Random frog: an efficient reversible jump Markov Chain Monte Carlo-like approach for variable selection with applications to gene selection and disease classification

Quantification of antibiotics in food by octahedral gold-silver nanocages-based SERS sensor coupling multivariate calibration

The abuse of antibiotics has caused gradually increases drug-resistant bacterial strains that pose health risks. Herein, a sensitive SERS sensor coupled multivariate calibration was proposed for quantification of antibiotics in milk. Initially, octahedral gold-silver nanocages (Au@Ag MCs) were synthesized by Cu2O template etching method as SERS substrates, which enhanced the plasmonic effect through sharp edges and hollow nanostructures. Afterwards, five chemometric algorithms, like partial least square (PLS), uninformative variable elimination-PLS (UVE-PLS), competitive adaptive reweighted sampling-PLS (CARS-PLS), random frog-PLS (RF-PLS), and convolutional neural network (CNN) were applied for TTC and CAP. RF-PLS performed optimally for TTC and CAP (Rc = 0.9686, Rp = 0.9648, RPD = 3.79 for TTC and Rc = 0.9893, Rp = 0.9878, RPD = 5.88 for CAP). Furthermore, the detection limit of 0.0001 µg/mL for both TTC and CAP was obtained. Finally, satisfactory (p > 0.05) results were obtained with the standard HPLC method. Therefore, SERS combined RF-PLS could be applied for fast, nondestructive sensing of TTC and CAP in milk.

An integrated CBLA-Net with fractional discrete wavelet transform and frequency-based CARS to predict heavy metal elements by XRF

BACKGROUND

X-ray fluorescence (XRF) emerges as a promising technique for estimating heavy metal elements. However, XRF spectra typically contain a significant amount of environmental information and signal noise, and the relationship between spectral intensity and element concentration is difficult to quantify using a single model, thereby reducing the predictive performance for low concentration elements.

RESULTS

This paper proposed a comprehensive framework for predicting elemental concentrations, encompassing preprocessing, variable selection, decision-making, to enable fast, non-destructive, and accurate estimation of element concentrations in soil. Firstly, an optimal denoising method based on fractional discrete wavelet transform (FDWT) was introduced to enhance signal quality. Furthermore, the frequency-based competitive adaptive reweighted sampling (FCARS) algorithm was employed for feature selection of XRF spectral variables, allowing extraction of the most informative features from the complex spectral data. Finally, a novel deep learning network, called ConvBiLSTM-Attention (CBLA-Net), was designed to achieve precise estimation of heavy metal elements concentration. Compared with other advanced algorithms, The CBLA-Net demonstrated the highest accuracy for V, Cr, Mn, Zn, Cd, and Pb, achieving the coefficient of determination (R2) of 0.9730, 0.9874, 0.9952, 0.9921, 0.9518, and 0.9741, respectively. The CBLA-Net not only effectively extract local features and capture global information, but also combines attention mechanism to focus on key information.

SIGNIFICANCE

The proposed novel deep learning quantitative framework, including preprocessing, feature selection, and CBLA-Net decision-making, significantly enhances the accuracy of elemental content prediction. It provides a new approach for accurately assessing the concentration of heavy metal elements in soil.

Using Spectroradiometry to Measure Organic Carbon in Carbonate-Containing Soils

This study explores the feasibility of analyzing soil organic carbon (SOC) in carbonate-rich soils using visible near-infrared spectroscopy (VIS-NIR). Employing a combination of datasets, feature groups, variable selection methods, and regression models, 22 modeling pipelines were developed. Spectral data and spectral data combined with carbonate contents were used as datasets, while raw reflectance, first-derivative (FD) reflectance, and second-derivative (SD) reflectance constituted the feature groups. The variable selection methods included Spearman correlation, Variable Importance in Projection (VIP), and Random Frog (Rfrog), while Partial Least Squares Regression (PLSR), Random Forest Regression (RFR), and Support Vector Regression (SVR) were the regression models. The obtained results indicated that the FD preprocessing method combined with RF, results in the model that is sufficiently robust and stable to be applied to soils rich in calcium carbonate.

Rapid quality evaluation and geographical origin recognition of ginger powder by portable NIRS in tandem with chemometrics

Ginger powder is an important spice that is susceptible to improper sales such as adulteration or geographical fraud. In this study, a portable near infrared spectroscopy was used to quantitatively predict the 6-gingerol content, an important quality index of ginger, as well as to identify the gingers from three origins in China. Specifically, the optimal preprocessing method was first investigated by comparing the predictions of models. Then three feature variable selection methods including PCA, CARS, and RFrog, on the quantitative analysis of 6-gingerol were also compared, respectively. After comparison, the PLS model established on the S-G combined with SNV preprocessing outperformed the others. The PLS regression of 6-gingerol with variables selected by RFrog possessed the Rc2 of 0.9463, Rp2 of 0.9497, and the RPD of 4.2257, respectively. Moreover, the results further verified that the LDA model by SPA variables extraction successfully identify gingers from different origins with 100 % accuracy.

Determination of seawater COD spectra using double-loop contraction and sorted frog optimization

This study develops a novel double-loop contraction and C value sorting selection-based shrinkage frog-leaping algorithm (double-contractive cognitive random field [DC-CRF]) to mitigate the interference of complex salts and ions in seawater on the ultraviolet-visible (UV-Vis) absorbance spectra for chemical oxygen demand (COD) quantification. The key innovations of DC-CRF are introducing variable importance evaluation via C value to guide wavelength selection and accelerate convergence; a double-loop structure integrating random frog (RF) leaping and contraction attenuation to dynamically balance convergence speed and efficiency. Utilizing seawater samples from Jiaozhou Bay, DC-CRF-partial least squares regression (PLSR) reduced the input variables by 97.5% after 1,600 iterations relative to full-spectrum PLSR, RF-PLSR, and CRF-PLSR. It achieved a test R2 of 0.943 and root mean square error of 1.603, markedly improving prediction accuracy and efficiency. This work demonstrates the efficacy of DC-CRF-PLSR in enhancing UV-Vis spectroscopy for rapid COD analysis in intricate seawater matrices, providing an efficient solution for optimizing seawater spectra.

Rapid evaluation of Ziziphi Spinosae Semen and its adulterants based on the combination of FT-NIR and multivariate algorithms

Ziziphi Spinosae Semen (ZSS) is a valued seed renowned for its sedative and sleep-enhancing properties. However, the price increase has been accompanied by adulteration. In this study, chromaticity analysis and Fourier transform near-infrared (FT-NIR) combined with multivariate algorithms were employed to identify the adulteration and quantitatively predict the adulteration ratio. The findings suggested that the utilization of chromaticity extractor was insufficient for identification of adulteration ratio. The raw spectrum of ZMS and HAS adulterants extracted by FT-NIR was processed by SNV + CARS and 1d + SG + ICO respectively, the average accuracy of machine learning classification model was improved from 77.06 % to 97.58 %. Furthermore, the R2 values of the calibration and prediction set of the two quantitative prediction regression models of adulteration ratio are greater than 0.99, demonstrating excellent linearity and predictive accuracy. Overall, this study demonstrated that FT-NIR combined with multivariate algorithms provided a significant approach to addressing the growing issue of ZSS adulteration.

Application of spectroscopic techniques combined with chemometrics to the authenticity and quality attributes of rice

Rice is a staple food for two-thirds of the world's population and is grown in over a hundred countries around the world. Due to its large scale, it is vulnerable to adulteration. In addition, the quality attribute of rice is an important factor affecting the circulation and price, which is also paid more and more attention. The combination of spectroscopy and chemometrics enables rapid detection of authenticity and quality attributes in rice. This article described the application of seven spectroscopic techniques combined with chemometrics to the rice industry. For a long time, near-infrared spectroscopy and linear chemometric methods (e.g., PLSR and PLS-DA) have been widely used in the rice industry. Although some studies have achieved good accuracy, with models in many studies having greater than 90% accuracy. However, higher accuracy and stability were more likely to be obtained using multiple spectroscopic techniques, nonlinear chemometric methods, and key wavelength selection algorithms. Future research should develop larger rice databases to include more rice varieties and larger amounts of rice depending on the type of rice, and then combine various spectroscopic techniques, nonlinear chemometric methods, and key wavelength selection algorithms. This article provided a reference for a more efficient and accurate determination of rice quality and authenticity.

Rapid authentication of variants of Gastrodia elata Blume using near-infrared spectroscopy combined with chemometric methods

The variety is one of the most important factors to generate difference of chemical compositions, which unavoidably influences the quality of natural medicine. Thus, simple and rapid authentication of different variants has great academic and practical significance. In this study, the goal was achieved with the help of near infrared spectroscopy (NIR) and chemometrics by using Gastrodia elata Blume as an example. A total of 540 samples including two classes of variants and their forms were investigated as a whole. The mean spectra of samples of each class and their 2-D synchronous correlation spectra were simultaneously applied to discover the difference of chemical characteristics. After hybrid pre-processing of the first and second derivative combined with Savitzky-Golay and Norris filtering, partial least squares discrimination analysis (PLS-DA) on the basis of latent variable projection was used to assess the feasibility for classification. The results show higher prediction accuracy in both internal test set and external prediction set. In order to further improve the robustness for modeling, three methods for wavelength selection were comprehensively compared to optimize PLS-DA models, including variable importance in the projection (VIP), random frog (RF), and Monte Carlo uninformative variable elimination (MC-UVE). The prediction accuracy of combination of the 2nd derivative, Norris, MC-UVE and PLS-DA achieved to 99.11% and 98.89% corresponding to the internal test set and external prediction set, respectively. The strategies proposed in this work perform effectiveness for rapid and accurate authentication of variants of plants with high chemical complexity.

Non-destructive detection of protein content in mulberry leaves by using hyperspectral imaging

Protein content is one of the most important indicators for assessing the quality of mulberry leaves. This work is carried out for the rapid and non-destructive detection of protein content of mulberry leaves using hyperspectral imaging (HSI) (Specim FX10 and FX17, Spectral Imaging Ltd., Oulu, Finland). The spectral range of the HSI acquisition system and data processing methods (pretreatment, feature extraction, and modeling) is compared. Hyperspectral images of three spectral ranges in 400-1,000 nm (Spectral Range I), 900-1,700 nm (Spectral Range II), and 400-1,700 nm (Spectral Range III) were considered. With standard normal variate (SNV), Savitzky-Golay first-order derivation, and multiplicative scatter correction used to preprocess the spectral data, and successive projections algorithm (SPA), competitive adaptive reweighted sampling, and random frog used to extract the characteristic wavelengths, regression models are constructed by using partial least square and least squares-support vector machine (LS-SVM). The protein content distribution of mulberry leaves is visualized based on the best model. The results show that the best results are obtained with the application of the model constructed by combining SNV with SPA and LS-SVM, showing an R 2 of up to 0.93, an RMSE of just 0.71 g/100 g, and an RPD of up to 3.83 based on the HSI acquisition system of 900-1700 nm. The protein content distribution map of mulberry leaves shows that the protein of healthy mulberry leaves distributes evenly among the mesophyll, with less protein content in the vein of the leaves. The above results show that rapid, non-destructive, and high-precision detection of protein content of mulberry leaves can be achieved by applying the SWIR HSI acquisition system combined with the SNV-SPA-LS-SVM algorithm.

IsoFrog: a reversible jump Markov Chain Monte Carlo feature selection-based method for predicting isoform functions

MOTIVATION

A single gene may yield several isoforms with different functions through alternative splicing. Continuous efforts are devoted to developing machine-learning methods to predict isoform functions. However, existing methods do not consider the relevance of each feature to specific functions and ignore the noise caused by the irrelevant features. In this case, we hypothesize that constructing a feature selection framework to extract the function-relevant features might help improve the model accuracy in isoform function prediction.

RESULTS

In this article, we present a feature selection-based approach named IsoFrog to predict isoform functions. First, IsoFrog adopts a reversible jump Markov Chain Monte Carlo (RJMCMC)-based feature selection framework to assess the feature importance to gene functions. Second, a sequential feature selection procedure is applied to select a subset of function-relevant features. This strategy screens the relevant features for the specific function while eliminating irrelevant ones, improving the effectiveness of the input features. Then, the selected features are input into our proposed method modified domain-invariant partial least squares, which prioritizes the most likely positive isoform for each positive MIG and utilizes diPLS for isoform function prediction. Tested on three datasets, our method achieves superior performance over six state-of-the-art methods, and the RJMCMC-based feature selection framework outperforms three classic feature selection methods. We expect this proposed methodology will promote the identification of isoform functions and further inspire the development of new methods.

AVAILABILITY AND IMPLEMENTATION

IsoFrog is freely available at https://github.com/genemine/IsoFrog.

Application of the combination method based on RF and LE in near infrared spectral modeling

The dimensionality of near-infrared (NIR) spectral data is often extremely large. Dimensionality reduction of spectral data can effectively reduce the redundant information and correlation between spectral variables and simplify the model, which is crucial to increasing the model's performance. As a nonlinear feature extraction method, Laplacian Eigenmaps (LE) may preserve the local neighborhood information of the dataset, has high robustness, and is simple to compute. However, when the LE algorithm maps the data from high-dimensional space to low-dimensional space, it is often disturbed by irrelevant information and multicollinearity in the spectral data, which lowers the model's prediction performance. Random Frog (RF) algorithm can eliminate noise and collinearity in the spectrum. Therefore, before using the LE algorithm, we use the RF algorithm to eliminate irrelevant information in the spectrum and reduce the correlation between the spectra variables to increase the efficiency of the LE algorithm. We used the RF + LE algorithm to reduce the dimensionality of two public NIRS datasets (soil datasets and pharmaceutical tablets datasets) and compared it with RF and LE algorithms alone. We utilized Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR) to establish regression models. The experimental findings demonstrate that compared with the RF algorithm and LE algorithm, the RF + LE combination method can reduce the dimension of spectral variables and model complexity, and improve regression models' prediction accuracy and stability. It is an effective dimensionality reduction method for the near-infrared spectrum.

Method Comparison of Sample Pretreatment and Discovery of Differential Compositions of Natural Flavors and Fragrances for Quality Analysis by Using Chemometric Tools

Natural flavors and fragrances or their extracts have been widely used in a large variety of areas, including food, cosmetic, and tobacco industrial processes, among others. The compositions and intrinsic attributes of flavors and fragrances were related to many factors, such as species, geographical origin, planting environment, storage condition, processing method, and so on. This not only increased the difficulty in analyzing the product quality of flavors and fragrances, but also challenged the idea of "quality-by-design (QbD)". This work proposed an integrated strategy for precise discovery of differential compounds among different classes and subsequent quality analysis of complex samples through flavors and fragrances used in tobacco industry as examples. Three pretreatment methods were first inspected to effectively characterize the sample compositions, including direct injection (DI), thermal desorption (TD), and stir bar sorptive extraction (SBSE)-TD, coupled with gas chromatography-mass spectrometry (GC-MS) analysis to obtain characteristic information of samples of flavors and fragrances. Then, principal component analysis (PCA) was applied to discover the relation and difference between chromatographic fingerprints and peak table data once significant components were recognized in a holistic manner. Model population analysis (MPA) was then used to quantitatively extract the characteristic chemicals representing the quality differences among different classes of samples. Some differential marker compounds were discovered for difference analysis, including benzyl alcohol, latin acid, l-menthol acid, decanoic acid ethyl ester, vanillin, trans-o-coumaric acid, benzyl benzoate, and so on. Furthermore, partial least squares-discriminant analysis (PLS-DA) and support vector machine (SVM) were respectively applied to construct multivariate models for evaluation of quality differences and variations. It was found that the accuracy attains to 100% for sample classification. With the help of optimal sample pretreatment technique and chemometric methods, the strategy for quality analysis and difference discovery proposed in this work can be widely delivered to more areas of complex plants with good interpretability and high accuracy.

Determination of Corn Protein Content Using Near-Infrared Spectroscopy Combined with A-CARS-PLS

In order to quickly and accurately determine the protein content of corn, a new characteristic wavelength selection algorithm called anchor competitive adaptive reweighted sampling (A-CARS) was proposed in this paper. This method first lets Monte Carlo synergy interval PLS (MC-siPLS) to select the sub-intervals where the characteristic variables exist and then uses CARS to screen the variables further. A-CARS-PLS was compared with 6 methods, including 3 feature variable selection methods (GA-PLS, random frog PLS, and CARS-PLS) and 2 interval partial least squares methods (siPLS and MWPLS). The results showed that A-CARS-PLS was significantly better than other methods with the results: RMSECV = 0.0336, R² _c = 0.9951 in the calibration set; RMSEP = 0.0688, R² _p = 0.9820 in the prediction set. Furthermore, A-CARS reduced the original 700-dimensional variable to 23 variables. The results showed that A-CARS-PLS was better than some wavelength selection methods, and it has great application potential in the non-destructive detection of protein content in corn.

Rapid Determination of Geniposide and Baicalin in Lanqin Oral Solution by Near-Infrared Spectroscopy with Chemometric Algorithms during Alcohol Precipitation

The selection of key variables is an important step that improves the prediction performance of a near-infrared (NIR) real-time monitoring system. Combined with chemometrics, NIR spectroscopy was employed to construct high predictive accuracy, interpretable models for the rapid detection of the alcohol precipitation process of Lanqin oral solution (LOS). The variable combination population analysis-iteratively retaining informative variables (VCPA-IRIV) was innovatively introduced into the variable screening process of the model of geniposide and baicalin. Compared with the commonly used synergy interval partial least squares regression, competitive adaptive reweighted sampling, and random frog, VCPA-IRIV achieved the maximum compression of variable space. VCPA-IRIV-partial least squares regression (PLSR) only needs to use about 1% of the number of variables of the original data set to construct models with Rp values greater than 0.95 and RMSEP values less than 10%. With the advantages of simplicity and strong interpretability, the prediction ability of the PLSR models had been significantly improved simultaneously. The VCPA-IRIV-PLSR models met the requirements of rapid quality detection. The real-time detection system can help researchers to understand the quality rules of geniposide and baicalin in the alcohol precipitation process of LOS and provide a reference for the optimization of a LOS quality control system.

Feasibility of identifying the authenticity of fresh and cooked mutton kebabs using visible and near-infrared hyperspectral imaging

Mutton kebab is an attractive type of meat product with high nutritional value, and is favored by consumers worldwide. However, mutton kebab is often subjected to adulteration due to its high price. Chicken, duck, and pork are frequently used as adulterated substitutes. The purpose of current study aims at developing a methodology based on hyperspectral imaging (HSI, 400-1000 nm) for identifying the authenticity of fresh and cooked mutton kebabs. Kebab samples were individually scanned using HSI system in their fresh and cooked states. Spectra of chicken, duck, pork, and mutton kebabs were first extracted from representative regions of interest (ROIs) identified in their calibrated hyperspectral images. After that, principal component analysis (PCA) was carried out, and results showed that the first three or two PCs were effective for identifying fresh or cooked samples of different meat species. Different effective modeling algorithms including k-nearest neighbor (KNN), partial least squares discriminant analysis (PLS-DA), and support vector machine (SVM) algorithms combined with different preprocessing methods were employed to develop classification models. Performances exhibited that PLS-DA models using raw spectra outperformed the KNN and SVM models, and the accuracies reached both 100 % in prediction sets for fresh and cooked meat kebabs, respectively. Moreover, compared to iteratively variable subset optimization (IVSO), random frog (RF), and successive projections algorithm (SPA) algorithms, the PC loadings successfully screened 14 and 8 effective wavelengths for fresh and cooked meat kebabs, respectively, from the complex original full-band wavelengths. The PC-PLS-DA models showed the optimal predicted performances with overall classification accuracies of 97.5 % and 100 %, sensitivity values of 1.00 and 1.00, specificity values of 0.97 and 1.00, precisions of 0.91 and 1.00, for fresh and cooked mutton kebabs, respectively. Furthermore, the visualization of classification maps confirmed the experimental results intuitively. Overall, it was evident that HSI showed immense potential to identify the authenticity of fresh and cooked mutton kebabs when substituted by different meats including chicken, duck, and pork.

Estimation of chemical oxygen demand in different water systems by near-infrared spectroscopy

To monitor environmental water pollution effectively and meet human water needs, it is crucial to develop a fast, simple, and accurate method for monitoring chemical oxygen demand (COD) in various water systems. In this study, COD prediction models for different water systems were developed by combining near-infrared (NIR) spectroscopy with partial least squares regression (PLSR). Samples of wastewater, surface water, and seawater were collected from Guangzhou, Guangdong Province, China. Three pretreatment methods were used to preprocess the spectra in order to improve the accuracy and minimalism of the model. We investigate the performance of two variable selection algorithms, namely, binary gray wolf optimization (BGWO) and competitive adaptive reweighting sampling (CARS). The results show that both BGWO and CARS improved the performance of the model in terms of higher accuracy and less wavelength input; both of the combined model performances were better than that of PLSR alone, and CARS-PLSR achieved the best results. Using CARS-PLSR, surface water, wastewater, and seawater model inputs were reduced by 96 %, 96 %, and 82 % as compared to the PLSR results, respectively, and the testing sets R² reached 0.860, 0.815, and 0.692, respectively. The spectral variable selection algorithm could identify the important spectral variables between COD content and NIR spectra in three water systems, thereby improving the accuracy and simplicity of the PLSR model for COD prediction. Our results have important practical value for predicting COD content in different water systems by NIR spectroscopy.

Intelligent evaluation of free amino acid and crude protein content in raw peanut seed kernels using NIR spectroscopy paired with multivariable calibration

Given the nutritional importance of peanuts, this study examined the free amino acid (FAA) and crude protein (CP) content in raw peanut seeds. Near-infrared spectroscopy (NIRS) was employed in combination with variable selection algorithms after successful reference data analysis using colorimetric and Kjeldahl methods. Ensuing the application of partial least squares (PLS) as a full spectral model, the genetic algorithm (GA), bootstrapping soft shrinkage (BOSS), uninformative variable elimination (UVE), and random frog (RF) models were tested and assessed. A comparison of correlation coefficients of prediction (R_p), root mean square error of prediction (RMSEP), and residual predictive deviation (RPD) was performed to appraise the performance of the built models. Using RF-PLS, an unsurpassed outcome was achieved for FAA (R_p = 0.937, RPD = 3.38) and CP (R_p = 0.9261, RPD = 3.66). These findings demonstrated that NIR in combination with RF-PLS could be utilized for quantitative, rapid, and nondestructive prediction of FAA and CP in raw peanut seed samples.

Citrus Huanglongbing detection and semi-quantification of the carbohydrate concentration based on micro-FTIR spectroscopy

Citrus Huanglongbing (HLB) is nowadays one of the most fatal citrus diseases worldwide. Once the citrus tree is infected by the HLB disease, the biochemistry of the phloem region in midribs would change. In order to investigate the carbohydrate changes in phloem region of citrus midrib, the semi-quantification models were established to predict the carbohydrate concentration in it based on Fourier transform infrared microscopy (micro-FTIR) spectroscopy coupled with chemometrics. Healthy, asymptomatic-HLB, symptomatic-HLB, and nutrient-deficient citrus midribs were collected in this study. The results showed that the intensity of the characteristic peak varied with the carbohydrate (starch and soluble sugar) concentration in citrus midrib, especially at the fingerprint regions of 1175-900 cm^-1, 1500-1175 cm^-1, and 1800-1500 cm^-1. Furthermore, semi-quantitative prediction models of starch and soluble sugar were established using the full micro-FTIR spectra and selected characteristic wavebands. The least squares support vector machine regression (LS-SVR) model combined with the random frog (RF) algorithm achieved the best prediction result with the determination coefficient of prediction ([Formula: see text]) of 0.85, the root mean square error of prediction (RMSEP) of 0.36%, residual predictive deviation (RPD) of 2.54, and [Formula: see text] of 0.87, RMSEP of 0.37%, RPD of 2.76, for starch and soluble sugar concentration prediction, respectively. In addition, multi-layer perceptron (MLP) classification models were established to identify HLB disease, achieving the overall classification accuracy of 94% and 87%, based on the full-range spectra and the optimal wavenumbers selected by the random frog (RF) algorithm, respectively. The results demonstrated that micro-FTIR spectroscopy can be a valuable tool for the prediction of carbohydrate concentration in citrus midribs and the detection of HLB disease, which would provide useful guidelines to detect citrus HLB disease.

Identification of Four Chicken Breeds by Hyperspectral Imaging Combined with Chemometrics

The current study aims to explore the potential of the combination of hyperspectral imaging and chemometrics in the rapid identification of four chicken breeds. The hyperspectral data of four chicken breeds were collected in the range of 400–900 nm. Five pretreatment methods were used to pretreat the original spectra. The important characteristic wavelength variables were extracted by random frog (RF), successive projection algorithm (SPA), and competitive adaptive reweighted sampling (CARS) algorithms. The classification models were established by using support vector machine (SVM), k-nearest neighbor (KNN), and partial least squares-discriminant analysis (PLS-DA). The results showed that the mean normalization pretreatment method was preferable, and overall classification accuracy of SVM-based models was higher than that of KNN-based and PLS-DA-based models. The correct classification rate (CCR) of the full-spectrum SVM model (Full-SVM) could reach 96.25%. The SPA method extracted 13 important wavelengths, and the SVM model based on SPA (SPA-SVM) achieved 90% CCR. This study can provide a theoretical reference for the discriminant analysis of chicken breeds.

Soluble Solids Content Binary Classification of Miyagawa Satsuma in Chongming Island Based on Near Infrared Spectroscopy

Citrus is one of the most important fruits in China. Miyagawa Satsuma, one kind of citrus, is a nutritious agricultural product with regional characteristics of Chongming Island. Near-infrared Spectroscopy (NIR) is a proper method for studying the quality of fruits, because it is low-cost, efficient, non-destructive, and repeatable. Therefore, the NIR technique is used to detect citrus's soluble solid content (SSC) in this study. After obtaining the original spectral data, the first 70% of them are divided into the training set and 30% into the test set. Then, the Random Frog algorithm is chosen to select characteristic wavelengths, which reduces the dimension of the data and the complexity of the model, and accordingly makes the generalization of the classification model better. After comparing the performance of various classifiers (AdaBoost, KNN, LS-SVM, and Bayes) under different characteristic wavelength numbers, the AdaBoost classifier outperforms using 275 characteristic wavelengths for modeling eventually. The accuracy, precision, recall, and F ₁-score are 78.3%, 80.5%, 78.3%, and 0.780, respectively and the ROC (Receiver Operating Characteristic Curve, ROC curve) is close to the upper left corner, suggesting that the classification model is acceptable. The results demonstrate that it is feasible to use the NIR technique to estimate whether the citrus is sweet or not. Furthermore, it is beneficial for us to apply the obtained models for identifying the quality of citrus correctly. For fruit traders, the model helps them to determine the growth cycle of citrus more scientifically, improve the level of citrus cultivation and management and the final fruit quality, and thus increase the economic income of fruit traders.

Non-Invasive Detection of Anti-Inflammatory Bioactivity and Key Chemical Indicators of the Commercial Lanqin Oral Solution by Near Infrared Spectroscopy

Quality control methods of current traditional Chinese medicine (TCM) preparation is time-consuming and difficult to assess in terms of overall efficiency of the drug. A non-destructive rapid near-infrared spectroscopy detection system for key chemical components and biological activity of Lanqin oral solution (LOS), one of the best-selling TCM formulations, was established for comprehensive quality evaluation. Near infrared spectral scanning was carried out on 101 batches of commercial LOS under the penetrated vial state and traditional state. RAW 264.7 cells were cultured to detect the anti-inflammatory ability of LOS, and the reference concentrations of epigoitrin, geniposide, and baicalin were obtained by HPLC. The quantitative models were optimized by three kinds of variable selection methods. The correlation coefficients of prediction value of the models were greater than 0.94. The system also passed the external validation. The performance of the non-invasive models was similar to the traditional models. The established non-destructive system can be applied to the rapid quality inspection of LOS to avoid unqualified drugs from entering the market and ensure drug effectiveness. The biological activity index of LOS was introduced and predicted by NIRs for the first time, which provides a new idea about the quality control of TCM formulations.

Quantitative assessment of wheat quality using near-infrared spectroscopy: A comprehensive review

Wheat is one of the most widely cultivated crops throughout the world. A great need exists for wheat quality assessment for breeding, processing, and products production purposes. Near-infrared spectroscopy (NIRS) is a rapid, low-cost, simple, and nondestructive assessment method. Many advanced studies associated with NIRS for wheat quality assessment have been published recently, either introducing new chemometrics or attempting new assessment parameters to improve model robustness and accuracy. This review provides a comprehensive overview of NIRS methodology including its principle, spectra pretreatments, spectral wavelength selection, outlier disposal, dataset division, regression methods, and model evaluation. More importantly, the applications of NIRS in the determination of analytical parameters, rheological parameters, and end product quality of wheat are summarized. Although NIRS showed great potential in the quantitative determination of analytical parameters, there are still challenges in model robustness and accuracy in determining rheological parameters and end product quality for wheat products. Future model development needs to incorporate larger databases, integrate different spectroscopic techniques, and introduce cutting-edge chemometrics methods. In addition, calibration based on external factors should be considered to improve the predicted results of the model. The NIRS application in micronutrients needs to be extended. Last, the idea of combining standard product sensory attributes and spectra for model development deserves further study.

Research and Application Validation of a Feature Wavelength Selection Method Based on Acousto-Optic Tunable Filter (AOTF) and Automatic Machine Learning (AutoML)

Near-infrared spectroscopy has been widely applied in various fields such as food analysis and agricultural testing. However, the conventional method of scanning the full spectrum of the sample and then invoking the model to analyze and predict results has a large amount of collected data, redundant information, slow acquisition speed, and high model complexity. This paper proposes a feature wavelength selection approach based on acousto-optical tunable filter (AOTF) spectroscopy and automatic machine learning (AutoML). Based on the programmable selection of sub nm center wavelengths achieved by the AOTF, it is capable of rapid acquisition of combinations of feature wavelengths of samples selected using AutoML algorithms, enabling the rapid output of target substance detection results in the field. The experimental setup was designed and application validation experiments were carried out to verify that the method could significantly reduce the number of NIR sampling points, increase the sampling speed, and improve the accuracy and predictability of NIR data models while simplifying the modelling process and broadening the application scenarios.

Simultaneous quantification of deoxymyoglobin and oxymyoglobin in pork by Raman spectroscopy coupled with multivariate calibration

Because of the nutritional advantages and customer acceptance, it is vital to ensure pork meat quality. This study examined the quantification of myoglobin proportions (deoxymyoglobin and oxymyoglobin) by coupling Raman spectroscopy with efficient variables selection chemometrics. Prior to acquiring Raman spectroscopic data, the fractions of myoglobin were determined. Afterward, multivariate calibration methods like partial least square (PLS), competitive adaptive reweighted sampling (CARS-PLS), genetic algorithm-PLS (GA-PLS), and random frog-PLS (RF-PLS) were applied and evaluated. The models' performance was assessed using correlation coefficients of prediction (Rp), root mean square error of prediction (RMSEP), and residual predictive deviation (RPD). The RF-PLS model achieved optimal results for both deoxymyoglobin and oxymyoglobin, with Rp = 0.8936; RMSEP = 2.91 and RPD = 1.97 for the former and Rp = 0.9762; RMSEP = 1.23 and RPD = 4.47 for the latter, respectively. Therefore, this work demonstrated that Raman spectroscopy paired with RF-PLS could be employed for nondestructive, fast, and easy detection of deoxymyoglobin and oxymyoglobin.

Surface-Enhanced Raman Scattering Spectroscopy Combined With Chemical Imaging Analysis for Detecting Apple Valsa Canker at an Early Stage

Apple Valsa canker (AVC) with early incubation characteristics is a severe apple tree disease, resulting in significant orchards yield loss. Early detection of the infected trees is critical to prevent the disease from rapidly developing. Surface-enhanced Raman Scattering (SERS) spectroscopy with simplifies detection procedures and improves detection efficiency is a potential method for AVC detection. In this study, AVC early infected detection was proposed by combining SERS spectroscopy with the chemometrics methods and machine learning algorithms, and chemical distribution imaging was successfully applied to the analysis of disease dynamics. Results showed that the samples of healthy, early disease, and late disease sample datasets demonstrated significant clustering effects. The adaptive iterative reweighted penalized least squares (air-PLS) algorithm was used as the best baseline correction method to eliminate the interference of baseline shifts. The BP-ANN, ELM, Random Forest, and LS-SVM machine learning algorithms incorporating optimal spectral variables were utilized to establish discriminative models to detect of the AVC disease stage. The accuracy of these models was above 90%. SERS chemical imaging results showed that cellulose and lignin were significantly reduced at the phloem disease-health junction under AVC stress. These results suggested that SERS spectroscopy combined with chemical imaging analysis for early detection of the AVC disease was feasible and promising. This study provided a practical method for the rapidly diagnosing of apple orchard diseases.

Online Detection of Watercore Apples by Vis/NIR Full-Transmittance Spectroscopy Coupled with ANOVA Method

Watercore is an internal physiological disorder affecting the quality and price of apples. Rapid and non-destructive detection of watercore is of great significance to improve the commercial value of apples. In this study, the visible and near infrared (Vis/NIR) full-transmittance spectroscopy combined with analysis of variance (ANOVA) method was used for online detection of watercore apples. At the speed of 0.5 m/s, the effects of three different orientations (O1, O2, and O3) on the discrimination results of watercore apples were evaluated, respectively. It was found that O3 orientation was the most suitable for detecting watercore apples. One-way ANOVA was used to select the characteristic wavelengths. The least squares-support vector machine (LS-SVM) model with two characteristic wavelengths obtained good performance with the success rates of 96.87% and 100% for watercore and healthy apples, respectively. In addition, full-spectrum data was also utilized to determine the optimal two-band ratio for the discrimination of watercore apples by ANOVA method. Study showed that the threshold discrimination model established based on O3 orientation had the same detection accuracy as the optimal LS-SVM model for samples in the prediction set. Overall, full-transmittance spectroscopy combined with the ANOVA method was feasible to online detect watercore apples, and the threshold discrimination model based on two-band ratio showed great potential for detection of watercore apples.

Chemometrics in Tandem with Hyperspectral Imaging for Detecting Authentication of Raw and Cooked Mutton Rolls

Authentication assurance of meat or meat products is critical in the meat industry. Various methods including DNA- or protein-based techniques are accurate for assessing meat authenticity, however, they are destructive, expensive, or laborious. This study explores the feasibility of chemometrics in tandem with hyperspectral imaging (HSI) for identifying raw and cooked mutton rolls substitution by pork and duck rolls. Raw or cooked samples (n = 180) of three meat species were prepared to collect hyperspectral images in range of 400-1000 nm. Spectra were extracted from representative regions of interest (ROIs), and spectral principal component analysis (PCA) revealed that PC1 and PC2 were effective for the identification. Different methods including standard normal variable (SNV), first and second derivatives, and normalization were individually employed for spectral preprocessing, and modeling methods of partial least squares-discriminant analysis (PLS-DA) and support vector machines (SVM) were also individually applied to develop classification models for both the raw and the cooked. Results showed that PLS-DA model developed by raw spectra presented the highest 100% correct classification rate (CCR) of success in all sets. After that, effective wavelengths selected by successive projections algorithm (SPA) built optimal simplified models which didn't influence the modeling results compared with full spectra regardless of the meat roll states. Therefore, SPA-PLS-DA models were subsequently used to visualize the raw and cooked meat rolls classification. As a consequence, the general meat species of both raw and cooked meat rolls were readily discernible in pixel-wise manner by generating classification maps. The results showed that HSI combined with chemometrics can be used to identify the authentication of raw and cooked mutton rolls substituted by pork and duck rolls accurately. This promising methodology provides a reference which can be extended to the classification or grading of other meat rolls.

Rice Leaf Blast Classification Method Based on Fused Features and One-Dimensional Deep Convolutional Neural Network

Rice leaf blast, which is seriously affecting the yield and quality of rice around the world, is a fungal disease that easily develops under high temperature and humidity conditions. Therefore, the use of accurate and non-destructive diagnostic methods is important for rice production management. Hyperspectral imaging technology is a type of crop disease identification method with great potential. However, a large amount of redundant information mixed in hyperspectral data makes it more difficult to establish an efficient disease classification model. At the same time, the difficulty and small scale of agricultural hyperspectral imaging data acquisition has resulted in unrepresentative features being acquired. Therefore, the focus of this study was to determine the best classification features and classification models for the five disease classes of leaf blast in order to improve the accuracy of grading the disease. First, the hyperspectral imaging data were pre-processed in order to extract rice leaf samples of five disease classes, and the number of samples was increased by data augmentation methods. Secondly, spectral feature wavelengths, vegetation indices and texture features were obtained based on the amplified sample data. Thirdly, seven one-dimensional deep convolutional neural networks (DCNN) models were constructed based on spectral feature wavelengths, vegetation indices, texture features and their fusion features. Finally, the model in this paper was compared and analyzed with the Inception V3, ZF-Net, TextCNN and bidirectional gated recurrent unit (BiGRU); support vector machine (SVM); and extreme learning machine (ELM) models in order to determine the best classification features and classification models for different disease classes of leaf blast. The results showed that the classification model constructed using fused features was significantly better than the model constructed with a single feature in terms of accuracy in grading the degree of leaf blast disease. The best performance was achieved with the combination of the successive projections algorithm (SPA) selected feature wavelengths and texture features (TFs). The modeling results also show that the DCNN model provides better classification capability for disease classification than the Inception V3, ZF-Net, TextCNN, BiGRU, SVM and ELM classification models. The SPA + TFs-DCNN achieved the best classification accuracy with an overall accuracy (OA) and Kappa of 98.58% and 98.22%, respectively. In terms of the classification of the specific different disease classes, the F1-scores for diseases of classes 0, 1 and 2 were all 100%, while the F1-scores for diseases of classes 4 and 5 were 96.48% and 96.68%, respectively. This study provides a new method for the identification and classification of rice leaf blast and a research basis for assessing the extent of the disease in the field.

A Robust Machine Learning Based Framework for the Automated Detection of ADHD Using Pupillometric Biomarkers and Time Series Analysis

Accurate and efficient detection of attention-deficit/hyperactivity disorder (ADHD) is critical to ensure proper treatment for affected individuals. Current clinical examinations, however, are inefficient and prone to misdiagnosis, as they rely on qualitative observations of perceived behavior. We propose a robust machine learning based framework that analyzes pupil-size dynamics as an objective biomarker for the automated detection of ADHD. Our framework integrates a comprehensive pupillometric feature engineering and visualization pipeline with state-of-the-art binary classification algorithms and univariate feature selection. The support vector machine classifier achieved an average 85.6% area under the receiver operating characteristic (AUROC), 77.3% sensitivity, and 75.3% specificity using ten-fold nested cross-validation (CV) on a declassified dataset of 50 patients. 218 of the 783 engineered features, including fourier transform metrics, absolute energy, consecutive quantile changes, approximate entropy, aggregated linear trends, as well as pupil-size dilation velocity, were found to be statistically significant differentiators (p < 0.05), and provide novel behavioral insights into associations between pupil-size dynamics and the presence of ADHD. Despite a limited sample size, the strong AUROC values highlight the robustness of the binary classifiers in detecting ADHD—as such, with additional data, sensitivity and specificity metrics can be substantially augmented. This study is the first to apply machine learning based methods for the detection of ADHD using solely pupillometrics, and highlights its strength as a potential discriminative biomarker, paving the path for the development of novel diagnostic applications to aid in the detection of ADHD using oculometric paradigms and machine learning.

Detection of chlorpyrifos and carbendazim residues in the cabbage using visible/near-infrared spectroscopy combined with chemometrics

Contamination of agricultural plants and food in the environment caused by pesticide residues has gained great attention of the world. Pesticide residues on vegetables constitute a potential risk to human health. A visible/near-infrared (Vis/NIR) spectroscopy combined with chemometric methods was employed to quantitatively determine chlorpyrifos and carbendazim residues in the cabbage (Brassica chinensis L.). Preprocessing methods were used for spectra denoising. Partial least squares regression (PLSR) and least squares-support vector machine (LS-SVM) were applied as the quantification models. Feature variables were selected by successive projection algorithms (SPA), random frog and regression coefficients in PLSR. As for the samples with chlorpyrifos residues, LS-SVM models based on the global spectra achieved best model performance. The best performance for carbendazim content prediction was achieved by the LS-SVM models based on the original global spectra. And modeling with SPA selected feature variables for carbendazim determination was as good as modeling with the global spectra. The results indicated that Vis/NIR spectroscopy coupled with chemometrics could be an efficient way for the assessment of the pesticide residues in vegetables, and was significant for detection of environmental pollution and ensuring food safety.

Reflectance spectroscopy with operator difference for determination of behenic acid in edible vegetable oils by using convolutional neural network and polynomial correction

A novel polynomial correction method, order-adaptive polynomial correction (OAPC), was proposed to correct reflectance spectra with operator differences, and convolutional neural network (CNN) was used to develop analysis model to predict behenic acid in edible oils. With application of OAPC, CNN performed well with coefficient of determination of correction (R²_cor) of 0.8843 and root mean square error of correction (RMSE_cor) of 0.1182, outperforming partial least squares regression, support vector regression and random forest with OAPC, as well as the cases without OAPC. Based on 16 effective wavelengths selected by combination of bootstrapping soft shrinkage, random frog and Pearson's correlation, CNN and OAPC exhibited excellent performance with R²_cor of 0.9560 and RMSE_cor of 0.0730. Meanwhile, only 5% correction samples were selected by Kennard-Stone for OAPC. Overall, the proposed method could alleviate the impact of operator differences on spectral analysis, thereby providing potential to correct differences from measurement instruments or environments.

Improved prediction of potassium and nitrogen in dried bell pepper leaves with visible and near-infrared spectroscopy utilising wavelength selection techniques

Wet chemistry analysis of agricultural plant materials such as leaves is widely performed to quantify key chemical components to understand plant physiological status. Visible and near-infrared (Vis-NIR) spectroscopy is an interesting tool to replace the wet chemistry analysis, often labour intensive and time-consuming. Hence, this study accesses the potential of Vis-NIR spectroscopy to predict nitrogen (N) and potassium (K) concentration in bell pepper leaves. In the chemometrics perspective, the study aims to identify key Vis-NIR wavelengths that are most correlated to the N and K, and hence, improves the predictive performance for N and K in bell pepper leaves. For wavelengths selection, six different wavelength selection techniques were used. The performances of several wavelength selection techniques were compared to identify the best technique. As a baseline comparison, the partial least-square (PLS) regression analysis was used. The results showed that the Vis-NIR spectroscopy has the potential to predict N and K in pepper leaves with root mean squared error of prediction (RMSEP) of 0.28 and 0.44%, respectively. The wavelength selection in general improved the predictive performance of models for both K and N compared to the PLS regression. With wavelength selection, the RMSEP's were decreased by 19% and 15% for N and K, respectively, compared to the PLS regression. The results from the study can support the development of protocols for non-destructive prediction of key plant chemical components such as K and N without wet chemistry analysis.

Nondestructive evaluation of soluble solids content in tomato with different stage by using Vis/NIR technology and multivariate algorithms

In this study Vis/NIR spectroscopy was applied to evaluate soluble solids content (SSC) of tomato. A total of 168 tomato samples with five different maturity stages, were measured by two developed systems with the wavelength ranges of 500-930 nm and 900-1400 nm, respectively. The raw spectral data were pre-processed by first derivative and standard normal variate (SNV), respectively, and then the effective wavelengths were selected using competitive adaptive reweighted sampling (CARS) and random frog (RF). Partial least squares (PLS) and least square-support vector machines (LS-SVM) were employed to build the prediction models to evaluate SSC in tomatoes. The prediction results revealed that the best performance was obtained using the PLS model with the optimal wavelengths selected by CARS in the range of 900-1400 nm (Rp = 0.820 and RMSEP = 0.207 °Brix). Meanwhile, this best model yielded desirable results with Rp and RMSEP of 0.830 and 0.316 °Brix, respectively, in 60 samples of the independent set. The method proposed from this study can provide an effective and quick way to predict SSC in tomato.

A novel variable selection method based on combined moving window and intelligent optimization algorithm for variable selection in chemical modeling

We propose a new wavelength selection algorithm based on combined moving window (CMW) and variable dimension particle swarm optimization (VDPSO) algorithm. CMW retains the advantages of the moving window algorithm, and different windows can overlap each other to realize automatic optimization of spectral interval width and number. VDPSO algorithms improve the PSO algorithm. They can search the data space in different dimensions, and reduce the risk of limited local extrema and over fitting. Four different high-performance variable selection algorithms-BOSS, VCPA, iVISSA and IRF-are compared in three NIR data sets (corn, beer and fuel). The results show that VDPSO-CMW has better performance. The Matlab codes for implementing PSO-CWM and VDPSO-CMW are freely available on the website: https://www.mathworks.com/matlabcentral/fileexchange/75828-a-variable-selection-method.

Estimating Leaf Nitrogen Content in Corn Based on Information Fusion of Multiple-Sensor Imagery from UAV

With the rapid development of unmanned aerial vehicle (UAV) and sensor technology, UAVs that can simultaneously carry different sensors have been increasingly used to monitor nitrogen status in crops due to their flexibility and adaptability. This study aimed to explore how to use the image information combined from two different sensors mounted on an UAV to evaluate leaf nitrogen content (LNC) in corn. Field experiments with corn were conducted using different nitrogen rates and cultivars at the National Precision Agriculture Research and Demonstration Base in China in 2017. Digital RGB and multispectral images were obtained synchronously by UAV in the V12, R1, and R3 growth stages of corn, respectively. A novel family of modified vegetation indices, named coverage adjusted spectral indices (CASIs (CASI =VI/1+FVcover, where VI denotes the reference vegetation index and FVcover refers to the fraction of vegetation coverage), has been introduced to estimate LNC in corn. Thereby, typical VIs were extracted from multispectral images, which have the advantage of relatively higher spectral resolution, and FVcover was calculated by RGB images that feature higher spatial resolution. Then, the PLS (partial least squares) method was employed to investigate the relationships between LNC and the optimal set of CASIs or VIs selected by the RFA (random frog algorithm) in different corn growth stages. The analysis results indicated that whether removing soil noise or not, CASIs guaranteed a better estimation of LNC than VIs for all of the three growth stages of corn, and the usage of CASIs in the R1 stage yielded the best R2 value of 0.59, with a RMSE (root mean square error) of 22.02% and NRMSE (normalized root mean square error) of 8.37%. It was concluded that CASIs, based on the fusion of information acquired synchronously from both lower resolution multispectral and higher resolution RGB images, have a good potential for crop nitrogen monitoring by UAV. Furthermore, they could also serve as a useful way for assessing other physical and chemical parameters in further applications for crops.

Hyperspectral imaging coupled with multivariate methods for seed vitality estimation and forecast for Quercus variabilis

In this study, the feasibility of estimation and forecast of different vitality Quercus variabilis seeds by a hyperspectral imaging technique were investigated. Artificially accelerated aging was conducive to achieve the division of four vitality levels. Hyperspectral data in the first 10 h of germination were continuously collected at one-hour intervals. The optimal band was selected for the original and pre-processed spectra which were treated by multiple scatter correction (MSC) and the Savitzky-Golay first derivative (SG 1st). Five characteristic wavelength methods were compared: successive projections algorithm (SPA), competitive adaptive reweighted sampling (CARS), genetic algorithm (GA), variable important in projection (VIP), and random frog (RF). Partial least square-discriminant analysis (PLS-DA) and K-nearest neighbor (KNN) built the vitality estimation model based on different data sets, and GA + PLS-DA constructed the optimal model with the highest accuracy. According to the weight coefficient and reflectance of the characteristic band extracted by the GA, the reflectance curves of different levels over time were plotted. The data of 0 h was employed to establish the vitality forecast model. The forecast model had a high recognition rate, with PLS-DA exceeding 99% and KNN exceeding 85%. This indicated that hyperspectral imaging of seed germination processes could achieve non-destructive estimation of Q. variabilis seed vitality, and accurate prediction in a shorter time is feasible.

Growth Stages Classification of Potato Crop Based on Analysis of Spectral Response and Variables Optimization

Potato is the world's fourth-largest food crop, following rice, wheat, and maize. Unlike other crops, it is a typical root crop with a special growth cycle pattern and underground tubers, which makes it harder to track the progress of potatoes and to provide automated crop management. The classification of growth stages has great significance for right time management in the potato field. This paper aims to study how to classify the growth stage of potato crops accurately on the basis of spectroscopy technology. To develop a classification model that monitors the growth stage of potato crops, the field experiments were conducted at the tillering stage (S1), tuber formation stage (S2), tuber bulking stage (S3), and tuber maturation stage (S4), respectively. After spectral data pre-processing, the dynamic changes in chlorophyll content and spectral response during growth were analyzed. A classification model was then established using the support vector machine (SVM) algorithm based on spectral bands and the wavelet coefficients obtained from the continuous wavelet transform (CWT) of reflectance spectra. The spectral variables, which include sensitive spectral bands and feature wavelet coefficients, were optimized using three selection algorithms to improve the classification performance of the model. The selection algorithms include correlation analysis (CA), the successive projection algorithm (SPA), and the random frog (RF) algorithm. The model results were used to compare the performance of various methods. The CWT-SPA-SVM model exhibited excellent performance. The classification accuracies on the training set (Atrain) and the test set (Atest) were respectively 100% and 97.37%, demonstrating the good classification capability of the model. The difference between the Atrain and accuracy of cross-validation (Acv) was 1%, which showed that the model has good stability. Therefore, the CWT-SPA-SVM model can be used to classify the growth stages of potato crops accurately. This study provides an important support method for the classification of growth stages in the potato field.