Seed quality affects crop yield and the quality of agricultural products, and traditional identification methods are time-consuming, complex, and irreversibly destructive. This study aims to establish a fast, non-destructive, and effective approach for defect detection in maize seeds based on hyperspectral imaging (HSI) technology combined with deep learning. Raw spectra collected from maize seeds (200 each healthy and worm-eaten) were pre-processed using detrending (DE) and multiple scattering correction (MSC) to highlight the spectral differences between samples. A convolutional neural network architecture (CNN-FES) based on a feature selection mechanism was proposed according to the importance of wavelength in the target classification task. The results show that the subset of 24 feature wavelengths selected by the proposed CNN-FES can capture important feature information in the spectral data more effectively than the conventional successive projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS) algorithms. In addition, a convolutional neural network architecture (CNN-ATM) based on an attentional classification mechanism was designed for one-dimensional spectral data classification and compared with three commonly used machine learning methods, linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM). The results show that the classification performance of the designed CNN-ATM on the full wavelength does not differ much from the above three methods, and the classification accuracy is above 90% on both the training and test sets. Meanwhile, the accuracy, sensitivity, and specificity of CNN-ATM based on feature wavelength modeling can reach up to 97.50%, 98.28%, and 96.77% at the highest, respectively. The study shows that hyperspectral imaging-based defect detection of maize seed is feasible and effective, and the proposed method has great potential for the processing and analysis of complex hyperspectral data.

In this paper, an adaptive header height control system is designed. Through the influence of the natural frequency, ${\omega }_n$, and damping ratio, $\zeta$, on the system’s dynamic index, the optimal hydraulic cylinder parameters are determined comprehensively. The ground profiling monitoring mechanism and the header height feedback mechanism based on the angle sensor are designed. An integrated electromagnetic proportional valve was installed to replace the original header-controlled, electronically-controlled reversing valve, and a PWM (pulse width modulation) control-simulated counterweight test was performed. The limitation of traditional PID facing the integral saturation state is analysed, and a new EVPIVS-PID algorithm is proposed and simulated. Through the analysis of multiple groups of sample data in the field test, the accuracy of the control system in the header height control and output PWM value is demonstrated. The effectiveness of the EVPIVS-PID control algorithm to change the corresponding PID parameters based on the monitoring operation speed is analysed and demonstrated. Experiments show that the adaptive control system of header height based on ground profiling has a stable control effect. The height error of cutting stubble is not more than 2 cm, which can meet the requirements of a 5–11 km/h harvesting speed in plain areas.

To address the limitations of the current discrete element model of rice plants in terms of accurately reflecting structural differences and threshing characteristics, this study proposes a whole-rice-plant modeling method based on segmented hollow stalks and establishes a whole-rice-plant model that accurately represents the bending and threshing characteristics of the actual rice plant. Initially, based on the characteristics of the rice plant, the rice stalk was segmented into three sections of hollow stalks with distinct structures—namely, the primary stalk, the secondary stalk, and the tertiary stalk—ensuring that the model closely resembles actual rice plants. Secondly, the mechanical and contact parameters for each structure of the rice plant were measured and calibrated through mechanical and contact tests. Subsequently, utilizing the Hertz–Mindlin contact model, a multi-dimensional element particle arrangement method was employed to establish a discrete element model of the entire rice plant. The bending characteristics of the stalk and the threshing characteristics of the rice were calibrated using three-point bending tests and impact threshing tests. The results indicated calibration errors in the bending resistance force of 4.46%, 3.95%, and 2.51% for the three-section stalk model, and the calibration error for the rice model’s threshing rate was 1.86%, which can accurately simulate the bending characteristics of the stalk and the threshing characteristics of the rice plant. Finally, the contact characteristics of the model were validated through a stack angle verification test, which revealed that the relative error of the stacking angle did not exceed 7.52%, confirming the accuracy of the contact characteristics of the rice plant model. The findings of this study provide foundational models and a theoretical basis for the simulation of and analytical applications related to rice threshing and cleaning.

Rice seed production is a critical step in breeding high-quality varieties. To ensure seed purity, it is essential that no residual grains or broken ears remain in the harvester header after harvesting each variety, thus preventing cross-contamination. This study addresses the issue of seed retention in existing rice harvesters, which lack efficient self-cleaning or other cleaning mechanisms and cannot be cleaned rapidly. A self-cleaning device for the harvester header was designed to enable one-click cleaning after harvesting a single variety. A novel cleaning nozzle was developed as the key component of the device, with its structure optimized through single-factor and orthogonal combination experiments. The number of nozzles was determined based on their spray width and the header width. A header-cleaning airflow simulation model based on Fluent–EDEM coupling was constructed to investigate the effects of nozzle inlet pressure, airflow incident angle, and nozzle outlet height on the self-cleaning rate. Optimal cleaning parameters were identified to maximize the self-cleaning rate, and the simulation results were validated. The study revealed that the nozzle’s expansion section length, throat diameter, and contraction section length significantly affect the spray width. When the expansion section length was 10 mm, the throat diameter was 8 mm, and the contraction section length was 8 mm, the nozzle achieved the largest jet angle, measuring 50.3 cm. Further analysis indicated that inlet air pressure had the greatest influence on the self-cleaning rate, followed by airflow incident angle and nozzle outlet height. The optimal parameter combination was identified as an inlet air pressure of 0.6 Mpa, an airflow incident angle of 118.25°, and a nozzle outlet height of 2.64 mm, achieving a maximum self-cleaning rate of 99.63%. A one-click cleaning system was designed using an STM32 microcontroller and hardware circuits. Field experiments under optimal parameters demonstrated a self-cleaning rate of 97.68% with a cleaning duration of 10 s per cycle. The findings provide theoretical guidance for the design and optimization of self-cleaning headers for rice seed production.

Traditional methods of pest control for sweet potatoes cause the waste of pesticides and land pollution, but the target detection algorithm based on deep learning can control the precise spraying of pesticides on sweet potato plants and prevent most pesticides from entering the land. Aiming at the problems of low detection accuracy of sweet potato plants and the complex of target detection models in natural environments, an improved algorithm based on YOLOv8s is proposed, which can accurately identify early sweet potato plants. First, this method uses an efficient network model to enhance the information flow in the channel, obtain more effective global features in the high-level semantic structure, and reduce model parameters and computational complexity. Then, cross-scale feature fusion and the general efficient aggregation architecture are used to further enhance the network feature extraction capability. Finally, the loss function is replaced with InnerFocaler-IoU (IFIoU) to improve the convergence speed and robustness of the model. Experimental results showed that the mAP0.5 and model size of the improved network reached 96.3% and 7.6 MB. Compared with the YOLOv8s baseline network, the number of parameters was reduced by 67.8%, the amount of computation was reduced by 53.1%, and the mAP0.5:0.95 increased by 3.5%. The improved algorithm has higher detection accuracy and a lower parameter and calculation amount. This method realizes the accurate detection of sweet potato plants in the natural environment and provides technical support and guidance for reducing pesticide waste and pesticide pollution.