Wheat Seed Detection and Counting Method Based on Improved YOLOv8 Model

Wheat seed detection has important applications in calculating thousand-grain weight and crop breeding. In order to solve the problems of seed accumulation, adhesion, and occlusion that can lead to low counting accuracy, while ensuring fast detection speed with high accuracy, a wheat seed counting method is proposed to provide technical support for the development of the embedded platform of the seed counter. This study proposes a lightweight real-time wheat seed detection model, YOLOv8-HD, based on YOLOv8. Firstly, we introduce the concept of shared convolutional layers to improve the YOLOv8 detection head, reducing the number of parameters and achieving a lightweight design to improve runtime speed. Secondly, we incorporate the Vision Transformer with a Deformable Attention mechanism into the C2f module of the backbone network to enhance the network's feature extraction capability and improve detection accuracy. The results show that in the stacked scenes with impurities (severe seed adhesion), the YOLOv8-HD model achieves an average detection accuracy (mAP) of 77.6%, which is 9.1% higher than YOLOv8. In all scenes, the YOLOv8-HD model achieves an average detection accuracy (mAP) of 99.3%, which is 16.8% higher than YOLOv8. The memory size of the YOLOv8-HD model is 6.35 MB, approximately 4/5 of YOLOv8. The GFLOPs of YOLOv8-HD decrease by 16%. The inference time of YOLOv8-HD is 2.86 ms (on GPU), which is lower than YOLOv8. Finally, we conducted numerous experiments and the results showed that YOLOv8-HD outperforms other mainstream networks in terms of mAP, speed, and model size. Therefore, our YOLOv8-HD can efficiently detect wheat seeds in various scenarios, providing technical support for the development of seed counting instruments.

Study on roadway layout and surrounding rock control of isolated island panel

During the re-mining of historical residual coal resources, the stress environment is complex, the surrounding rock conditions are bad, the mining roadway is significantly affected by ground pressure, the layout is difficult, and the safety is poor. Taking the recovery of isolated island coal pillar in 4# coal seam as the research background, based on the difference in the distribution morphology of the goaf on both sides of the isolated island coal pillar, the stress and failure law of the isolated island panel boundary are studied by numerical simulation method. (1) The peak stress difference of multiple goaf boundaries on both sides of the isolated island coal pillar is between 0.18 and 4.51 MPa. The peak stress is affected by the change of the length of the roof "cantilever beam" at the stopping line of the goaf, so that the peak stress of the goaf boundary is periodic. (2) The high stress is mainly concentrated in the center of the pillar. The peak stress at the end of each pillar is 35-40 MPa. The coal pillar bears high stress, and the stress zone of the original rock moves to the end of the coal pillar. (3) There is a plastic zone of 8-20 m at the corner of the end of each coal pillar. On the basis of the stress zone and failure zone distribution of the goaf boundary on both sides of the isolated island panel, the roadway layout of the isolated island panel is determined, that is, the air-return roadway of the isolated island panel is arranged at random, and the width of the isolated island coal pillar d1 is selected as 10 m. The transport roadway is arranged straight, and the transport roadway of the isolated island panel is in the width section area of the goaf X4103. The width d1 of the isolated island coal pillar is selected to be 8 m, and the length d5-d7 of the mining roadway layout in the width of the coal pillar is 24 m. The roadway of isolated island panel is divided into 4 areas for support control, and the drilling pressure relief technology is proposed for high stress roadway. Through the field monitoring data, it can be seen that the mining roadway can meet the requirements of isolated island coal pillar recovery, which provides reference for the layout and control of abandoned coal roadway in this mine and other mines.

Research on the improvement of wetted perimeter method and application in seasonal rivers-a case study of Fenhe River, China

Environmental flow is vital for maintaining river ecosystem health and ensuring the normal growth of aquatic organisms. The wetted perimeter method is indeed very useful in the assessment of environmental flow due to consideration of stream forms and minimum flow for aquatic life habitat. In this study, a river with obvious seasonality and external water diversion was selected as the typical research object; taking Jingle, Lancun, Fenhe Reservoir, and Yitang hydrological sections as control sections, we improved the existing wetted perimeter method in three aspects: (1) We improved the selection of hydrological data series. The selected hydrological data series should be of a certain length and can well reflect the hydrological changes of wet, normal, and dry years. (2) Different from the traditional wetted perimeter method, which only gives one environmental flow value, the improved method calculates the environmental flow month by month. (3) The improved wetted perimeter method establishes the relationship between native fish survival and environmental flow. Results indicated that the improved wetted perimeter took the survival of the main fishes into consideration, the ratio of the calculated results by the slope method to the multi year average flow was greater than 10%, which can ensure the fishes' habitat is not being destroyed, and the results are more reasonable. Furthermore, the monthly environmental flow processes obtained were better than the annual unified environmental flow value determined by the existing method and are consistent with the natural hydrological situation and water diversion situation of the river. This study shows that the improved wetted perimeter method is feasible for research of river environmental flow with strong seasonal and large variation of annual flow.