Non-destructive prediction of lead content in oilseed rape leaves by fluorescence hyperspectral technology based on neural network.

Hyperspectral Prediction of Variety, SPAD Value, and Water Content of Oilseed Rape Leaves Using an Improved WGAN-GP

Rapid and non-destructive identification of oilseed rape varieties and prediction of leaf Soil Plant Analysis Development (SPAD) value and water content are important for variety evaluation and plant-status monitoring. However, hyperspectral prediction is constrained by small sample size, costly reference measurements, and acquisition-batch differences. Samples were collected over four consecutive days. Day 1 samples were used for training and Wasserstein generative adversarial network with gradient penalty (WGAN-GP) generation. For Days 2–4, 20% of the samples from each day were combined to form a selection-validation set, while the remaining 80% were retained separately as Test sets 1–3. The improved WGAN-GP integrated principal component analysis–Gaussian mixture model (PCA–GMM)-based class assignment, a partial least squares regression (PLSR) consistency loss, physicochemical distribution control, and generated-sample selection. Under single-task modeling, the improved framework enhanced all three tasks across the test sets. On Test set 3, variety accuracy increased from 0.4672 to 0.6100, SPAD root mean square error of prediction (RMSEP) decreased from 6.2201 to 4.6303, and the water-content test-set correlation coefficient (rp) increased from 0.6803 to 0.7842. The improved multi-task model also enhanced all three tasks. These findings support the framework within the investigated three-variety, four-day leaf setting.

Multistage Fourier Volumetric Attention Transformer for Hyperspectral Reconstruction and Physicochemical Quality Prediction of Brewing Sorghum

Hyperspectral monitoring of wheat gluten index based on multi-model ensemble and adaptive learning.

Wheat gluten index is an important indicator of food quality inspection. Although hyperspectral technology offers non-destructive and rapid detection potential, its application faces challenges including high-dimensional data redundancy, multicollinearity among adjacent wavelengths, and model overfitting risks in small-sample scenarios. This study proposes a rapid prediction method for wheat gluten index integrating adaptive spectral preprocessing with machine learning. Based on 89 wheat flour samples, adaptive wavelet denoising (average SNR improvement: 11.92 dB), successive projections algorithm (SPA) feature selection, and variance inflation factor (VIF) collinearity diagnosis reduced 2001 spectral features to 5 core variables (587 nm, 436nm_square, 1241 nm, 1080nm_square, 940nm_square). Ten algorithms including PLSR, SVR, XGBoost, and CatBoost were systematically compared under 70% training/30% test split with 10-fold cross-validation. CatBoost achieved optimal performance (R2 = 0.8552, RMSE = 7.8602), surpassing PLSR (R2 = 0.7090) and XGBoost (R2 = 0.7923) by 20.6% and 8.0% respectively. Notably, the well-tuned single CatBoost model outperformed stacking ensemble methods (R2 = 0.7803), providing empirical evidence for model selection in small-sample spectral analysis. SHAP interpretability analysis identified critical spectral bands corresponding to protein and starch absorption characteristics, offering guidance for portable equipment optimization and mechanistic understanding of spectral-quality relationships.

GCT-BCLN: a bidirectional closed-loop network for nondestructive detection of rice seed vigor using hyperspectral imaging

Rice seed vigor is a key determinant of germination performance and final crop yield, making its rapid and non-destructive assessment essential for seed quality evaluation. Conventional vigor detection methods are often destructive, labor-intensive, and time-consuming. Hyperspectral imaging provides a promising non-destructive alternative, but hyperspectral data are typically high-dimensional, redundant, and susceptible to noise and scattering interference. Moreover, existing models still have limited ability to discriminate subtle spectral differences among seed vigor levels. To address these challenges, this study proposes a gated recurrent unit (GRU)-guided closed-loop CNN-Transformer network (GCT-BCLN) for accurate, non-destructive identification of rice seed vigor. The model establishes bidirectional information flow between CNN and Transformer via the GRU, enabling dynamic and synergistic optimization of local spectral features and global spectral representations. In addition, a combined preprocessing strategy integrating adaptive iteratively reweighted penalized least squares (AirPLS), Savitzky–Golay (SG) smoothing, and multiplicative scatter correction (MSC) was adopted to improve spectral quality. Experimental results showed that GCT-BCLN achieved a test accuracy of 0.9795 for hybrid indica rice, outperforming the CNN-Transformer fusion model by 1.37%. The model also achieved accuracies of 0.9793 and 0.9758 on conventional japonica rice and glutinous japonica rice, respectively, showing consistent performance across the three evaluated variety-specific datasets under the controlled experimental protocol. These results support the feasibility of GCT-BCLN for laboratory-scale, non-destructive discrimination of aging-induced rice seed categories under controlled conditions, while practical application requires further external validation.

Optimizing Leaf Nitrogen Estimation Using Hyperspectral Sensing and Machine Learning With Band Selection Techniques

Integrating hyperspectral sensing with machine learning (ML) enables efficient monitoring of crop health, improved fertilization management, and data-driven precision agriculture. This study proposes a framework for leaf-level nitrogen estimation using full-range visible, near-infrared, and short-wave infrared hyperspectral reflectance data combined with advanced ML models and band selection (BS) strategies. Seven regression models, including linear, kernel-based, and neural-network approaches, were benchmarked using both full-spectrum data and reduced bands subsets. BS was addressed through variable importance in projection (VIP), a constrained genetic algorithm (GA) specifically designed to enforce fixed subset cardinality, and by introducing a two-step hybrid VIP–GA (HVG) strategy, in which VIP was first applied as a coarse filtering step followed by GA-based refinement, thus combining computational efficiency with refined combinatorial optimization. Experiments were conducted on a large public leaf-level spectral dataset using both repeated random subsampling and leave-one-group-out (LOGO) validation to assess robustness to training set size and transferability under heterogeneous conditions. Results showed that regularized linear models achieved, on these data, performance comparable to or better than more complex nonlinear approaches, with ridge regression yielding the lowest $\text{NRMSE}{{\mathrm{P}}_{\text{IQR}}}$ values, ranging between 0.27 and 0.30 across different BS scenarios. LOGO experiments indicated moderate transferability across heterogeneous conditions, although performance remained dependent on group-specific characteristics. Overall, BS considerably reduced spectral dimensionality and computational cost while preserving predictive performance, with the HVG approach providing an effective tradeoff between accuracy, efficiency, and interpretability. These findings highlight the potential of combining ML and BS for efficient leaf-level nutrient monitoring and for providing a priori spectral insights for future hyperspectral sensing studies.

A Novel Hyperspectral Imaging based Framework for Tomato Leaf Disease Detection

A lightweight deep learning model, the Dual Multi-Layer Perceptron Feature Fusion Network (DMLPFFN) as a hyperspectral-based approach to the four tomato disease stages, which includes healthy, asymptomatic, early and late infection is demonstrated.