Development of single row automatic transplanting device for potted vegetable seedlings

The number of wheat spikes has an important influence on wheat yield, and the rapid and accurate detection of wheat spike numbers is of great significance for wheat yield estimation and food security. Computer vision and machine learning have been widely studied as potential alternatives to human detection. However, models with high accuracy are computationally intensive and time consuming, and lightweight models tend to have lower precision. To address these concerns, YOLO-FastestV2 was selected as the base model for the comprehensive study and analysis of wheat sheaf detection. In this study, we constructed a wheat target detection dataset comprising 11,451 images and 496,974 bounding boxes. The dataset for this study was constructed based on the Global Wheat Detection Dataset and the Wheat Sheaf Detection Dataset, which was published by PP Flying Paddle. We selected three attention mechanisms, Large Separable Kernel Attention (LSKA), Efficient Channel Attention (ECA), and Efficient Multi-Scale Attention (EMA), to enhance the feature extraction capability of the backbone network and improve the accuracy of the underlying model. First, the attention mechanism was added after the base and output phases of the backbone network. Second, the attention mechanism that further improved the model accuracy after the base and output phases was selected to construct the model with a two-phase added attention mechanism. On the other hand, we constructed SimLightFPN to improve the model accuracy by introducing SimConv to improve the LightFPN module. The results of the study showed that the YOLO-FastestV2-SimLightFPN-ECA-EMA hybrid model, which incorporates the ECA attention mechanism in the base stage and introduces the EMA attention mechanism and the combination of SimLightFPN modules in the output stage, has the best overall performance. The accuracy of the model was P=83.91%, R=78.35%, AP= 81.52%, and F1 = 81.03%, and it ranked first in the GPI (0.84) in the overall evaluation. The research examines the deployment of wheat ear detection and counting models on devices with constrained resources, delivering novel solutions for the evolution of agricultural automation and precision agriculture.

In order to improve the efficiency and quality of transplanting vegetables in dry land, based on the seedling technology, transplanting effect and analysis of the planting process, a potted vegetable seedlings transplanting machine was designed. It mainly comprised rotary disc type feeding mechanism, five-bar duckbill type planting mechanism (simulated duckbill mechanism, disc cam, connecting rod, crank, fork, cable) and power transmission system. Based on the physical parameters of the seedlings and design requirements, it was determined that the diameter of the duckbill was D=90 mm, the opening and closing angle was 25°, and the taper angle was 17°. A test bench with adjustable parameters was built by analyzing the structure of the planting mechanism and the motion of the working process. The digital speckle technique was used to optimize the parameters, so that the length of the crankshaft Ι was S1=100 mm, the length of the crankshaft ΙΙ was S2= 80 mm, the length of the connecting rod Ι was S3=140 mm, the length of the connecting rod ΙΙ was S4=260 mm, the length of rod to connecting the rack was S6=314 mm, and the height of planting track was H=450 mm. According to the above parameters and the control requirements of the duckbill mechanism, the cam stroke was determined to be S=15 mm. And the initial phase difference between the two cams was 180°. The experiment was carried out with pepper seedlings as transplanting objects. The results showed that when the planting frequency was 50-70 plants/min, the seedling upright rate was 93%-91.1%, the planting depth qualified rate was 96%-92%. The leakage rate was 0-0.26%, the variation coefficient of plant spacing was 0.37%-0.67%, the injury rate was 0%, and the mechanical damage degree of the mining surface was 2.43-3.77 mm/m2. The machine can effectively improve the quality of transplanting, which can meet the production needs and design requirements of the mechanism. Keywords: seedling transplanting, planting mechanism, digital speckle, parameter optimization, test analysis DOI: 10.25165/j.ijabe.20201301.5494 Citation: Jin X, Cheng Q, Zhao B, Ji J T, Li M Y. Design and test of 2ZYM-2 potted vegetable seedlings transplanting machine. Int J Agric & Biol Eng, 2020; 13(1): 101–110.

Wheat, being a crucial global food crop, holds immense significance for food safety and agricultural economic stability, as the quality and condition of its grains are critical factors. Traditional methods of wheat grain detection are inefficient, and the advancements in deep learning offer a novel solution for fast and accurate grain recognition. This study proposes an improved deep learning model based on YOLOv8n, referred to as YOLO-SDL, aiming to achieve efficient wheat grain detection. A high-quality wheat grain dataset was first constructed, including images of perfect, germinated, diseased, and damaged grains. Multiple data augmentation techniques were employed to enhance the dataset’s complexity and diversity. The YOLO-SDL model incorporates the ShuffleNetV2 architecture in its backbone and combines depthwise separable convolutions (DWConv) with the large separable kernel attention (LSKA) mechanism in its neck structure, significantly improving detection speed and accuracy while ensuring the model remains lightweight. The results indicate that YOLO-SDL achieves superior performance in wheat grain detection, balancing lightweight design and performance optimization. The model achieved a P of 0.942, R of 0.903, mAP50 of 0.965, and mAP50-95 of 0.859, with low computational complexity, making it suitable for resource-constrained environments. These findings demonstrate the efficiency of the ShuffleNetV2, DWConv, and LSKA structures. The proposed YOLO-SDL model provides a new technical solution for agricultural automation and serves as a reliable reference for detecting other crops.

In order to improve the quality of oil peony transplanting, based on the characteristics of peony seedlings and the transplanting effect, the transplanting process was analyzed. A peony seedling transplanting machine was designed, which was mainly composed of a chain-clamp type planting mechanism (clamp, transmission chain, sprocket), slideway, profiling wheels, trencher, the power transmission system and rack. Through the movement analysis of the planting mechanism and its operation process, its structural parameters were optimized. The effective length of the clamp was determined as L1=235 mm, the adjustment range was 235-285 mm, and the height of the planting trajectory was H=340 mm. Based on the physical characteristics and mechanical characteristics of the peony seedlings, the parameters of the slideway and the putter were determined. The effective length of the slideway was determined as L2=420 cm, the width T2=105 mm, the distance between the slideways S=45 mm. The initial angle difference of the putter was θ=13°, and the putter rotation angle β=37°, putter height ratio ε=2. Based on the above parameters and design requirements, the radius of the planting sprocket R1=42 mm, R2=80 mm. “Feng Dan” peony seedlings were used as transplantation test objects. The results showed that: when the planting frequency was 60-75 plants/min, the upright rate was 89.2%-92.8%, the leakage rate was 0-0.28%, the injury rate of seedlings was 0, and the qualified rate of planting depth was 91.7%-95.3%, the variation coefficient of plant spacing was 0.25%-0.54%. This machine can effectively improve the quality and efficiency of transplanting, meeting the requirements of mechanical design and production needs. Keywords: peony transplanting, planting mechanism, oil peony, peony seedlings, parameter design, test analysis DOI: 10.25165/j.ijabe.20201304.5695 Citation: Ji J T, Cheng Q, Jin X, Zhang Z H, Xie X L, Li M Y. Design and test of 2ZLX-2 transplanting machine for oil peony. Int J Agric & Biol Eng, 2020; 13(4): 61–69.

In order to improve the automation and reliability of the mechanized transplanting, a vegetable seedling automatic conveying system was designed. The functional execution component of the system consists of a tank-wheel storage seedling mechanism, a seedling tray conveyor, and a hanging cup circulating conveying mechanism. Combined with the communication technology between host computer and programmable logic controller (PLC), the user interface software of automatic conveying system for potted vegetable was developed by using Visual Studio to control and monitor the transmission progress. The PVDF pressure sensor was used to detect the pressure change of the seedling storage disk to obtain the information of the loading seedling tray. The limit switch and stroke switch were combined to obtain the information of the push rod and feed plate. The timing of each component is controlled by the PLC controller combined with the feedback information of each sensor. The test results show the effect of taking an individual seedling is excellent, the error rate of the pushing mechanism is about 2.08%, and the success rate was 97.91%.