst. Environ.,2022,326:107821.
Zhang H.RNA-Seq Analysis of Tilletia Controversa Kühninfection Wheat Grains. Shihezi:Shihezi University(张菡.小麦矮腥黑粉菌侵染小麦穗部的RNA-Seq分析. 石河子:石河子大学),2020.
Wang Y,Lu X Y,Wang Y,Gao R F,Xiang C Y,Xu L,Zhang G M,Huang H Q.Plant Quar. 汪莹,卢小雨,王颖,高瑞芳,向才玉,徐浪,章桂明,黄河清.植物检疫),2020,34(6):45-49.
GB/T 18085-2000. Plant Quarantine-Methods for Inspection and Identification of Tilletia Controversa Kühn. National Standards of the People’s Republic of China(植物检疫 小麦矮化腥黑穗病菌检疫鉴定方法. 中华人民共和国国家标准).
GB/T 28080-2011. Detection and Identification of Tilletia Indica Mitra. National Standards of the People’s Republic of China(小麦印度腥黑穗病菌检疫鉴定方法. 中华人民共和国国家标准).
Liu J J,Zhang J M,Chen W Q,Liu T G,Gao L.Jiangsu Agric. Sci. (刘俭俭,张建民,陈万权,刘太国,高利.江苏农业科学),2020,48(16):126-130.
Cao X R,Zhan H Y,Zhou Y L,Duan X Y.J. Biosaf. (曹学仁,詹浩宇,周益林,段霞瑜.生物安全学报),2011,20(2):171-174.
Guo Z Y,Zhang J,Ma C Y,Yin X,Guo Y M,Sun X,Jin C Q.J. Food Compos. Anal.,2022,116:105048.
Du Z,Tian W,Tilley M,Wang D,Zhang G,Li Y.Compr. Rev. Food Sci. Food Saf.,2022,21(3):2956-3009.
Wu T T,Yu K Q,Zhang H H,Feng Y,Zhang X,Wang H H.Spectrosc. Spectral Anal. (吴婷婷,余克强,张海辉,冯毅,张晓,汪辉辉.光谱学与光谱分析),2018,38(2):3912-3916.
Liu S,Tan X,Liu C Y,Zhu C L,Li W H,Cui S,Du Y F,Huang D C,Xie F.Spectrosc. Spectral Anal. 刘爽,谭鑫,刘成玉,朱春霖,李文昊,崔帅,杜懿峰,黄殿成,谢锋.光谱学与光谱分析),2019,39(11):3540-3546.
Shen F,Zhao T X,Jiang X S,Liu X,Fang Y,Liu Q,Hu Q H,Liu X Q.LWT,2019,109:216-224.
Jie Z W,Liu Z,Wang J F,Gu K S,Wang Z Y.China Oils Fats(接昭玮,刘卓,王继芬,古锟山,王之宇.中国油脂),2023,48(1):79-83,93.
Bejani M M,Ghatee M.IEEE Trans. Intell. Transp. Syst.,2020,21(2):543-552.
Liu S,Wang S,Hu C,Kong D,Yuan Y.Spectrochim. Acta A,2023,291:122371.
Qiu R,Zhao Y,Kong D,Wu N,He Y.Spectrochim. Acta A,2023,285:121838.
Chen G X,Zhou S B,Chen Q,Liu Y S,Zhao L L,Han W.Spectrosc. Spectral Anal. (陈国喜,周松斌,陈颀,刘忆森,赵路路,韩威.光谱学与光谱分析),2022,42(12):3811-3816.
Huang Y Q,Liu X H,Huang Z Y,Qian W Q,Liu S Y,Qiao X.Trans. Chin. Soc. Agric. Mach. (黄亦其,刘祥焕,黄震宇,钱万强,刘双印,乔曦.农业机械学报),2023,54(4):259-267.
Sheng Y,Jiao J,Teng Y,Ma X,Li C,Jiang L.J. Anal. Sci. (圣阳,焦俊,滕燕,马鑫,李春,蒋玲.分析科学学报),2022,38(5):552-560.
Lu X C,Su C L,He J Y.Mod. Agric. (陆小成,苏聪莉,何俊英.现代化农业),2022,44(2):22-23.
Matanguihan J B,Murphy K M,Jones S S.Plant Dis.,2011,95(2):92-103.
Salas E A L,Henebry G M.Israel J. Plant Sci.,2012,60(1/2):65-76.
Basati Z,Jamshidi B,Rasekh M,Abbaspour-Gilandeh Y.Spectrochim. Acta A,2018,203:308-314.
0
浏览量
17
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构