During the grain drying process, porosity directly affects the pore curvature, effective evaporation area, and ventilation resistance of the granular bed, which is an important factor in determining the drying rate, energy consumption and drying uniformity. Existing research on the porosity of flowing maize beds is relatively limited and has primarily focused on static pile conditions. Therefore, this study designed an air convection measurement system consisting of a variable-frequency fan, a valve opening adjustment device, and a differential pressure sensor. The following factors were selected: moisture content (13.03–33.37% w.b.), air velocity (0.6–1.0 m/s), maize flow velocity (0.04–0.06 m/s), and bed thickness (20–40 cm). Experiments were conducted using a Box–Behnken design, and the measured results were compared with theoretical values from the Ergun equation and results from EDEM–Fluent coupled simulations. The dynamic porosity ranged from 57.81% to 63.30%. Porosity decreased as moisture content and pile thickness increased, while it increased as the corn flow velocity increased; air speed had a minor effect. The response surface model had an R² value of 0.9600, and the relative deviations between the measured, simulated, and predicted values did not exceed ±1%. The conclusion is that water content, corn flow velocity, and pile thickness are the primary factors affecting dynamic porosity.

In response to the subjective issues, inconsistent quality standards, high labor intensity and low sorting efficiency during the drying process of green pepper, an improved YOLOv11 algorithm was proposed for quality detection. A multi-scale edge enhancement module (MEEM) is introduced into the backbone network, replacing the original basic C3K2 module with C3K2-MEEM to enhance the extraction of detailed features in images of dried green Sichuan pepper and prevent missed detections, false detections, and boundary confusion. The LRSA module is integrated into the 10th layer of the backbone network to improve the clarity of the tumor-like texture of the Sichuan pepper and reduce the influence of impurities, automatically allocating attention based on feature similarity to preserve local information. In the neck layer, the DPCF module is added to the FPN+PAN feature fusion stage to achieve multi-scale feature collaboration, meeting the detection requirements of dried green Sichuan pepper. The results show that the accuracy recall rate, mean average precision, and model size of the improved MLD-YOLOv11 algorithm are 92.1%, 96.6%, 95.6%, and 11.06 MB, respectively. Compared with the training results of the original YOLOv11 model, the average accuracy of the improved model has increased by 2.2 percentage points, and GFLOPs have definitely decreased by 2 G, with parameter reduction of approximately 3.10%. Compared with other mainstream models, the MLD-YOLOv11 model has significant advantages in terms of mean average precision, model size, and floating point operations per second, making it more suitable for industrial applications and providing an efficient, accurate, and lightweight solution for the quality detection of dried green Sichuan pepper.