Objective Aiming to address the challenge of effectively capturing and reliably discriminating the subtle spectral changes caused by early infection of tobacco mosaic virus (TMV),a precise hyperspectral re-cognition framework based on physiologically sensitive feature selection and interpretable machine learning was developed,to provide technical support for early TMV detection and precise control. Method Hyperspectral (400-1 000 nm) reflectance data of healthy and early TMV infected tobacco leaves were collected.After pretreatment with whiteboard correction and region of interest extraction,fifty-two candidate vegetation indices, including normalized pigment chlorophyll index (NPCI) and water index (WI),were calculated.Recursive feature elimination (RFE) was used to select the optimal feature subset.Based on the selected subset,four categories of machine-learning models,including k-nearest neighbors (KNN) and support vector machine (SVM), were developed and comparatively evaluated.Hyperparameters were optimized using random search and grid search,and model generalization performance was assessed using an independent test set.Furthermore,shapley additive explanations (SHAP) was introduced to interpret the best-performing model by quantifying the contribution of key features to classification decisions. Result A dual-feature subset consisting of NPCI and WI was obtained through RFE screening,demonstrating that pigment and water-related changes are key information for early TMV identification,which effectively mitigates the curse of dimensionality in hyperspectral data.The KNN model built on the dual-feature subset achieved the best performance,with receiver operating characteristic-area under curve (ROC-AUC) of 0.987,five-fold cross-validation accuracy (ACC) of 0.955 and balanced F score (F1-score) of 0.961.On the independent test set,the ROC-AUC remained as high as 0.960,indicating good generalization performance.SHAP analysis further showed that NPCI and WI were the dominant drivers of model decision-making:an increase in their values positively drove samples to be classified as “TMV-infected”,and its variation was consistent with TMV-induced chlorophyll degradation and water-stress processes. Conclusion Early and precise identification of TMV can be achieved using only two vegetation indices (NPCI and WI), substantially reducing model complexity.The proposed “RFE-based feature selection+KNN+SHAP” methodological framework not only improves identification performance but also reveals an early spectral response mechanism dominated by pigment changes and supplemented by water status variations.This provides a reference for the early and rapid identification of TMV,as well as for the development of low-cost, portable TMV monitoring equipment for tobacco.
采用受试者工作特征曲线下面积(receiver ope-rating characteristic-area under curve,ROC-AUC)、准确率(accuracy,ACC)及F1分数(balanced F score,F1-score)作为核心评价指标,分别评估模型的综合判别能力、整体预测精度及正负样本的平衡性。
CULVERJ N, ALWYNG C, LINDBECKG C,et al.Virus-host interactions:induction of chlorotic and necrotic responses in plants by tobamoviruses[J].Annual Review of Phytopathology,1991,29:193-217.
[2]
BALACHANDRANS, HURRYV M, KELLEYS E,et al.Concepts of plant biotic stress:some insights into the stress physiology of virus‐infected plants, from the perspective of photosynthesis[J].Physiologia Plantarum,1997,100(2):203-213.
[3]
MAHLEINA K, OERKEE C, STEINERU,et al.Recent advances in sensing plant diseases for precision crop protection[J].European Journal of Plant Pathology,2012,133(1):197-209.
[4]
LAWALB Y.基于高光谱成像技术的烟叶病害识别方法研究[D].杭州:浙江大学,2012.
[5]
LAWALB Y.Study on tobacco leaf disease recognition methods based on hyperspectral imaging technology[D].Hangzhou:Zhejiang University,2012.
XUD Y, LIX J, YANGY H,et al.Study on monitoring mosaic virus infected tobacco based on remote sensing technology[J].Acta Tabacaria Sinica,2016,22(1):76-83.
[8]
GUQ, SHENGL, ZHANGT,et al.Early detection of tomato spotted wilt virus infection in tobacco using the hyperspectral imaging technique and machine learning algorithms[J].Computers and Electronics in Agriculture,2019,167:105066.
[9]
HAAGSMAM, HAGERTYC H, KROESED R,et al.Detection of soil-borne wheat mosaic virus using hyperspectral imaging: from lab to field scans and from hyperspectral to multispectral data[J].Precision Agriculture,2023,24(3):1030-1048.
[10]
ZHUH, CENH, ZHANGC,et al.Early detection and classification of tobacco leaves inoculated with tobacco mosaic virus based on hyperspectral imaging technique[C]//2016 ASABE Annual International Meeting.Orlando,Florida:American Society of Agricultural and Biological Engineers,2016:1-7.
[11]
XUB, LIX, HOUW,et al.A similarity-based ranking method for hyperspectral band selection[J].IEEE Transactions on Geoscience and Remote Sensing,2021,59(11):9585-9599.
[12]
YANK, GAOS, YANG,et al.A global systematic review of the remote sensing vegetation indices[J].International Journal of Applied Earth Observation and Geoinformation,2025,139:104560.
[13]
KAMENOVAI, DIMITROVP.Evaluation of Sentinel-2 vegetation indices for prediction of LAI,fAPAR and fCover of winter wheat in Bulgaria[J].European Journal of Remote Se-nsing,2021,54(1):89-108.
[14]
LOWEA, HARRISONN, FRENCHA P.Hyperspectral image analysis techniques for the detection and classification of the early onset of plant disease and stress[J].Plant Methods,2017,13(1):80.
[15]
MAHLEINA K, RUMPFT, WELKEP,et al.Development of spectral indices for detecting and identifying plant diseases[J].Remote Sensing of Environment,2013,128:21-30.
[16]
MUSTAFAG, ZHENGH, LIUY,et al.Leveraging machine learning to discriminate wheat scab infection levels through hyperspectral reflectance and feature selection methods[J].European Journal of Agronomy,2024,161:127372.
[17]
WATTM S, BUDDENBAUMH, LEONARDOE M C,et al.Using hyperspectral plant traits linked to photosynthetic efficiency to assess N and P partition[J].ISPRS Journal of Photogrammetry and Remote Sensing,2020,169:406-420.
[18]
JOSHIP, SANDHUK S, DHILLONG S,et al.Detection and monitoring wheat diseases using unmanned aerial vehicles (UAVs)[J].Computers and Electronics in Agriculture,2024,224:109158.
ZHANGN, YANGG J, ZHAOC J,et al.Progress and prospects of hyperspectral remote sensing technology for crop diseases and pests[J].National Remote Sensing Bulletin,2021,25(1):403-422.
[21]
ZHANGH, ZHAOJ, HUANGL,et al.Development of new indices and use of CARS-Ridge algorithm for wheat fusarium head blight detection using in-situ hyperspectral data[J].Biosystems Engineering,2024,237:13-25.
[22]
MORELL-MONZÓS, SEBASTIÁ-FRASQUETM T, ESTORNELLJ,et al.Detecting abandoned citrus crops using Sentinel-2 time series:a case study in the comunitat Valenciana region (Spain)[J].ISPRS Journal of Photogrammetry and Remote Sensing,2023,201:54-66.
[23]
GUYONI, WESTONJ, BARNHILLS,et al.Gene selection for cancer classification using support vector machines[J].Machine Learning,2002,46(1):389-422.
[24]
ZHAON, HUANGY, MEIY,et al.Random forest-guided decoding of multiplex role of initial brine salinity in shaping dichotomous equilibrium of microbial assembly and flavor compound in liquid phase and solid phase of radish Paocai during fermentation[J].Food Chemistry,2025,502:147699.
[25]
FALAMAKIA, SHAFIEEA H, SHAFIEEA.Assessing liquefaction potential using Arias intensity:a logistic regression approach with an updated database[J].Soil Dynamics and Ear-thquake Engineering,2026,202:110028.
[26]
PARKM, SOMBORNA, SCHLEHUBERD,et al.Non-destructive quantification of lutein and beta-carotene in spinach by Raman spectroscopy under optimized conditions for linear discriminant analysis[J].Food Chemistry,2025,493(4):146062.
[27]
PRIYAS R K, BALAMBIGAR K, MISHRAP,et al.Sugarcane yield forecast using weather based discriminant analysis[J].Smart Agricultural Technology,2023,3:100076.
[28]
MONDALD, KOLED K, ROYK.Gradation of yellow mosaic virus disease of okra and bitter gourd based on entropy based binning and Naive Bayes classifier after identification of leaves[J].Computers and Electronics in Agriculture,2017,142:485-493.
[29]
RAYK K, KUMARIA, KUMARS,et al.Guava leaf disease detection using support vector machine (SVM)[J].Smart Agricultural Technology,2025,12:101190.
[30]
DENGN, XUR, ZHANGY,et al.Forest biomass carbon stock estimates via a novel approach:K-nearest neighbor-based weighted least squares multiple birth support vector regression coupled with whale optimization algorithm[J].Computers and Electronics in Agriculture,2025,232:110020.
[31]
HANIFM F, VAN DAMMEL G W, VANDEN HOLEC,et al.The right fit:a decision tree to select a protocol for assessing the welfare of laying hens[J].Poultry Science,2025,104(12):105935.
[32]
ZHAIX, LIUY, HONGY,et al.Improved digital mapping of soil texture using the kernel temperature-vegetation dryness index and adaptive boosting[J].Ecological Informatics,2025,87:103083.
[33]
DUZ, YANGL, ZHANGD,et al.Corn variable-rate seeding decision based on gradient boosting decision tree model[J].Computers and Electronics in Agriculture,2022,198:107025.
[34]
ANDRADE-AMBRIZY A, PERÉZ-GARCÍAV, SERRA-NO-ARELLANOJ,et al.Application of eXtreme Gradient Boosting in the performance prediction of a refrigeration syste-m working with alternative refrigerants[J].International Journal of Refrigeration,2026,183:254-264.
[35]
NIC, HUANGH, CUIP,et al.Light Gradient Boosting Machine (LightGBM) to forecasting data and assisting the defrosting strategy design of refrigerators[J].International Journal of Refrigeration,2024,160:182-196.
[36]
YUANY, LINS, DONGX,et al.Construction of a multilayer perceptron based intelligent platform for dynamic quality monitoring and shelf-life prediction of bivalves[J].Food Che-mistry,2025,502:147717.
[37]
LUNDBERGS M, ERIONG, CHENH,et al.From local explanations to global understanding with explainable AI for trees[J].Nature Machine Intelligence,2020,2(1):56-67.
[38]
GRIMMERM K, JOHN FOULKESM, PAVELEYN D. Foliar pathogenesis and plant water relations:a review[J].Journal of Experimental Botany,2012,63(12):4321-4331.
[39]
DAS S, PRATIHERS, KYALC,et al.Sparsity regularized deep subspace clustering for multicriterion-based hyperspectral band selection[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2022,15:4264-4278.
[40]
GRIFFELL M, DELPARTED, EDWARDSJ.Using support vector machines classification to differentiate spectral signatures of potato plants infected with potato virus Y[J].Computers and Electronics in Agriculture,2018,153:318-324.
[41]
ARAÚJOM C U, SALDANHAT C B, GALVAOR K H,et al.The successive projections algorithm for variable selection in spectroscopic multicomponent analysis[J].Chemometrics and Intelligent Laboratory Systems,2001,57(2):65-73.