To solve the problems of high kernel breakage and low threshing efficiency in the threshing operation of high-moisture maize, this study designs an adaptive threshing device based on the coupled working principle of vortex airflow driving and flexible collision. The adaptive threshing mode enables maize ears to move spirally upward under vortex airflow and make compliant contact with flexible components. By adopting repeated mild collisions instead of the rigid violent impact used in traditional devices, low-damage maize threshing is achieved. Preliminary experiments verify that the layout density of flexible threshing units, tangential airflow velocity, and feeding speed are the key factors affecting threshing performance. A regression orthogonal rotational combination test is conducted to systematically explore the single-factor and interactive effects on threshing efficiency, and the optimal parameter configuration is obtained. The test results show that, under the conditions of circumferential angular spacing of 21.5°, tangential velocity of 45.9 m/s and feed rate of 0.65 kg/s, the maize threshing rate reaches 96.1% while the grain breakage rate is controlled below 0.1%, which is significantly superior to conventional rigid threshing methods. This research provides a new technical scheme and experimental data reference for the low-damage threshing study of high-moisture maize.

To investigate the dynamic threshing and separation process of white radish seed pods using the Discrete Element Method (DEM) and improve the accuracy of DEM simulations, a study on parameter calibration for the DEM model of white radish seed pods was carried out through experimental and simulation methods. Firstly, the static friction coefficient and rolling friction coefficient between seed pods and steel plates and between seed pods themselves, as well as restitution coefficient between seed pods and steel plates, were measured using the inclined plane method and the flat plate collision method. The angle of repose test for white radish seed pods was conducted using the funnel accumulation method, and the contour of the angle of repose image was fitted by Origin 2018 software, yielding a measured angle of repose of 31.03°. Key contact parameters between white radish seed pods and operating equipment were determined using the Plackett–Burman design and the steepest ascent method, and the model reliability was verified through the angle of repose test. The results indicated that restitution coefficient, static friction coefficient, and rolling friction coefficient between white radish seed pods and operating equipment were 0.5, 0.488, and 0.025, respectively. The optimal values of restitution coefficient, static friction coefficient, and rolling friction coefficient between white radish seed pods themselves were 0.4, 0.219, and 0.055, respectively. In the angle of repose verification test, the relative error between the simulated and measured values was 2.56%. A DEM-based threshing simulation of white radish seed pods was conducted, achieving a threshing efficiency of 96.47%, with a relative error of 2.24% compared to the actual threshing efficiency. The DEM model of white radish seed pods showed good agreement with the actual situation, and the calibrated DEM simulation parameters were reliable, providing a reference for the DEM simulation studies of white radish seed pods.

Maize kernel damage detection is critical for grain quality assessment and post-harvest processing. However, existing deep learning methods struggle to balance accuracy, model complexity, and multi-class recognition under real-world conditions. To address these issues, this paper proposes YOLOv11n_DEG, an improved lightweight detection model based on YOLOv11n. The model uses the first ten pretrained layers as a feature extractor, replaces standard convolutions in the backbone with depthwise separable convolutions to reduce parameters, integrates an Efficient Channel Attention (ECA) module to enhance feature representation, and employs a dual-dropout strategy in the classification head to mitigate overfitting and improve generalization. Additionally, to account for potential discrepancies in damage characteristics between the obverse and reverse sides of maize kernels, a dual-sided synchronous image feature fusion method is introduced. The output layer classifies five target categories for multi-class damage detection. On an independent test set, the proposed model achieves a precision of 92.5%, a recall of 89.6%, and a mean average precision (mAP) of 93.4%, outperforming the original YOLOv11n by 6.2%, 4.0%, and 3.7%, respectively, while reducing parameter count and computational complexity by 32.9% and 6.5%. To validate practical deployability, a custom testbed with dual-camera synchronous acquisition and geometry-based matching was developed. On this platform, the model with dual-sided fusion achieves a single-side recognition accuracy of 94.1% and a dual-sided recognition accuracy of 89.3%, with an average detection time of 0.8 s per batch. These results demonstrate that YOLOv11n_DEG provides an accurate and practical solution for intelligent maize kernel damage detection, with strong potential for real-world deployment in grain inspection systems.

Weed interference is a major constraint in the production of Angelica dahurica, where seedling–weed similarity, small target size, soil background variation, and occlusion make reliable field recognition difficult. This study develops a lightweight attention-enhanced one-stage detector for real-time crop-versus-weed localization in A. dahurica fields. The detector combines a MobileNetV3 backbone for computationally efficient feature extraction, the parameter-free SimAM mechanism for discriminative feature enhancement, and the Inner-EIoU loss for improved localization of small and partially occluded targets. A field image dataset was collected in Luoning County, Luoyang City, Henan Province, China, comprising a total of 2286 original images. To avoid potential data leakage, the dataset was first divided into training, validation, and test sets at the original-image level. Data augmentation was then performed separately within each subset, resulting in a final dataset of 11,430 images. Under the augmented-test evaluation protocol, the proposed detector achieved 92.5% precision, 94.8% recall, 93.6% F1-score, and 95.2% mAP@0.5, while requiring 0.275 million parameters and 1.1 GFLOPs and reaching 215 FPS in offline inference. For an independent field-validation dataset collected after model training and deployment, precision, recall, F1-score, and mAP@0.5 were 91.6%, 88.4%, 90.0%, and 91.8%, respectively. On an NVIDIA Jetson Orin Nano 8GB edge platform, TensorRT FP16 acceleration produced an average speed of 46.8 FPS. The results show that the proposed detector offers a practical balance among detection accuracy, model compactness, inference speed, and field adaptability. Accordingly, it can serve as a visual-perception component for precision-weeding systems in Angelica dahurica fields.