In order to address the challenge of early detection of cotton verticillium wilt disease, naturally infected cotton plants in the field, which were divided into five categories based on the degree of disease severity, have been investigated in this study. Canopies of infected cotton plants were analyzed with spectral data measured, and various preprocessing techniques, including multiplicative scatter correction (MSC) and MSC-continuous wavelet analysis algorithms, were used to predict the disease severity. With a combination of support vector machine (SVM) models with such optimization algorithms as genetic algorithm (GA), grid search (GS), particle swarm optimization (PSO), and grey wolf optimizer (GWO), a grading model of cotton verticillium wilt disease was established in this study. The study results show that the MSC-PSO-SVM model outperforms the other three models in terms of classification accuracy, and the accuracy, macro precision, macro recall, and macro F1-score of this model are 80%, 81.26%, 80%, and 79.57%, respectively. Among those eight models constructed on the basis of continuous wavelet analyses using mexh and db3, the MSC-db3(2³)-PSO-SVM and MSC-db3(2³)-GWO-SVM models perform best, with the latter having a shorter running time. An overall evaluation shows that the MSC-db3(2³)-GWO-SVM model is an optimal model, with values of its accuracy, macro precision, macro recall, and macro F1-score indicators being 91.2%, 92.02%, 91.2%, and 91.16%, respectively. Moreover, under this model, the prediction accuracy on disease levels 1 and 5 has achieved the highest rate of 100%, with a prediction accuracy rate of 88% on disease level 2 and the lowest prediction accuracy rate of 84% on both disease levels 3 and 4. These results demonstrate that it is effective to use spectral technology in classifying the cotton verticillium wilt disease and satisfying the needs of field detection and grading. This study provides a new approach for the detection and grading of cotton verticillium wilt disease and offered a theoretical basis for early prevention, precise drug application, and instrument development for the disease.

Rapid and non-destructive estimation of the chlorophyll content in cotton leaves is of great significance for the real-time monitoring of cotton growth under verticillium wilt (VW) stress. The spectral reflectance of healthy and VW cotton leaves was determined using hyperspectral technology, and the original spectra were processed using Savitzky–Golay (SG) smoothing, and on its basis through mean centering, standard normal variate (SG-SNV), multiplicative scatter correction (SG-MSC), reciprocal second-order differentiation, and logarithmic second-order differentiation ([lg(SG)]″) preprocessing operations. The characteristic bands were selected based on the correlation coefficient, vegetation index, successive projection algorithm (SPA), and competitive adaptive reweighted sampling (CARS). The single-factor model, back propagation neural network of particle swarm optimization algorithm, and extreme learning machine (ELM) of a grey wolf optimizer (GWO) algorithm were constructed to compare and explore the ability of each model to estimate the soil plant analysis development (SPAD) value of cotton under VW stress. The results showed that spectral pretreatment could improve the correlation between characteristic bands and SPAD values. SG-MSC and SG-SNV showed better changes in the five pretreatments, and the maximum correlation coefficients of healthy and VW cotton leaves were higher than 0.74. Compared with SPA, the accuracy of model estimation based on CARS-extracted characteristic bands was higher, and the estimation accuracy of the multi-factor model was better than that of the single-factor model under each pretreatment. For healthy cotton leaves, [lg(SG)]″–CARS–GWO–ELM was the optimal model, with a modeling and validation set R² of 0.956 and 0.887, respectively. For VW cotton leaves, SG-MSC–CARS–GWO–ELM was the optimal model, with a modeling and validation set R² of 0.832 and 0.824, respectively. Therefore, the GWO–ELM model constructed under different pretreatments combined with characteristic extraction methods can be used for the estimation of leaf SPAD values under VW stress to dynamically monitor VW stress in cotton and provide a theoretical reference for precision agriculture.

Securing authentic cottonseed identity information is crucial for preserving the livelihoods of farmers. Traditional seed identification methods are generally time-consuming, and have a high degree of difficulty. Raman spectroscopy, in combination with machine learning (ML), has opened up new avenues for seed identification. In this study, we explored the feasibility of using Raman spectroscopy combined with ML for cottonseed identification. Using Raman confocal microscopy, we constructed fingerprints of cottonseeds and analyzed their important Raman peaks. We integrated two feature exploration methods (Principal Component Analysis and Harris Hawk optimization) and three ML algorithms (Support Vector Machine, eXtreme Gradient Boosting, and Multi-Layer Perceptron) into a Raman spectroscopy analysis framework to accurately identify cottonseed cultivars. Through the utilization of SHapley Additive exPlanations (SHAP), we provide an in-depth explanation of the model’s decision-making process. Our results demonstrate that XGBoost, a tree-based model, exhibits outstanding accuracy (overall accuracy of 0.94–0.88) in cottonseed identification. Notably, lignin emerged as a pivotal factor that strongly influenced the model’s prediction of cottonseed cultivars, as revealed by the XGBoost interpretation. Overall, our study illustrates the effectiveness of combining Raman spectroscopy with ML to precisely identify cottonseed cultivars. The SHAP framework used in our study enables seed-related personnel to better comprehend the model’s prediction mechanism. These valuable insights are expected to enhance seed planting and management practices in the future.

Accurate detection and counting of wheat spikes are crucial for yield estimation and variety selection in precision agriculture. However, challenges such as complex field environments, morphological variations, and small target sizes hinder the performance of existing models in real-world applications. This study proposes FEWheat-YOLO, a lightweight and efficient detection framework optimized for deployment on agricultural edge devices. The architecture integrates four key modules: (1) FEMANet, a mixed aggregation feature enhancement network with Efficient Multi-scale Attention (EMA) for improved small-target representation; (2) BiAFA-FPN, a bidirectional asymmetric feature pyramid network for efficient multi-scale feature fusion; (3) ADown, an adaptive downsampling module that preserves structural details during resolution reduction; and (4) GSCDHead, a grouped shared convolution detection head for reduced parameters and computational cost. Evaluated on a hybrid dataset combining GWHD2021 and a self-collected field dataset, FEWheat-YOLO achieved a COCO-style AP of 51.11%, AP@50 of 89.8%, and AP scores of 18.1%, 50.5%, and 61.2% for small, medium, and large targets, respectively, with an average recall (AR) of 58.1%. In wheat spike counting tasks, the model achieved an R² of 0.941, MAE of 3.46, and RMSE of 6.25, demonstrating high counting accuracy and robustness. The proposed model requires only 0.67 M parameters, 5.3 GFLOPs, and 1.6 MB of storage, while achieving an inference speed of 54 FPS. Compared to YOLOv11n, FEWheat-YOLO improved AP@50, AP_s, AP_m, AP_l, and AR by 0.53%, 0.7%, 0.7%, 0.4%, and 0.3%, respectively, while reducing parameters by 74%, computation by 15.9%, and model size by 69.2%. These results indicate that FEWheat-YOLO provides an effective balance between detection accuracy, counting performance, and model efficiency, offering strong potential for real-time agricultural applications on resource-limited platforms.

The early detection of Verticillium wilt (VW) in cotton is a critical challenge in agricultural disease management. Cotton, a vital global textile resource, is severely threatened by this devastating disease. Traditional diagnostic methods, which often rely on manual expertise or destructive sampling, are limited by low efficiency and high subjectivity. In recent years, Raman spectroscopy has emerged as a promising solution due to its rapid, non-destructive, and highly sensitive characteristics for plant disease detection. In this study, we analyzed cotton stems using Raman spectroscopy, applying Savitzky-Golay (SG) smoothing combined with multiple preprocessing methods including Scaling and Shifting (SS), Standard Normal Variate (SNV), inverse first-order differential (1/SG)′, and multiplicative scatter correction (MSC). For baseline correction, we employed polynomial fitting (PolyFit) and adaptive iterative weighted penalized least squares (airPLS). Feature selection was performed using principal component analysis (PCA), successive projection algorithm (SPA), and competitive adaptive reweighted sampling (CARS).Three optimized models were developed: support vector machine (SVM) with weighted mean of vectors (INFO) algorithm, random forest (RF) enhanced by particle swarm optimization (PSO), and long short-term memory (LSTM) network optimized via chameleon swarm algorithm (CSA).The results show that the INFO-SVM model with SG-airPLS-(1/SG)′ -CARS preprocessing demonstrated superior performance, achieving 97.5% accuracy (0.974 F1-score) on training data and 90.0% accuracy (0.867 F1-score) on validation data, outperforming both PSO-RF and CSA-LSTM models. These results confirm that Raman spectroscopy integrated with optimized machine learning enables accurate VW classification in cotton stems. This method enables early disease detection during infection, facilitating timely fungicide application and reducing yield losses.

Cotton is a globally strategic crop that plays a crucial role in sustaining national economies and livelihoods. To address the challenges of accurate cotton field extraction in the complex planting environments of Xinjiang’s Alaer reclamation area, a cotton field identification model was developed that integrates multi-source satellite remote sensing data with machine learning methods. Using imagery from Sentinel-2, GF-1, and Landsat 8, we performed feature fusion using principal component, Gram–Schmidt (GS), and neural network techniques. Analyses of spectral, vegetation, and texture features revealed that the GS-fused blue bands of Sentinel-2 and Landsat 8 exhibited optimal performance, with a mean value of 16,725, a standard deviation of 2290, and an information entropy of 8.55. These metrics improved by 10,529, 168, and 0.28, respectively, compared with the original Landsat 8 data. In comparative classification experiments, the endmember-based random forest classifier (RFC) achieved the best traditional classification performance, with a kappa value of 0.963 and an overall accuracy (OA) of 97.22% based on 250 samples, resulting in a cotton-field extraction error of 38.58 km². By enhancing the deep learning model, we proposed a U-Net architecture that incorporated a Convolutional Block Attention Module and Atrous Spatial Pyramid Pooling. Using the GS-fused blue band data, the model achieved significantly improved accuracy, with a kappa coefficient of 0.988 and an OA of 98.56%. This advancement reduced the area estimation error to 25.42 km², representing a 34.1% decrease compared with that of the RFC. Based on the optimal model, we constructed a digital map of continuous cotton cropping from 2021 to 2023, which revealed a consistent decline in cotton acreage within the reclaimed areas. This finding underscores the effectiveness of crop rotation policies in mitigating the adverse effects of large-scale monoculture practices. This study confirms that the synergistic integration of multi-source satellite feature fusion and deep learning significantly improves crop identification accuracy, providing reliable technical support for agricultural policy formulation and sustainable farmland management.