Rapid determination of oil content of single peanut seed by near-infrared hyperspectral imaging

Oil content is a crucial indicator for evaluating the quality of peanuts.A rapid and non-destructive method to determine oil content of individual peanut seed can provide robust technical support for breeding high-oil-content peanut varieties.In this study,we established a rapid determination method using near-infrared hyperspectral imaging and chemometrics to assess the oil content of single peanut seed.After selecting key wavelengths through competitive adaptive reweighted sampling(CARS),uninformative variable elimination(UVE),and random frog(RF),we constructed an oil content calibration model based on partial least squares regression for single peanut seed.Validation results demonstrated that the correlation coefficient was 0.8393 with a root mean square error of 1.7771 in the calibration set,while it was 0.7915 with a root mean square error of 2.2943 in the independent prediction set.Most samples exhibited relative errors below 5%,confirming the reliability of this model in predicting oil content of single peanut seed.

Application of Hyperspectral Imaging Technology in Rapid Detection of Preservative in Milk

To ensure the quality and safety of pure milk,detection method of typical preservative-potassium sorbate in milk was researched in this paper.Hyperspectral imaging technology was applied to realize rapid detection.Influence factors for hyperspectral data collection for milk samples were firstly researched,including height of sample,bottom color and sample filled up container or not.Pretreatment methods and variable selection algorithms were applied into original spectral data.Rapid detection models were built based on support vector machine method(SVM).Finally,standard normalized variable(SNV)-competitive adaptive reweighted sampling(CARS)and SVM model was chosen in this paper.The accuracies of calibration set and testing set were 0.97 and 0.97,respectively.Kappa coefficient of the model was 0.93.It could be seen that hyperspectral imaging technology could be used to detect for potassium sorbate in milk.Meanwhile,it also provided methodological supports for the rapid detection of other preservatives in milk.

Observation of Atherosclerotic Plaque Phantoms through Saline or Blood Layers by Near-Infrared Hyperspectral Imaging

We observed atherosclerotic plaque phantoms using a novel near-infrared (NIR) hyperspectral imaging (HSI) technique. Data were obtained through saline or blood layers to simulate an angioscopic environment for the phantom. For the study, we developed a NIR-HSI system with an NIR supercontinuum light source and mercury-cadmium-telluride camera. Apparent spectral absorbance was obtained at wavelengths of 1150 - 2400 nm. Hyperspectral images of lipid were constructed using a spectral angle mapper algorithm. Bovine fat covered with saline or blood was observed using hyperspectral images at a wavelength around 1200 nm. Our results show that NIR-HSI is a promising angioscopic technique with the potential to identify lipid-rich plaques without clamping and saline injection.

Detection and Discrimination of Tea Plant Stresses Based on Hyperspectral Imaging Technique at a Canopy Level

Tea plant stresses threaten the quality of tea seriously.The technology corresponding to the fast detection and differentiation of stresses is of great significance for plant protection in tea plantation.In recent years,hyperspectral imaging technology has shown great potential in detecting and differentiating plant diseases,pests and some other stresses at the leaf level.However,the lack of studies at canopy level hampers the detection of tea plant stresses at a larger scale.In this study,based on the canopy-level hyperspectral imaging data,the methods for identifying and differentiating the three commonly occurred tea stresses(i.e.,the tea leafhopper,anthrax and sun burn)were studied.To account for the complexity of the canopy scenario,a stepwise detecting strategy was proposed that includes the process of background removal,identification of damaged areas and discrimination of stresses.Firstly,combining the successive projection algorithm(SPA)spectral analysis and K-means cluster analysis,the background and overexposed non-plant regions were removed from the image.Then,a rigorous sensitivity analysis and optimization were performed on various forms of spectral features,which yielded optimal features for detecting damaged areas(i.e.,YSV,Area,GI,CARI and NBNDVI)and optimal features for stresses discrimination(i.e.,MCARI,CI,LCI,RARS,TCI and VOG).Based on this information,the models for identifying damaged areas and those models for discriminating different stresses were established using K-nearest neighbor(KNN),Random Forest(RF)and Fisher discriminant analysis.The identification model achieved an accuracy over 95%,and the discrimination model achieved an accuracy over 93%for all stresses.The results suggested the feasibility of stress detection and differentiation using canopy-level hyperspectral imaging techniques,and indicated the potential for its extension over large areas.

Research on the detection of early caries based on hyperspectral imaging

Objective:We applied hyperspectral imaging(HSI)system to distinguish early caries from soundand pigmented areas.It will provide a theoretical basis and technical support,for research anddevelopment of an instrument that could be used for screening and detection of early dentalcaries.Methods:Eighteen extracted human teeth(molars and premolars),with varying degrees ofnatural pathology and no degree of decay involving dentin were obtained.HSI system with awavelength range from 400 to 1000nm was used to obtain images of all 18 teeth containingsound,carious and pigmented areas.We compared the spectra of the wavebands at both 500 nmand 780 nm from the different tooth states,and the reflectance diference bet ween sound versuscarious lesions and sound versus pigmented areas,respectively.Results:There was a slight diference in refectance bet ween carious areas and pigmented areas at500 nm.A substantial difference was additionally noted in refectance bet ween carious areas andpigmented areas at 780 nm.Conclusion:The results have shown that the interference of tooth surface pigment can be elim-inated in the near-infrared(NIR)waveband,and the caries can be effectively identifed from the pigmented areas.Thus,it could be used to detect carious areas of teeth in place of the traditionalvisual inspection method or white light endoscopy.Clinical significance:The NIR difused light signal enables the identification of early caries frompigment and other interference,providing a reasonable detection tool for early detection andearly treatment of teeth diseases.

In-situ monitoring of saccharides removal of alcohol precipitation using near-infrared spectroscopy

As unsafe components in herbal medicine(HM),saccharides can affect not only the drug appearance and stabilization,but also the drug efficacy and safety.The present study focuses on the in-line monitoring of batch alcohol precipitation processes for saccharide removal using nearinfrared(NIR)spectroscopy.NIR spectra in the 4000–10,000-cm^(-1)wavelength range are acquired in situ using a transflectance probe.These directly acquired spectra allow characterization of the dynamic variation tendency of saccharides during alcohol precipitation.Calibration models based on partial least squares(PLS)regression have been developed for the three saccharide impurities,namely glucose,fructose,and sucrose.Model errors are estimated as the root-meansquare errors of cross-validation(RMSECVs)of internal validation and root-mean-square errors of prediction(RMSEPs)of external validation.The RMSECV values of glucose,fructose,and sucrose were 1.150,1.535,and 3.067 mg·mL^(-1),and the RMSEP values were 0.711,1.547,and 3.740 mg·mL^(-1),respectively.The correlation coeffcients(r)between the NIR predictive and the reference measurement values were all above 0.94.Furthermore,NIR predictions based on the constructed models improved our understanding of sugar removal and helped develop a control strategy for alcohol precipitation.The results demonstrate that,as an alternative process analytical technology(PAT)tool for monitoring batch alcohol precipitation processes,NIR spectroscopy is advantageous for both efficient determination of quality characteristics(fast,in situ,and requiring no toxic reagents)and process stability,and evaluating the repeatability.

Hyperspectral Images-Based Crop Classification Scheme for Agricultural Remote Sensing

Hyperspectral imaging is gaining a significant role in agricultural remote sensing applications.Its data unit is the hyperspectral cube which holds spatial information in two dimensions while spectral band information of each pixel in the third dimension.The classification accuracy of hyperspectral images(HSI)increases significantly by employing both spatial and spectral features.For this work,the data was acquired using an airborne hyperspectral imager system which collected HSI in the visible and near-infrared(VNIR)range of 400 to 1000 nm wavelength within 180 spectral bands.The dataset is collected for nine different crops on agricultural land with a spectral resolution of 3.3 nm wavelength for each pixel.The data was cleaned from geometric distortions and stored with the class labels and annotations of global localization using the inertial navigation system.In this study,a unique pixel-based approach was designed to improve the crops'classification accuracy by using the edge-preserving features(EPF)and principal component analysis(PCA)in conjunction.The preliminary processing generated the high-dimensional EPF stack by applying the edge-preserving filters on acquired HSI.In the second step,this high dimensional stack was treated with the PCA for dimensionality reduction without losing significant spectral information.The resultant feature space(PCA-EPF)demonstrated enhanced class separability for improved crop classification with reduced dimensionality and computational cost.The support vector machines classifier was employed for multiclass classification of target crops using PCA-EPF.The classification performance evaluation was measured in terms of individual class accuracy,overall accuracy,average accuracy,and Cohen kappa factor.The proposed scheme achieved greater than 90%results for all the performance evaluation metrics.The PCA-EPF proved to be an effective attribute for crop classification using hyperspectral imaging in the VNIR range.The proposed scheme is well-suited for practical applications of crops and landfill estimations using agricultural remote sensing methods.

Grain Protein Content Phenotyping in Rice via Hyperspectral Imaging Technology and a Genome-Wide Association Study

Efficient and accurate acquisition of the rice grain protein content(GPC)is important for selecting high-quality rice varieties,and remote sensing technology is an attractive potential method for this task.However,the majority of multispectral sensors are poor predictors of GPC due to their broad spectral bands.Hyperspectral technology provides a new analytical technology for bridging the gap between phenomics and genomics.However,the small size of typical datasets is a constraint for model construction for estimating GPC,limiting their accuracy and reducing their ability to generalize to a wide range of varieties.In this study,we used hyperspectral data of rice grains from 515 japonica varieties and deep convolution generative adversarial networks(DCGANs)to generate simulated data to improve the model accuracy.Features sensitive to GPC were extracted after applying a continuous wavelet transform(CWT),and the estimated GPC model was constructed by partial least squares regression(PLSR).Finally,a genome-wide association study(GWAS)was applied to the measured and generated datasets to detect GPC loci.The results demonstrated that the simulated GPC values generated after 8,000 epochs were closest to the measured values.The wavelet feature(WF_(1743,2)),obtained from the data with the addition of 200 simulated samples,exhibited the highest GPC estimation accuracy(R^(2)=0.58 and RRMSE=6.70%).The GWAS analysis showed that the estimated values based on the simulated data detected the same loci as the measured values,including the OsmtSSB1L gene related to grain storage protein.This study provides a new technique for the efficient genetic study of phenotypic traits in rice based on hyperspectral technology.

Identifying the geographical origin and processing technology of Moyao(Myrrh)on the basis of near-infrared spectroscopy combined with chemometrics

OBJECTIVE:To evaluate the quality of Moyao(Myrrh)in the identification of the geographical origin and processing of the products.METHODS:Raw Moyao(Myrrh)and two kinds of Moyao(Myrrh)processed with vinegar from three countries were identified using near-infrared(NIR)spectroscopy combined with chemometric techniques.Principal component analysis(PCA)was used to reduce the dimensionality of the data and visualize the clustering of samples from different categories.A classical chemometric algorithm(PLS-DA)and two machine learning algorithms[K-nearest neighbor(KNN)and support vector machine]were used to conduct a classification analysis of the near-infrared spectra of the Moyao(Myrrh)samples,and their discriminative performance was evaluated.RESULTS:Based on the accuracy,precision,recall rate,and F1 value in each model,the results showed that the classical chemometric algorithm and the machine learning algorithm obtained positive results.In all of the chemometric analyses,the NIR spectrum of Moyao(Myrrh)preprocessed by standard normal variation or Multivariate scattering correction combined with KNN achieved the highest accuracy in identifying the geographical origins,and the accuracy of identifying the processing technology established by the KNN method after first-order derivative pretreatment was the best.The best accuracy of geographical origin discrimination and processing technology discrimination were 0.9853 and 0.9706 respectively.CONCLUSIONS:NIR spectroscopy combined with chemometric technology can be an important tool for tracking the origin and processing technology of Moyao(Myrrh)and can also provide a reference for evaluations of its quality and the clinical use.

Modeling for mung bean variety classification using visible and near-infrared hyperspectral imaging

This study was carried out to investigate the feasibility of using visible and near infrared hyperspectral imaging for the variety classification of mung beans.Raw hyperspectral images of mung beans were acquired in the wavelengths of 380-1023 nm,and all images were calibrated by the white and dark reference images.The spectral reflectance values were extracted from the region of interest(ROI)of each calibrated hyperspectral image,and then they were treated as the independent variables.The dependent variables of four varieties of mung beans were set as 1,2,3 and 4,respectively.The extreme learning machine(ELM)model was established using full spectral wavelengths for classification.Modified gram-schmidt(MGS)method was used to identify effective wavelengths.Based on the selected wavelengths,the ELM and linear discriminant analysis(LDA)models were built.All models performed excellently with the correct classification rates(CCRs)covering 99.17%-99.58% in the training sets and 99.17%-100%in the testing sets.Fifteen wavelengths(432 nm,455 nm,468 nm,560 nm,705 nm,736 nm,760 nm,841 nm,861 nm,921 nm,930 nm,937 nm,938 nm,959 nm and 965 nm)were recommended by MGS.The results demonstrated that hyperspectral imaging could be used as a non-destructive method to classify mung bean varieties,and MGS was an effective wavelength selection method.

Grading method of soybean mosaic disease based on hyperspectral imaging technology

Soybean is a crop with a long cultivation history that occupies an important position in agricultural production.Soybean mosaic virus disease(SMV)has caused a rapid decline in soybean yields,causing huge losses to the soybean industry,wherefrom its early detec-tion is particularly important.This study proposes a new classification method for the early SMV,dividing its severity into grades 0,1 and 2.In the case of a small number of experi-mental samples of soybeans,this study proposes a combined convolutional neural network and support vector machine(CNN-SVM)method for the early detection of SMV.Experimen-tal results showed that the accuracy of the training set of the CNN-SVM model reached 96.67%,and the accuracy rate of the test set reached 94.17%.The experiment proved the feasibility of using the proposed CNN-SVM model to classify early SMV under the new clas-sification method,and provided a new direction for early SMV detection based on hyper-spectral images.

Rapid detection of total nitrogen content in soil based on hyperspectral technology

Soil total nitrogen content(TN)is a crucial factor in boosting the growth of crops.Its sur-plus or scarcity will alter the quality and yield of crops to a certain extent.Traditional methods such as chemical analysis is complicated,laborious and time-consuming.A faster and more efficient method to detect TN should be explored to address this problem.The hyperspectral technology integrates conventional energy and spectroscopy which aids in the simultaneous collection of spatial and spectral information from an object.It has grad-ually proved its significance and gained popularity in the analysis of soil composition.This study discussed the possibility of using hyperspectral technology to detect TN,analyzed six spectral data preprocessing methods and five modeling methods:partial least squares(PLS),back-propagation(BP)neural network,radial basis function(RBF)neural network,extreme learning machine(ELM)and support vector regression(SVR)with evaluation index R^(2) and RMSE.Setting the content of chemical analysis as the control and comparing the errors from spectral analysis.According to the results,all five models can be used for TN detection,and the SVR model with R^(2) 0.9121 and RMSE 0.7581 turned to the best method.The study showed that the spectral model can detect TN quickly,providing a reference for the detection of elements in soil with favorable research significance.

Hyperspectral Imaging Technology for Revealing the Original Handwritings Covered by the Same Inks

This manuscript presents a preliminary investigation on the applicability of hyperspectral imaging technology for nondestructive and rapid analysis to reveal covered original handwritings.The hyperspectral imager Nuance‑Macro was used to collect the reflected light signature of inks from the overlapping parts.The software Nuance1p46 was used to analyze the reflected light signature of inks which shows the covered original handwritings.Different types of black/blue ballpoint pen inks and black/blue gel pen inks were chosen for sample preparation.From the hyperspectral images examined,the covered original handwritings of application were revealed in 90.5%,69.1%,49.5%,and 78.6%of the cases.Further,the correlation between the revealing effect and spectral characteristics of the reflected light of inks at the overlapping parts was interpreted through theoretical analysis and experimental verification.The results indicated that when the spectral characteristics of the reflected light of inks at the overlapping parts were the same or very similar to that of the ink that was used to cover the original handwriting,the original handwriting could not be shown.On the contrary,when the spectral characteristics of the reflected light of inks at the overlapping parts were different to that of the ink that was used to cover the original handwriting,the original handwriting was revealed.

光谱技术在果品无损检测中的应用研究进展

介绍了常用的果品无损检测方法,包括电子鼻法、X射线计算机断层成像法、机械方法、磁共振(核磁共振)成像光谱学法、拉曼光谱学法、近红外光谱技术等;重点综述了高光谱成像技术在无损检测水果结构特性、生化成分和安全性等方面的研究进展,以期为水果品质的无损检测研究提供参考。

基于高光谱成像技术的南果梨酸度无损检测方法

南果梨是一种重要的水果品种,其酸度是评估果品质量的重要指标之一。然而,传统的南果梨酸度检测方法通常需要破坏性采样和化学分析,不仅耗时费力,而且容易导致样品污染和浪费。因此,旨在探索一种基于高光谱成像技术的无损检测方法,以实现对南果梨酸度的快速、准确、无损检测。首先,采集室温20℃下不同贮藏天数南果梨的高光谱数据,其光谱波长范围为400~1000 nm,并且通过理化实验测量南果梨样本的可滴定酸;其次,采用多元散射校正(multipli⁃cative scatter correction,MSC)、标准正态变换(standard normal variate,SNV)、Savitzky-Golay平滑滤波等多种方法对光谱数据进行预处理,建立偏最小二乘回归模型(partial least squares regression,PLSR),选择出建模效果最佳的预处理方法,结果显示MSC方法效果最优;然后结合连续投影算法(successie projection algorithm,SPA)提取特征波段,在700~900 nm范围内确定9个特征光谱变量;最后,以提取出的9个特征光谱变量作为输入矢量,分别建立PLSR模型、极限学习机(extreme learning machine,ELM)模型以及遗传算法(genetic algorithm,GA)和粒子群算法(particle swarm op⁃timization,PSO)优化的BP神经网络模型。研究结果表明,基于MSC预处理和SPA算法特征提取的PSO-BP模型预测精度最高,效果最好,预测集决定系数R^(2)_(p)=0.911,RMSEP=0.032。可见,基于高光谱成像技术的SPA-PSO-BP模型可用于南果梨酸度的检测,为南果梨的品质评价提供参考。

高光谱偏振技术的研究进展及展望

高光谱偏振技术是一种融合了高光谱和偏振成像的新兴技术,其在多个科学领域成为研究热点。本文旨在全面综述高光谱偏振技术的研究进展,并展望其未来发展方向。首先介绍了高光谱偏振技术的基本原理,解释了高光谱和偏振成像相结合的优势。然后,根据不同的设计原理介绍了偏振光谱仪器的分类。接下来详细讨论了该技术在遥感、医学、环境监测、地球科学和材料科学等领域的广泛应用。通过对不同领域的案例研究进行梳理,展示了高光谱偏振技术在提供更为丰富、精确信息方面的独特优势。最后对高光谱偏振技术目前面临的挑战进行了分析,包括仪器设备的精密性、数据处理的复杂性以及与其他传感设备的有效融合的问题。针对这些挑战,探讨了未来技术发展方向。未来的研究应着重于提升该技术的高光谱和时间分辨率,提高数据处理和分析准确性,扩展不同应用场景的适用性,以更好地满足不同领域的需求。综合而言,高光谱偏振技术作为一种全面、高效的信息获取手段,在多个领域取得了显著的研究进展。通过优化高光谱偏振技术满足更宽广的应用领域,高光谱偏振技术有望成为未来科学研究和实际应用中的重要工具。

基于人工智能技术的高光谱人脸自动化识别系统设计

为有效识别人脸区域,提升其在多种领域应用效果,设计基于人工智能技术的高光谱人脸自动化识别系统。以模块化思想设计嵌入式系统架构,采集与预处理高光谱人脸图像,并将预处理后的图像数据放入RAM存储器;人脸检测模块调用RAM存储器存储数据,并加载Haar人脸分类器,完成人脸区域检测提取工作;之后由人脸特征提取与识别模块经人脸区域LBP特征提取、LeNet-5卷积神经网络人脸识别模型构建与训练等操作,输出人脸识别结果。实验结果表明,该系统能够在较短时间内完成LeNet-5卷积神经网络人脸识别模型训练。

基于希尔伯特滤波的可擦笔油墨光谱模式识别

文件的真实性是当前诉讼审查阶段的重要工作,可擦笔在司法案件中常被用来进行伪造文书、合同等犯罪行为。针对其油墨成分、笔迹修改等方面的辨识是文件检验领域的重点研究。特殊热感变色颜料是可擦笔油墨的主要成分,其变色原理是随着温度变化会产生笔迹的消失与复现,在65℃以上颜色褪去,在-18℃以下颜色复现。对其进行种属认定可以对案件证据的真实性进行鉴别,为案件诉讼过程提供支持。高光谱的超高光谱分辨率对高分子材料具有较好的特征选择性,能够有效的对常见油墨成分进行数据采集。该实验收集22个品牌共45份可擦笔油墨样本,可以分为碳化钨笔珠、子弹头笔珠、全针管、半针管四种类型,统一采集450~950 nm波段的高光谱信息。关于光谱数据背景噪声冗余的问题,选用主成分分析法(PCA)对数据进行降维处理,提取特征变量。基于降维后的数据选用不同类型的希尔伯特变换(HT)进行信号滤波,进一步选择有效信号,提升建模效果。样本识别上选用多层感知器(MLP)和径向基函数神经网络(radial basis function neural network,RBFNN)两种人工神经网络模型,基于23维主成分提取的特征变量类建模准确率分别为81%,84%,通过希尔伯特高通滤波处理后可以将分类准确率提升至88.9%,92%,能够有效提升识别准确率。为进一步区分不同样本的种类,选择Fisher判别分析方法进行建模,各样本原始数据在FDA模型中识别准确率为44%,经最优PCA-HT处理的FDA建模准确率为93.3%,能够区分出不同的可擦笔油墨类型。结果表明,PCA能够在保留光谱有效信息的基础上进行降维,提升模型精度并且缩短运行时间,相较于原始光谱数据建模效果较好,通过希尔伯特变换后的光谱数据能够进一步完善有效光谱信息,使得建模准确率进一步提升。该实验确定PCA-HT-FDA模型为最佳可擦笔油墨高光谱识别模型,能够为司法鉴定人员提供一定参考。

基于高光谱成像技术对菜心种子霉变的识别

为了鉴别健康与霉变菜心种子,本研究通过高光谱成像技术获得健康与霉变菜心种子光谱,建立判别模型。基于原始光谱和9种预处理后光谱建立支持向量机判别(SVM-DA)模型,发现基于一阶导数预处理后光谱的模型表现最出色,建模集和预测集的准确率分别为95.87%和95.74%。为了去除冗余或不必要的信息,采用遗传算法(GA)对原始光谱和各种预处理后光谱进行波段筛选,并再次建立SVM-DA模型,在这些模型中,FD-GA-SVM-DA模型性能最优,建模集和预测集准确率分别达97.71%和96.81%。研究表明,基于高光谱技术鉴别健康和霉变菜心种子具有可行性。

高光谱图像结合卷积神经网络的马铃薯干腐病潜育期识别

马铃薯是世界第四大粮食作物,具有丰富的营养价值。但其在贮藏和运输过程中易被镰刀真菌侵染而产生干腐病,最终造成巨大资源浪费和经济损失,因此实现马铃薯干腐病的早期快速无损检测是必要的。在样品被病原菌侵染时,经历了健康—潜育期—轻度病害—重度病害的阶段,其中潜育期的样品难以识别,主要源于病害发生时间较短,表面未形成肉眼可见的病斑,与健康样品相似。为了实现马铃薯干腐病潜育期的识别,结合高光谱成像和深度学习展开马铃薯干腐病早期诊断研究。以健康和不同腐败程度马铃薯为实验对象,获取健康和不同病害等级的马铃薯高光谱图像。然后基于ENVI人工选取健康部位和不同腐败程度样品的病斑部位为感兴趣区域(ROI),并计算ROI的平均光谱值作为该样品的最终光谱信息。以光谱数据作为输入变量,病害等级作为输出变量,建立卷积神经网络(CNN)模型,并对其网络结构进行优化,对比分析不同模型的预测结果,筛选出最优网络层模型为Model_3_3。并基于此结构进行学习率的优化,得到Model_0.0001识别效果最好,其总体准确率、精度、灵敏度和特异性分别为99.68%、99.76%、98.82%、99.54%。为了进一步突显CNN应用于马铃薯干腐病潜育期识别的优势,建立了最小二乘支持向量机(LS-SVM)、随机森林(RF)、K-近邻法(KNN)和线性判别分析(LDA)模型。结果显示,四种常规算法模型的识别准确率分别为90.77%、92.30%、93.10%和92.34%,其中对潜育期样品识别率分别为91.00%、85.58%、94.18%和90.33%。对于总体准确率,CNN模型较几种常规方法提高了6.58%~8.91%;对于潜育期样品的识别,CNN模型较常规方法提高了5.55%~14.15%。研究表明,高光谱成像技术结合CNN可以有效实现马铃薯干腐病潜育期识别,为提高马铃薯病害早期诊断的智能化水平提供了参考方法。