Zhenfa Dong

Intelligent detection of tomato seedling transplant quality represents a core technology for advancing agricultural automation. However, in practical applications, existing algorithms still face numerous technical challenges, particularly with prominent issues of false detections and missed detections during recognition. To address these challenges, we developed the ESG-YOLO object detection model and successfully deployed it on edge devices, enabling real-time assessment of tomato seedling transplanting quality. Our methodology integrates three key innovations: First, an EMA (Efficient Multi-scale Attention) module is embedded within the YOLOv8 neck network to suppress interference from redundant information and enhance morphological focus on seedlings. Second, the feature fusion network is reconstructed using a GSConv-based Slim-neck architecture, achieving a lightweight neck structure compatible with edge deployment. Finally, optimization employs the GIoU (Generalized Intersection over Union) loss function to precisely localize seedling position and morphology, thereby reducing false detection and missed detection. The experimental results demonstrate that our ESG-YOLO model achieves a mean average precision mAP of 97.4%, surpassing lightweight models including YOLOv3-tiny, YOLOv5n, YOLOv7-tiny, and YOLOv8n in precision, with improvements of 9.3, 7.2, 5.7, and 2.2%, respectively. Notably, for detecting key yield-impacting categories such as “exposed seedlings” and “missed hills”, the average precision (AP) values reach 98.8 and 94.0%, respectively. To validate the model’s effectiveness on edge devices, the ESG-YOLO model was deployed on an NVIDIA Jetson TX2 NX platform, achieving a frame rate of 18.0 FPS for efficient detection of tomato seedling transplanting quality. This model provides technical support for transplanting performance assessment, enabling quality control and enhanced vegetable yield, thus actively contributing to smart agriculture initiatives.

To address the critical challenges of low success rates and high seedling damage in automatic transplanters for vegetable substrate block seedlings, this study took cabbage substrate block seedlings as the research object and designed a silica gel wheel–synchronous belt clamping seedling separation device. An experimental platform was constructed to perform a three-factor, three-level orthogonal test, investigating the effects of the wheelbase of the silica gel wheel, the inclination angle of the conveyor belt, and the wheelbase of the silica gel wheel and the synchronous belt on seedling separation success rate and substrate block breakage rate. A quadratic regression model was established to analyze the influence of each factor on the index and to optimize the parameter combination verification test. The results showed that the seedling separation effect was better when the wheelbase of the silica gel wheel was 60.47 mm, the inclination angle of the conveyor belt was 8.67°, and the wheelbase of the silica gel wheel and seedling separation synchronous belt was 39.8 mm. The success rate of seedling separation was 90.21% and the substrate block breakage rate was 6.88% in the field verification test of this parameter combination. When the operating speed is 60 plants/min, there is a higher success rate of seedling separation and a lower substrate block breakage rate. This study explored the conditions for stable seedling separation using the seedling separation device, and provided practical reference for the study of the automatic seedling separation of substrate block seedlings.