Search Results (87)

Moisture content (MC) is an important indicator affecting the quality of dried lemon slices. To achieve rapid and non-destructive MC detection, this study developed a lemon slice MC detection model based on the fusion of image texture and spectral features. A total of 240 lemon slices were dried at 80 ${}^{\circ }$C, and hyperspectral imaging (HSI) data and reference MC values were collected at different drying times. Competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), and uninformative variable elimination (UVE) were used to select characteristic wavelengths. Image texture features were extracted using the gray-level co-occurrence matrix (GLCM), and the spectral features and image texture features were concatenated and fused. Kernel principal component analysis (KPCA) was then applied to reduce the dimensionality of the fused feature set. Finally, support vector machine (SVM), general regression neural network (GRNN), and partial least squares (PLS) models were established for MC detection. The results showed that the spectral-feature-based models achieved good predictive performance. The image texture-feature-based models also demonstrated predictive capability, whereas spectral–texture feature fusion further improved prediction accuracy. Among all models, the PLS model based on the spectral–texture fused features achieved the best performance, with a coefficient of determination of prediction ($R_p^2$) of 0.9890 and a root mean square error of prediction (RMSEP) of 0.1916 g/g in the prediction set. These results indicate that HSI combined with spectral–texture feature fusion provides a promising approach for rapid MC detection in lemon slices. Full article

Soybean protein content is a key indicator of nutritional value and quality grade, and its determination is important for quality evaluation and cultivar selection. To overcome the time-consuming and costly limitations of conventional chemical assays, this study proposed a multiple linear learner ensemble importance-score wavelength selection (MLLEISWS) method to identify informative wavelengths from soybean near-infrared spectra and establish a partial least squares (PLS) model. MLLEISWS was compared with competitive adaptive reweighted sampling, successive projections algorithm, and uninformative variable elimination. Shapley additive exPlanations (SHAP) were applied to the MLLEISWS algorithm to interpret the selected wavelengths. Results showed that the PLS model developed using MLLEISWS achieved the best performance. With only 29 selected wavelengths, the coefficients of determination for the training and test sets reached 0.941 and 0.933, respectively. Root mean square errors were 0.490% and 0.514%, relative root mean square errors were 1.32% and 1.37%, and residual predictive deviation was 3.863, indicating predictive accuracy and stability. SHAP analysis showed that the selected wavelengths were located in protein-related spectral regions and corresponded to overtone and combination bands information from functional groups. MLLEISWS effectively reduced variable dimensionality while maintaining model performance. Full article

Early-stage fungal contamination in maize kernels is difficult to identify visually and it can cause severe quality and safety risks during storage and transportation. Short-wave infrared (SWIR) hyperspectral imaging offers a rapid, non-destructive approach by capturing chemical information related to water, proteins, and lipids. This study investigates the early detection and classification of Gibberella zeae contamination in maize kernels using SWIR hyperspectral imaging combined with machine learning. Two maize varieties were artificially inoculated and cultured under controlled conditions, followed by hyperspectral data collection over six contamination stages. Various preprocessing techniques including standard normal variate (SNV), second derivative (SD), multiplicative scatter correction (MSC), and derivatives were evaluated to enhance data quality. Feature wavelength selection was performed using successive projections algorithm (SPA), competitive adaptive reweighted sampling (CARS), and uninformative variable elimination (UVE), significantly reducing redundancy and improving classification performance. Multiple models, including linear discriminant analysis (LDA), multilayer perceptron (MLP), support vector machine (SVM), a convolutional neural network (CNN), long short-term memory (LSTM) network, and a hybrid architecture Transformer that integrated a CNN, a LSTM network, and a Transformer (abbreviated as CLT), were constructed for both binary (healthy vs. contaminated) and multiclass classification tasks. Specifically, the multiclass task consisted of six contamination stages corresponding to contamination time from Day 0 to Day 5. The best binary classification task accuracy of 100% was achieved using SNV-preprocessed data with the MLP model. For multiclass classification task, the SD-preprocessed LDA model reached a test accuracy of 92.56%. Combined with appropriate preprocessing, feature selection and modeling, these results demonstrate that hyperspectral imaging is a powerful tool for the non-destructive, early-stage identification of fungal contamination in maize kernels, offering strong support for food safety and quality monitoring. Full article

Near-infrared (NIR) spectroscopy combined with multivariate calibration methods is widely used for the quantitative analysis of complex samples. However, the high-dimensional redundancy of spectra may compromise model predictive accuracy, making it necessary to select variables before modeling. The beluga whale optimization (BWO) algorithm is known for its fast convergence speed, high accuracy and few parameters. The present study employed the discretized BWO (DBWO) algorithm in conjunction with partial least squares (PLS) for spectral quantitative analysis of complex samples. After the optimal number of iterations and transfer function were determined, the PLS models were established based on the randomization test (RT), uninformative variable elimination (UVE) and Monte Carlo uninformative variable elimination (MC-UVE). The predictive performance of DBWO-PLS was compared with full-spectrum PLS, RT-PLS, UVE-PLS and MC-UVE-PLS using wheat, tablet and cocoa bean samples. The results show that all four variable selection methods enhanced model prediction accuracy, with the DBWO-PLS model notably achieving superior performance. Full article

Rolling is a critical step in the processing of black tea, marking the beginning of fermentation. At this stage, the formation of tea pigments causes significant changes in the color of the processed leaves, laying the essential groundwork for the development of color and flavor quality components in subsequent fermentation processes. However, the rapid and non-destructive sensing of tea pigments during black tea rolling remains challenging. This study focused on black tea products undergoing rolling as its research subject, utilizing electrical characteristic detection technology to collect time-series electrical parameters of rolling leaves at various testing frequencies. The original electrical parameters were preprocessed using multiplicative scatter correction (MSC), min-max normalization (Min-Max), and smoothing (Smooth). Various selection methods, including the competitive adaptive reweighting algorithm (CARS), uninformative variable elimination (UVE), and the variable combination population analysis and iterative retained information variable algorithm (VCPA-IRIV), were employed to identify electrical parameters relevant to the targeted attributes. Quantitative prediction models for the content of tea pigments were established using partial least squares regression (PLSR) and support vector machine regression (SVR). The results demonstrated that the Smooth-VCPA-IRIV-SVR model exhibited superior performance in predicting the contents of theaflavins (TFs), thearubigins (TRs), and theabrownins (TBs). Correlation coefficients of prediction (R_p) all exceeded 0.99, and Relative prediction deviation (RPD) values were all above 6.5, indicating that the model enables rapid and non-destructive detection of tea pigment content during black tea rolling. These findings provide preliminary technical support and reference for the digital production of black tea. Full article

This study developed a rapid, non-destructive method for the quantitative detection of protein in cottonseed by integrating near-infrared (NIR) fiber spectroscopy with chemometric machine learning. The establishment of this method holds significant importance for the rational and efficient utilization of cottonseed resources, advancing research on the genetic improvement of cottonseed nutritional quality, and promoting the development of equipment for raw cottonseed protein detection. Fuzzy cottonseed samples from three varieties were collected, and their NIR fiber-optic spectra were acquired. Reference protein contents were measured using the Kjeldahl method. Spectra were denoised through preprocessing, after which informative wavelengths were selected by combining Uninformative Variable Elimination (UVE) with Competitive Adaptive Reweighted Sampling (CARS) and the Random Frog (RF) algorithm. Partial least squares regression (PLSR), least-squares support vector machine (LSSVM), and support vector regression (SVR) models were then constructed to predict protein content. Model performance was assessed using the coefficient of determination (R²), root-mean-square error (RMSE), residual predictive deviation (RPD), and range error ratio (RER). The results indicate that the standard normal variate (SNV) is the most effective preprocessing step. The best performance was achieved by the LSSVM model coupled with UVE + CARS, yielding R² = 0.8571, RMSE = 0.0033, RPD = 2.7078, and RER = 10.72, outperforming the PLSR and SVR counterparts. These findings provide technical support for the rapid detection of fuzzy cottonseed protein and lay the groundwork for the development of related detection equipment. Full article

Fusarium head blight (FHB), a globally significant agricultural disaster, causes annual losses of dozens of millions of tons of wheat toxins produced by FHB, such as deoxyroscyliaceol, further pose serious threats to human and livestock health. Consequently, rapid and non-destructive determination of FHB severity is crucial for implementing timely and precise scientific control measures, thereby ensuring wheat supply security. Therefore, this study adopts hyperspectral imaging (HSI) combined with a lightweight deep learning model. Firstly, the wheat ears were inoculated with Fusarium fungi at the spike’s midpoint, and HSI data were acquired, yielding 1660 samples representing varying disease severities. Through the integration of multiplicative scatter correction (MSC) and uninformative variable elimination (UVE) methods, features are extracted from spectral data in a manner that optimizes the reduction of feature dimensionality while preserving elevated classification accuracy. Finally, a lightweight FHB severity discrimination model based on MobileNetV2 was developed and deployed as an easy-to-use analysis system. Analysis revealed that UVE-selected characteristic bands for FHB severity predominantly fell within 590–680 nm (chlorophyll degradation related), 930–1043 nm (water stress related) and 738 nm (cell wall polysaccharide decomposition related). This distribution aligns with the synergistic effect of rapid chlorophyll degradation and structural damage accompanying disease progression. The resulting MobileNetV2 model achieved a mean average precision (mAP) of 99.93% on the training set and 98.26% on the independent test set. Crucially, it maintains an 8.50 MB parameter size, it processes data 2.36 times faster, significantly enhancing its suitability for field-deployed equipment by optimally balancing accuracy and operational efficiency. This advancement empowers agricultural workers to implement timely control measures, dramatically improving precision alongside optimized field deployment. Full article

Amylose content significantly influences the technological, quality, and nutritional properties of starchy foods. This study developed a rapid, non-destructive method to quantify amylose content in starch using mid-infrared (MIR) spectroscopy combined with chemometric techniques. Manually prepared starch mixtures with varying amylose levels were scanned to obtain MIR spectra, which were preprocessed using smoothing and z-score normalization to reduce operational variability. Three variable selection methods, including bootstrap soft shrinkage (BOSS), competitive adaptive reweighted sampling (CARS), and uninformative variable elimination (UVE), were applied to select the useful spectra. A partial least square (PLS) model was then constructed to correlate selected spectral data with amylose content. The results revealed that the number and position of selected variables differed across different optimization methods, which influenced the model’s performance. It is worth noting that the optimized PLS model significantly reduced the root mean squared error of cross-validation (RMSECV) and improved prediction accuracy in 50 runs. In particular, the CARS-PLS model showed superior performance, achieving a correlation coefficient (R_p) of 0.964 and a root mean squared error of prediction (RMSEP) of 4.59, a 60% improvement over the original PLS model, which had an RMSEP of 11.56. These results highlight MIR spectroscopy’s potential, combined with optimized chemometric models, for accurate amylose quantification in food quality control. Full article

Early and accurate detection of mechanical damage in prunes is crucial for preserving postharvest quality and enabling automated sorting. This study proposes a practical and reproducible method for identifying cumulative bruising in prunes using visible–near-infrared (Vis–NIR) reflectance spectroscopy coupled with machine learning techniques. A self-developed impact simulation device was designed to induce progressive damage under controlled energy levels, simulating realistic postharvest handling conditions. Spectral data were collected from the equatorial region of each fruit and processed using a hybrid modeling framework comprising continuous wavelet transform (CWT) for spectral enhancement, uninformative variable elimination (UVE) for optimal wavelength selection, and support vector machine (SVM) for classification. The proposed CWT-UVE-SVM model achieved an overall classification accuracy of 93.22%, successfully distinguishing intact, mildly bruised, and cumulatively damaged samples. Notably, the results revealed nonlinear reflectance variations in the near-infrared region associated with repeated low-energy impacts, highlighting the capacity of spectral response patterns to capture progressive physiological changes. This research not only advances nondestructive detection methods for prune grading but also provides a scalable modeling strategy for cumulative mechanical damage assessment in soft horticultural products. Full article

Aflatoxin B₁ (AFB₁) contamination in corn silage poses significant risks to livestock and human health. This study developed a non-destructive detection method for AFB₁ using color-sensitive arrays (CSAs). Twenty self-developed CSAs were employed to react with samples, with reflectance spectra collected using a portable spectrometer. Spectral data were optimized through seven preprocessing methods, including Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), first-order derivative (1st D), second-order derivative (2nd D), wavelet denoising, and their combinations. Key variables were selected using five feature selection algorithms: Competitive Adaptive Reweighted Sampling (CARS), Principal Component Analysis (PCA), Random Forest (RF), Uninformative Variable Elimination (UVE), and eXtreme Gradient Boosting (XGBoost). Five machine learning models were constructed: Light Gradient Boosting Machine (LightGBM), XGBoost, Support Vector Regression (SVR), RF, and K-Nearest Neighbor (KNN). The results demonstrated significant AFB₁-responsive characteristics in three dyes: (2,3,7,8,12,13,17,18-octaethylporphynato)chloromanganese(III) (Mn(OEP)Cl), Bromocresol Green, and Cresol Red. The combined 1st D-PCA-KNN model showed optimal prediction performance, with determination coefficient ($R_p^2$ = 0.87), root mean square error (RMSEP = 0.057), and relative prediction deviation (RPD = 2.773). This method provides an efficient solution for silage AFB₁ monitoring. Full article

This study addresses the urgent need for the rapid, non-destructive assessment of key soybean components, including moisture, fat, and protein, using near-infrared (NIR) spectroscopy. This study provides technical and theoretical support for achieving the efficient and accurate detection of major soybean components and for the development of portable near-infrared (NIR) instruments. Thirty soybean samples from diverse sources were collected, and 360 spectral measurements were acquired using a 900–1700 nm NIR spectrometer after grinding and standardized sampling. To improve model robustness, preprocessing strategies such as standard normal variate (SNV), multiplicative scatter correction (MSC), and Savitzky–Golay derivatives were applied. Feature selection was conducted using competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), and uninformative variable elimination (UVE), followed by model construction with partial least squares regression (PLSR), support vector regression (SVR), and random forest (RF). Comparative analysis revealed that the RF model consistently outperformed the others across most combinations. Specifically, the SPASNV + D₁–RF combination achieved an RPD of 14.7 for moisture, CARS–SNV + D₁–RF reached 5.9 for protein, and CARS–SG + D₂–RF attained 12.0 for fat, all significantly surpassing alternative methods and demonstrating a strong nonlinear learning capacity and predictive precision. These findings show that integrating optimal preprocessing and feature selection strategies can markedly enhance the predictive accuracy in NIR-based soybean analyses. The RF model offers exceptional stability and performance, providing both technical reference and theoretical support for the development of portable NIR devices and practical rapid-quality assessment systems for soybeans in industrial applications. Full article

Xishuangbanna cucumber (Cucumis sativus L. var. xishuangbannanesis), as a rare and endangered cucumber germplasm resource, possesses certain irreplaceable characteristics that make it difficult to reacquire once lost. To ensure long-term preservation of this germplasm, immediate propagation and regeneration are required after successful collection. Current germplasm management relying on conventional viability testing methods often leads to seed loss. Therefore, there is an urgent need to develop a rapid and non-destructive testing technology for assessing the seed viability of Xishuangbanna cucumber. This study integrated hyperspectral imaging technology with various data preprocessing methods, feature wavelength selection algorithms, and classification models to achieve rapid and non-destructive detection of Xishuangbanna cucumber seed viability. Hyperspectral imaging was employed to acquire spectral data from the seeds. Preprocessing methods including MSC (Multivariate Scattering Correction), SNV (Standard Normal Variety), FD (First Derivative), SD (Second Derivative), and L2NN (L2 Norm Normalization) were applied to enhance spectral data quality. Feature selection algorithms such as UVE (Uninformative Variables Elimination), SPA (Successive Projections Algorithm), and CARS (Competitive Adaptive Reweighted Sampling) were utilized to identify optimal spectral bands. Combined with KNN (K-Nearest Neighbor) and LogitBoost algorithms, predictive models for seed viability were established. The results demonstrated that the L2NN-KNN model outperformed other models, achieving an accuracy of 83.33%, precision of 86.99%, and an F1-score of 0.83. This study confirms that hyperspectral imaging combined with machine learning can effectively predict the viability of Xishuangbanna cucumber seeds, providing a novel technical approach for the conservation of rare and endangered cucumber germplasm resources. The findings hold significant implications for promoting long-term preservation and sustainable utilization of this valuable genetic material. Full article

This study aims to comprehensively evaluate the internal quality changes in apples during storage via near-infrared spectroscopy. Specifically, we focus on the performance differences in different apple varieties under diverse storage conditions and construct predictive models to determine the optimal storage period. By using near-infrared spectroscopy technology, 384 samples of four apple varieties (Xinjiang Akesu, Wafangdian Huangyuanshuai, Shandong Fuji, and Luochuan Fuji) were analyzed to monitor the changes in their soluble solid content (SSC) and fruit firmness within 7 weeks. The results indicated that, under cold storage conditions, SSC and firmness gradually decreased after peaking between the third and fifth weeks, while the opposite trend was observed at room temperature. To enhance the predictive accuracy of the model, several pretreatment methods were employed, including standardization, multiplicative scatter correction (MSC), and standard normal variate transformation (SNV). Additionally, competitive adaptive reweighted sampling (CARS) and uninformative variable elimination (UVE) were utilized for band selection. These pretreatment and selection processes significantly reduced noise and improved model reliability. The best results were achieved with the Normalization-CARS-PLS model for the sugar content at 1 °C, which demonstrated an optimal predictive correlation coefficient (Rp) of 0.904 and a root mean square error of prediction (RMSEP) of 0.67. For firmness at room temperature, the Normalization-CARS-PLS model also showed an excellent performance, with an Rp of 0.823 and an RMSEP of 0.809. The study of the quality of four varieties of apples under three storage conditions in this paper was able to analyze the changes in the internal quality of apples and predict the optimal storage period of different varieties of apples, which is important for guiding the optimal storage period of apples before ripening. Full article

The soluble solids content (SSC) of blueberry is a key index for evaluating its quality. In view of the demand for rapid non-destructive testing of blueberry SSC and the shortcomings of the existing single-variety testing models in cross-variety applications, a universal prediction model construction method based on hyperspectral imaging (HSI) technology is proposed in this study. The spectral data of three blueberry varieties were obtained by using a 935∼1720 nm hyperspectral imaging system. A partial least squares regression (PLSR) model was constructed by combining different preprocessing methods such as Savitzky–Golay (S-G), multiplicative scatter correction (MSC) and standard normal variable transformation (SNV). The results showed that the PLSR model pretreated by S-G-MSC-SNV had the best performance, and the determination coefficient, root mean square error and residual prediction deviation of the prediction set were 0.94, $0.33\%$ and 3.94, respectively. The characteristic wavelengths were optimized in stages by uninformative variables elimination (UVE) and the successive projections algorithm (SPA), and the model was simplified by multiple linear regression (MLR). Finally, a high-precision UVE-PLSR model and a simple and efficient UVE-SPA-MLR hybrid model were obtained. The construction of this universal model effectively solves the limitation of the single-variety model and has important application value in the optimization of food industry production and quality control. Full article

Total soluble solids (TSSs) serve as a crucial maturity indicator and quality determinant in apricots, influencing harvest timing and postharvest management decisions. This study develops an advanced framework integrating adaptive boosting (Adaboost) ensemble learning with high-frequency spectral variables selected by uninformative variable elimination (UVE) for the rapid non-destructive detection of fruit quality. Near-infrared (NIR) spectra (1000~2500 nm) were acquired and then preprocessed through robust principal component analysis (ROBPCA) for outlier detection combined with z-score normalization for spectral pretreatment. Subsequent data processes included three steps: (1) 100 continuous runs of UVE identified characteristic wavelengths, which were classified into three levels—high-frequency (≥90 times), medium-frequency (30–90 times), and low-frequency (≤30 times) subsets; (2) the development of the base optimal partial least squares regression (PLSR) models for each wavelength subset; and (3) the execution of adaptive weight optimization through the Adaboost ensemble algorithm. The experimental findings revealed the following: (1) The model established based on high-frequency wavelengths outperformed both full-spectrum model and full-characteristic wavelength model. (2) The optimized UVE-PLS-Adaboost model achieved the peak performance (R = 0.889, RMSEP = 1.267, MAE = 0.994). This research shows that the UVE-Adaboost fusion method enhances model prediction accuracy and generalization ability through multi-dimensional feature optimization and model weight allocation. The proposed framework enables the rapid, non-destructive detection of apricot TSSs and provides a reference for the quality evaluation of other fruits in agricultural applications. Full article