Detection of maize leaf diseases using improved MobileNet V3-small

Achieving fast and accurate prediction of the fruit mechanical damage area is important to improve the accuracy and efficiency of apple quality grading. In this paper, the spectral data of all samples in the wavelength range from 376 to 1011 nm were collected, the sample set was divided by the physicochemical coeval distance method, and the spectral preprocessing methods were evaluated by establishing a full-wavelength artificial neural network model. The wavelength selection of spectral data was performed by competitive adaptive reweighted sampling, L1 parameter method, and the Pearson correlation coefficient method, and the partial least squares, artificial neural network, and support vector machine (Gaussian kernel) prediction models were established to predict the fruit bruise area size. The surface fitting was performed using the actual apple bruise area, and the regression surface equation of the damage time and damage height of the fruit was established. The results showed that (1) the preprocessing method of first-order difference + SG smoothing can make the prediction model more accurate; (2) the CARS-ANN prediction model has better prediction performance and higher operation efficiency, with the prediction set root mean square error of prediction and R-value of 0.1150 and 0.8675, respectively; (3) the sparrow search algorithm was used to optimize the model, which improved the accuracy of the prediction model. The root mean square error of prediction reached 0.0743 and The R-value reached 0.9739. (4) The relationship between spectral information, bruise area, damage time, and damage degree was obtained by combining the establishment of the fitted surface of the apple bruise area with the prediction model. This study is of application and extension value for the rapid nondestructive prediction of fruit bruise area.

Aimed at the problems of difficult treatment, unreasonable utilization and serious waste of fruit tree residue, combined with the terrain and planting characteristics of hilly orchards, a self-propelled monorail branch chipper was developed. It can realize long distances and large ranges of crushing operations and debris tiling in the garden. Because the monorail branch chipper adopts the half-empty suspension support method, the moving operations and discontinuous cutting of branches can lead to vibration failures or hazards. In response to this problem, modal analysis of a track system with different numbers of nodes by ANSYS software showed that an increase in the number of track sections decreases the natural frequency of each order under the condition of rigid fixation of the support rods, and weakness of vibration is especially seen in sections 1–4, but after a certain amount, the vibrational change tends to level off. The number of lateral rods should be increased for distal short rail branches of less than four sections to reduce operational and operational vibration. The vibration level test results of field multi-condition and multi-point grinding operations showed that the static vibration amplitude of the crusher is basically the same as that of the mobile state. The vibration amplitude of the chipper is significantly increased when in the states from no-load to grinding operation, and the maximum vibration occurs in the left and right direction of the transverse rail, which provides a theoretical basis and optimization direction for further optimization of the single-track branch chipper operational stability problem.

In response to the issues of low-velocity zones and non-uniform jet velocity distribution in the airflow field of traditional air-assisted orchard sprayers, an arc-shaped air outlet suitable for axial-flow air-assisted systems is designed. This article employs the method of CFD numerical simulation and experimental verification to compare and analyze the internal flow field of the air-assisted system and validates the reliability of the numerical simulation results through calculation error and chi-square test. The wind speed of the cross-section is measured at different distances from the outlet, and the distribution characteristics of the outflow field wind speed before and after the structural optimization of the air-assisted system are compared. The horizontal distribution of fog droplets is collected using a fog collection chamber. The experimental results show that the design of the arc-shaped outlet increases the average wind speed of the annular outlet from 14.95 m/s to 18.20 m/s and reduces the proportion of low-speed area from 20.83% to 0.71%. When the rounded corner radius of the air outlet is 50 mm, optimal parameters are attained. The maximum error between the simulated and experimental values is 9.52%. At a significance level of 0.05, the χ² value is 0.252, indicating that the simulated values follow the distribution of the actual measurement values. On the cross-sections located at distances of 0.5, 0.75, 1, 1.25, and 1.5 m from the air outlet, the wind speed distribution with no arc-shaped air outlets exhibits a “low left and high right” type, tending to shift towards the right as a whole. Fog droplets also display a drift tendency towards the right side. The wind speed distribution with arc-shaped air outlets shows a symmetric “high in the middle and low on the sides” type. Fog droplets concentrate in the central position. The optimized air-assisted system can reduce the air field’s low-flow area, increase the airflow distribution uniformity, improve the average wind speed at the outlet, and decrease fog droplet drift. This provides a reference for the structural design of air-assisted systems in current orchard sprayers of the same type.

Accurate detection of Jinxiu Malus fruits in unstructured orchard environments is hampered by frequent overlap, occlusion, and variable illumination. To address these challenges, we propose YOLOv8-MSP-PD (YOLOv8 with Multi-Scale Pyramid Fusion and Proportional Distance IoU), a lightweight model built on an enhanced YOLOv8 architecture. We replace the backbone with MobileNetV4, incorporating unified inverted bottleneck (UIB) modules and depth-wise separable convolutions for efficient feature extraction. We introduce a spatial pyramid pooling fast cross-stage partial connections (SPPFCSPC) module for multi-scale feature fusion and a modified proportional distance IoU (MPD-IoU) loss to optimize bounding-box regression. Finally, layer-adaptive magnitude pruning (LAMP) combined with knowledge distillation compresses the model while retaining performance. On our custom Jinxiu Malus dataset, YOLOv8-MSP-PD achieves a mean average precision (mAP) of 92.2% (1.6% gain over baseline), reduces floating-point operations (FLOPs) by 59.9%, and shrinks to 2.2 MB. Five-fold cross-validation confirms stability, and comparisons with Faster R-CNN and SSD demonstrate superior accuracy and efficiency. This work offers a practical vision solution for agricultural robots and guidance for lightweight detection in precision agriculture.

In order to improve the adaptability of sweet cherry sorters to sweet cherries, a full-view-type grading cup for automated sweet cherry sorters was developed based on the physical characteristics of three species of sweet cherries: Tieton, Huang Mi, and Lapins. The structure and working principle of the full-view-type grading cup are described in this paper. The main factors affecting the operating stability of the grading cup were identified in this study by analyzing the mechanical properties of sweet cherries during the conveying and rotating process. According to the Box-Behnken test design method, we took the operating speed, the Young’s modulus, and the friction coefficient of the double-roller supporter as the test factors and we took the success rate as the test index. The operating parameters of the full-view-type grading cup were tested and studied, and the regression model between the test index and the test factors was established. The influence law of each factor on the test index was analyzed, and the test factors were comprehensively optimized according to the regression model. The results showed that when the operating speed was 48 sweet cherries/min; the butadiene rubber (BR) 9000, with a Young’s modulus of 0.012 GPa was selected as the material for making the double-support rollers; and the friction coefficient was designed to be 15, the success rate was 86.0%. The results of this research have provided a theoretical basis for the design and optimization of grading cups for automated sweet cherry sorters.