The fruit and vegetable classification problem is an inseparable branch in the field of image recognition. GoogLeNet provides a more ideal solution for problem. We use network to classify apples, lemons, oranges, pomegranates, tomatoes, colored peppers. Through experiments, we obtained training accuracy as 96.88%, testing 96%, speed 11.38 sheets/second. recognition this model can meet basic requirements, but low. Therefore, decided optimize significantly improve further enhance GoogLeNet....

To study the occurrence mechanism of rock burst during mining irregular working face, took panel 7447 near fault tectonic as an engineering background. The spatial fracture characteristic overlying strata was analyzed by Winkler elastic foundation beam theory. Furthermore, influence law width to suspended and limit breaking span key were also thin plate Through micro-seismic monitoring, theoretical analysis, numerical simulation resistance support field measurement, this investigated in...

This paper presents a new numerical modeling procedure and design principle for longwall pillar with the assistance of simulation FLAC3D. A coal mine located in Yanzhou city, Shandong Province, China, was selected this case study. meticulously validated model developed to investigate stress changes across various sizes. In order improve reliability modeling, calibration is undertaken match Salamon Munro strength formula pillar, while similar used estimate stress–strain response gob. The...

In response to the navigation requirements of robots in orchard environments, this paper presents a method for based on point cloud maps. Firstly, pre-acquired maps are preprocessed with pass-through filtering and PCA algorithms obtain suitable path planning. Additionally, tree rows within clustered segmented map orientations. Subsequently, using combination least squares A* algorithm, global operation planning is conducted local Finally, TEB algorithm employed ensure that robot navigates...

Achieving the stringent demands of sustainable tribological industrial applications poses a significant challenge, particular in optimizing self-lubricant performance nitride-based films. This paper tackled this challenge by designing and depositing series ZrN-MoSN composite films with varying (Mo + S)/Zr ratios, employing RF magnetron sputtering, aimed to enhance properties through utilizing high loading capacity ZrN matrix exceptional self-lubricating attributes Mo-S-N additives. After...

Sandstone-like specimens with fissures (fissure inclination 0°, 30°, 45°, 60°, 75°, 90°) were made for investigation into the influence of fissure on deformation damaged rock mass. The high-speed camera was used to record images specimen. Based digital image correlation (DIC) method, strain distribution calculated. evolution pattern fissured mass under effect analyzed via maximum equivalent strain, average and distribution, respectively. results revealed that prefabricated exhibit four...

In order to investigate the effect of fissure inclination on mechanical properties, deformation, and crack evolution surrounding rock in roadway, uniaxial compression experiments were conducted sandstone-like materials with prefabricated fissures. The high-speed camera DIC (digital image correlation) method employed analyze strain distribution specimen. results demonstrated that presence fissures reduces stress for initiation, intact specimens producing new cracks from about 75% peak...

The efficient detection of grapes is a crucial technology for fruit-picking robots. To better identify from branch shading that similar to the fruit color and improve accuracy green due cluster adhesion, this study proposes Shine-Muscat Grape Detection Model (S-MGDM) based on improved YOLOv3 ripening stage. DenseNet fused in backbone feature extraction network extract richer underlying grape information; depth-separable convolution, CBAM, SPPNet are added multi-scale module increase...

In order to solve the problems of domestic and foreign vegetable precision metering devices that easily damage seeds have low seeding efficiency poor qualification rate, this paper designs a one-machine for two-row pneumatic combination seeder, which is mainly composed mechanism honeycomb mechanism. The other components theoretical analysis working process combined seed device clarifies minimum negative pressure adsorbing during suction stage, along with required seed-carrying stage. It also...

To address the current problems of poor seed-filling performance and seed leakage pneumatic filler for cabbage, we designed a disturbing precision cabbage based on auxiliary method. This seeder can completely perturb population entering bottom box casing during process, thereby increasing initial velocity seeds, facilitating smooth progress process. Firstly, carried out theoretical analysis particle dynamics process obtained influencing factors affecting seeder. Secondly, through EDEM...

Vegetable precision planting agronomy is suitable for my country's current vegetable system, and the air-suction planter currently most important work tool in country. This paper designs a kind of air-absorbing sower problems small seeds with grains, poor mobility high difficulty achieving uniform sowing seeds. First, Fluent software used to simulate analyze flow field air chamber seed metering device, pressure velocity fluid are analyzed. Through comparison distribution cloud chart chart,...

Taking the large storage tank as cleaning object, a robot integrating shoveling, crushing, sweeping and dust-absorbing was developed, its system analyzed optimized. Firstly, Fluent-EDEM gas-solid coupling utilized to simulate system. By analyzing fluid distribution particle trajectory, internal structure of dust collection box optimized reduce particles entering fan box. Then, by structural parameters suction nozzle, influence nozzle shoulder angle, length, height on effect explored, were...

As the feeding mass of harvester continues to be increased, accumulation maize mixture on screen more serious, leading a decrease in screening efficiency cleaning screen. The was quickly transported backward through bionic based motion characteristics earthworms reduce accumulation. However, impact main operating parameters this performance not yet clear. Therefore, that affect loss percentage and impurity grain device were simulated using CFD-DEM. Analysis variance, regression response...

Crop diseases have an important impact on the safe production of food. Therefore, automated identification pre-crop is very for farmers to increase and income. In this paper, a tomato leaf disease method based optimized MobileNetV2 model proposed. A dataset 20,400 images was created taken from greenhouse obtained PlantVillage database. The trained with obtain classification diseases. average recognition accuracy 98.3% recall rate 94.9%, which 1.2% 3.9% higher than original model,...