Intelligent agricultural management system and method based on Internet of Things

Abstract

The invention relates to the technical field of agricultural management, in particular to an intelligent agricultural management system and method based on the Internet of Things. The method comprises the following steps: acquiring farmland geographic data and a farmland environment image; extracting scene feature points of the farmland environment image, and performing farmland three-dimensional modeling in combination with farmland geographic data to generate a digital twin farmland; acquiring to-be-sown seed data; analyzing development characteristics of the to-be-sown seed data to obtain to-be-sown seed development characteristic data; the development characteristic data of the seeds to be sown are imported into the digital twin farmland for sowing simulation and growth simulation, and sowing simulation data and growth simulation data are generated; and performing seeding profile analysis on the seeding simulation data to obtain a seeding profile map. According to the invention, the precision and efficiency of agricultural seeding management are improved by combining digital twinning, the Internet of Things and an intelligent simulation technology.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural management, and particularly to an intelligent agricultural management system and method based on the Internet of Things. Background Art

[0002] Initially, agricultural production mainly relied on manual operations, with relatively traditional monitoring and management means and a low level of informatization. With the development of information technology, technologies such as sensors and automation equipment have gradually been introduced into the agricultural field to improve production efficiency and precision. However, these early technologies were mostly limited to single devices and data collection, and failed to achieve real-time interconnection and comprehensive utilization of information. The Internet of Things technology enables various devices and sensors in the agricultural production process to be connected through the Internet, real-time collect data such as soil humidity, temperature, and climate, and analyze and process them through a cloud platform. Agricultural production can not only be managed in a refined manner, but also automated operations can be realized, such as automatic control of irrigation systems and intelligent fertilization, thereby greatly improving production efficiency and resource utilization rate. However, currently, traditional sowing methods usually rely on experience, lack targeted analysis of the development characteristics of different seeds, and at the same time cannot provide real-time feedback on the impact of sowing depth or position on crop growth, especially the problem of root-shoot ratio imbalance, which further leads to low precision and efficiency in agricultural sowing management. Summary of the Invention

[0003] Based on this, it is necessary to provide an intelligent agricultural management system and method based on the Internet of Things to solve at least one of the above technical problems.

[0004] To achieve the above object, an intelligent agricultural management method based on the Internet of Things, the method includes the following steps:

[0005] Step S1: Obtain farmland geographical data and farmland environment images; extract the scene feature points of the farmland environment images, and combine the farmland geographical data to perform 3D modeling of the farmland to generate a digital twin farmland;

[0006] Step S2: Obtain the data of the seeds to be sown; analyze the development characteristics of the data of the seeds to be sown to obtain the development characteristic data of the seeds to be sown; import the development characteristic data of the seeds to be sown into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data;

[0007] Step S3: Perform a sowing profile analysis on the sowing simulation data to obtain a sowing profile diagram; analyze the root-shoot ratio imbalance of the sowing profile diagram based on the growth simulation data, and adjust the sowing simulation data based on the analysis result of the root-shoot ratio imbalance to obtain optimized sowing simulation data;

[0008] Step S4: Use the Internet of Things to import the sowing optimization simulation data into the intelligent seeder for sowing operation execution, and synchronously collect the sowing operation data; perform real-time sowing position correction on the sowing operation data to execute intelligent agricultural sowing management operations.

[0009] Through the combination of three-dimensional modeling and digital twin farmland, the present invention can accurately simulate the sowing and crop growth processes in a virtual environment. By updating and analyzing data in real time, fine adjustments can be made according to environmental changes and farmland conditions. Such an improvement in accuracy can ensure that crops grow in the most suitable environment, maximizing crop yields. By obtaining and analyzing the development characteristics of the seeds to be sown, customized sowing plans can be provided for different crop seeds, optimizing the growth conditions of different crops. This personalized management can improve the germination rate and growth efficiency of crops and reduce seed waste. The optimization of the root-shoot ratio is a key factor affecting plant health and yield. By analyzing the imbalance of the root-shoot ratio, parameters such as sowing depth and soil fertilization can be adjusted in a timely manner to avoid growth obstacles caused by an unbalanced root-shoot ratio, improve the root development and leaf growth of crops, and promote overall growth health. By continuously collecting and analyzing sowing simulation data, growth simulation data, and sowing operation data, real-time decision-making support can be provided for agricultural managers. Specifically, the growth trend, pest and disease possibilities, etc. of crops can be predicted based on the growth simulation data, so as to take measures for intervention in advance, improving yields and quality. The introduction of intelligent seeders not only realizes precise sowing but also can make dynamic adjustments according to real-time data. Such automated operations can significantly reduce manual intervention and operation errors, improve sowing efficiency, and ensure that crops grow at appropriate depths and spacings, contributing to the uniform distribution of crops, reducing waste and resource waste. Therefore, the present invention improves the accuracy and efficiency in agricultural sowing management by combining digital twin, Internet of Things, and intelligent simulation technologies.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Obtain farmland geographical data through GIS technology; obtain farmland environment images through cameras;

[0012] Step S12: Standardize the coordinates of the farmland geographical data to generate standardized farmland geographical data;

[0013] Step S13: Detect and match feature points in the farmland environment image, and extract the scene feature points of the farmland environment image to obtain scene feature points;

[0014] Step S14: Combine the standardized farmland geographical data and the scene feature points to perform farmland point cloud reconstruction to generate initial farmland point cloud data;

[0015] Step S15: Perform three-dimensional surface modeling on the initial farmland point cloud data to generate a digital twin farmland.

[0016] The present invention obtains farmland geographical data through GIS technology and combines it with the farmland environment images captured by cameras, enabling comprehensive collection of the spatial information and environmental characteristics of the farmland, which provides a high-quality data foundation for subsequent modeling and analysis. The application of GIS technology ensures the accuracy of the farmland geographical data, while the introduction of image data supplements the rich information of the farmland environment. By standardizing the coordinates of the farmland geographical data, the unity of data from different sources in the same coordinate system is ensured, eliminating the errors caused by coordinate system differences. This step lays a solid foundation for subsequent reconstruction and modeling, ensuring the high precision of the three-dimensional modeling of the farmland. By detecting and matching the feature points of the farmland environment images, the key scene feature points in the farmland environment images are extracted, enabling accurate capture of the geometric features of the farmland environment. These feature points provide the necessary data for the point cloud reconstruction of the farmland, further enhancing the accuracy and stability of the modeling process. Combining the standardized farmland geographical data and the extracted scene feature points for farmland point cloud reconstruction can form the initial three-dimensional point cloud data of the farmland. The point cloud data provides detailed structural information of the farmland space, providing important three-dimensional spatial data support for further modeling and analysis. By performing three-dimensional surface modeling on the initial point cloud data of the farmland, a digital twin farmland can be generated, accurately restoring the three-dimensional terrain and environment of the farmland. This modeling method not only improves the visualization effect of the farmland data but also provides virtual environment support for subsequent intelligent management, sowing, irrigation, crop growth monitoring, etc.

[0017] Preferably, step S15 includes the following steps:

[0018] Step S151: Denoise the initial point cloud data of the farmland to generate purified point cloud data;

[0019] Step S152: Perform point cloud surface reconstruction based on the purified point cloud data to generate three-dimensional surface data of the farmland;

[0020] Step S153: Extract the texture features of the farmland environment images and perform surface texture mapping on the three-dimensional surface data of the farmland to generate three-dimensional surface mapping data of the farmland;

[0021] Step S154: Integrate the three-dimensional surface mapping data of the farmland into a virtual scene to generate a digital twin farmland.

[0022] By denoising the initial point cloud data of the farmland, the present invention purifies the point cloud data, can effectively remove the noise and abnormal data in the point cloud, and improve the accuracy and quality of the data. This provides a clean basis for subsequent surface reconstruction and mapping, and ensures the stability and accuracy of the modeling process. By performing point cloud surface reconstruction based on the purified point cloud data, three-dimensional surface data of the farmland can be generated. This step ensures the fine restoration of the farmland terrain, makes the digital twin farmland model more realistic and operable, and can provide a reliable three-dimensional space basis for intelligent agricultural management. Extracting the texture features of the farmland environment images and mapping them onto the three-dimensional surface data of the farmland can enhance the visual effect of the digital twin farmland model. Texture mapping adds color, details and realism to the three-dimensional model of the farmland, helps users more intuitively understand the actual situation of the farmland, and provides more perceptual information for agricultural decision-making. Integrating the three-dimensional surface mapping data of the farmland into a virtual scene can create a complete and interactive digital twin farmland environment. This virtual scene provides users with a panoramic view of the farmland, can simulate and observe the changes of the farmland in real time, and optimize the management and decision-making of the farmland.

[0023] Preferably, step S2 includes the following steps:

[0024] Step S21: Obtain the data of the seeds to be sown;

[0025] Step S22: Extract the seed type, variety and growth cycle of the seeds to be sown, classify the seeds to be sown based on the data, and generate seed classification data;

[0026] Step S23: Analyze the development characteristics of the seeds to be sown based on the seed classification data to obtain the development characteristic data of the seeds to be sown;

[0027] Step S24: Perform time series analysis on the development characteristic data of the seeds to be sown, extract the development parameters at different growth stages, and generate stage development characteristic data; import the stage development characteristic data into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data.

[0028] By obtaining the data of the seeds to be sown, the invention can accurately record key information such as the types, varieties, and sources of different seeds, which provides detailed basic data for subsequent seed classification and analysis of developmental characteristics, ensuring more scientific seed selection and management during the sowing process. By extracting information such as the types, varieties, and growth cycles of the seeds to be sown, the seeds are accurately classified to generate seed classification data. This classification data provides a clear basis for subsequent seed management and sowing decisions, ensuring the selection of the most suitable seeds under different farmland conditions, thereby improving crop yield and quality. By analyzing the developmental characteristics based on the seed classification data, the specific growth characteristics of the seeds to be sown can be obtained. These developmental characteristic data include the germination speed, root system development, resistance characteristics, etc. of the seeds, which can provide a scientific basis for the formulation of sowing strategies and management measures, ensuring the healthy growth of the seeds after sowing. By performing time series analysis on the developmental characteristic data of the seeds to be sown, key developmental parameters at different growth stages can be extracted. These stage data can help agricultural managers accurately grasp the requirements of the seeds at different growth stages, optimize sowing timing, depth, and density, etc., thereby improving the efficiency and quality of crop growth. Importing the stage developmental characteristic data into the digital twin farmland for sowing simulation and growth simulation can generate more accurate sowing simulation data and growth simulation data, which enables farmland managers to preview the sowing and growth processes in a virtual environment, adjust the planting plan in a timely manner, and reduce potential risks.

[0029] Preferably, the importing the stage developmental characteristic data into the digital twin farmland for sowing simulation and growth simulation in step S24 includes:

[0030] Importing the stage developmental characteristic data into the digital twin farmland for sowing simulation parameter setting to obtain sowing simulation setting parameters: the sowing depth is set to 3 cm, the seed density is 100 seeds per square meter, the seed germination rate is 85%, the soil temperature is set to 20 °C, and the soil humidity is 0.22 cm 3 / cm 3 , the initial growth rate of the seeds is 1.2 mm per day, the temperature adaptability range is 18 °C to 30 °C, the water demand is 5 mm per day, the light demand is 12 hours per day, the nutrient demand is 10 kg / ha of nitrogen, 5 kg / ha of phosphorus, and 8 kg / ha of potassium, the crop growth cycle is 120 days, the root growth rate is 0.8 cm per day, and the crop stress resistance score is 4 points;

[0031] Importing the stage developmental characteristic data into the digital twin farmland for growth simulation parameter setting to obtain growth simulation setting parameters: the initial growth rate is 1.5 mm per day, the nutrient demand is 15 kg / ha of nitrogen, 10 kg / ha of phosphorus, and 12 kg / ha of potassium, the maximum leaf area index is 3.5, and the photosynthetically active radiation demand is 15 MJ / m 2 / day, the water evaporation rate is set at 4 mm / day, the root system expansion rate is 2 cm / day, the temperature requirement for the growth period is between 15°C and 30°C, the growth stage distribution is 30 days for the initial growth period, 60 days for the rapid growth period, 30 days for the maturity period, the light response coefficient is 0.9, the impact of pests and diseases is 0.05, the environmental humidity requirement is 70% to 85%, and the crop maturity period is 90 days;

[0032] Use agricultural simulation software to perform simulation analysis on the sowing simulation setting parameters and the growth simulation setting parameters to generate sowing simulation data and growth simulation data.

[0033] By setting specific sowing parameters of the present invention, such as sowing date, sowing depth, seed density, germination rate, soil temperature and humidity, etc., the sowing process under different environmental conditions can be accurately simulated. Through these detailed parameter settings, precise guidance can be provided for sowing operations to ensure that the sowing depth and density meet the growth requirements of seeds, thereby improving the germination rate and initial growth rate of seeds. By setting key parameters such as the germination rate and initial growth rate of seeds, the growth of seeds under different environments can be better simulated. Setting reasonable conditions such as seed density, soil temperature and humidity helps to optimize the germination environment of seeds, thereby increasing the germination rate and the speed of initial growth and reducing the failure rate in the initial planting stage. In growth simulation, by setting detailed growth parameters, such as nutrient requirements, leaf area index, photosynthetically active radiation requirements, water evaporation rate, root system expansion rate, etc., the requirements of crops in different growth stages can be more accurately simulated. These data can help managers adjust water and fertilizer management, environmental control, etc. in the planting process to ensure the best environmental conditions for crops in each growth stage. By accurately dividing the growth stages (initial growth period, rapid growth period, maturity period), the different requirements of crop growth can be better understood and predicted. Applying these phased requirements to the simulation helps to optimize farmland management, improve the growth efficiency and health of crops, and avoid waste of resources. In growth simulation, the settings of factors such as light response coefficient, impact of pests and diseases, and environmental humidity can simulate the performance of crops under different environmental conditions. By adjusting these parameters, the light, water and nutrient supply can be optimized in a virtual environment to ensure the healthy development of crops throughout the growth cycle.

[0034] Preferably, the sowing profile analysis of the sowing simulation data in step S3 includes:

[0035] Extract the seed position information and soil temperature information of the sowing simulation data;

[0036] Based on the seed position information, analyze the sowing simulation depth of the sowing simulation data to obtain the seed distribution data at different depths;

[0037] Extract the soil temperature and soil humidity from the soil temperature information, and perform soil temperature interpolation and soil humidity interpolation on the soil temperature and soil humidity respectively to generate continuous soil temperature profile data and continuous soil humidity profile data;

[0038] Visualize the soil temperature profile data and the soil humidity profile data according to the seed distribution data to generate a seeding profile diagram.

[0039] By extracting the seed position information in the seeding simulation data, the present invention can clearly understand the position and distribution of each seed. Analyzing the seeding simulation depth can determine the accurate distribution of seeds in the soil, which provides data support for the optimization and management of seeding depth and helps improve the germination rate and initial growth effect of seeds. By extracting the soil temperature and humidity information and performing interpolation respectively, continuous soil temperature profile data and soil humidity profile data can be obtained, which provides a detailed reference for accurately understanding the impact of soil conditions on crop growth. Temperature and humidity are key factors affecting seed germination and initial growth, and continuous soil temperature and humidity data help analyze how these factors affect different stages of crop growth. Combining the seed distribution data with the soil temperature and humidity data can comprehensively evaluate the impact of seeding depth on seed growth. At different depths, the germination rate and growth status of seeds are affected by different temperature and humidity conditions. Providing this comprehensive perspective helps optimize the seeding depth, thereby improving the healthy growth and stress resistance of crops. By combining the soil temperature profile data, the soil humidity profile data with the seed distribution data, a seeding profile diagram is generated. The visualized seeding profile diagram can help agricultural managers intuitively understand the impact of soil environmental conditions during seeding on seed germination and growth, so as to better manage the land and regulate crop growth. By analyzing the seed distribution at different depths, the optimal seeding depth can be found, so that the seeds can obtain the most suitable soil temperature and humidity conditions at the initial germination stage. By adjusting the seeding depth, the germination rate and early growth rate of seeds are improved, and the seeding effect is optimized.

[0040] Preferably, the root-shoot ratio imbalance in analyzing the seeding profile diagram according to the growth simulation data in step S3 includes:

[0041] Extract the morphological characteristics of the seeded crops in the growth simulation data, and calculate the growth rate of the growth simulation data according to the morphological characteristics to obtain dynamic growth data;

[0042] Based on the dynamic growth data, perform growth stratification of the seeded crops on the growth simulation data to obtain upper-layer growth data and lower-layer growth data; analyze the root spatial structure of the seeded crops for the lower-layer growth data to generate root characteristic data;

[0043] Calculate the biomass of the upper-layer growth data to obtain canopy characteristic data;

[0044] Ratio calculations are performed using root characteristic data and canopy characteristic data to obtain the root-to-shoot ratio of the crop;

[0045] An imbalance assessment of the root-to-shoot ratio of the crop is carried out according to the sowing profile diagram to generate an analysis result of the root-to-shoot ratio imbalance.

[0046] In the present invention, by extracting the morphological characteristics of the sown crop from the growth simulation data and calculating the growth rate, the growth process of the crop can be dynamically tracked. This dynamic growth data provides key data support for analyzing the growth trend, periodic changes, and growth stages of the crop, helping farmers to timely adjust agricultural management strategies to ensure that the crop is provided with appropriate resource supply at different growth stages. Conducting growth stratification of the sown crop on the growth simulation data can separate the upper and lower layer growth of the crop for separate analysis. This hierarchical analysis method makes the management of different growth regions (such as between the ground and the canopy, between the root system and the ground) more targeted, and can adjust management measures such as fertilization and irrigation according to the needs of different layers. Analyzing the spatial structure of the sown crop root system for the lower layer growth data to generate root characteristic data provides a detailed reference for understanding the health and growth status of the crop root system. The root system is the key part for the crop to obtain water and nutrients, and accurate root characteristic analysis helps to improve the accuracy of water and fertilizer management to ensure that the crop root system is fully supported. By calculating the biomass of the upper layer growth data, the characteristic data of the crop canopy can be obtained, reflecting the photosynthesis ability and overall growth status of the crop. The biomass of the canopy is an important indicator for measuring the health of the crop, and high-quality canopy biomass contributes to the photosynthesis of the crop, thereby increasing the crop yield and stress resistance. Through the ratio calculation of the root characteristic data and the canopy characteristic data, the root-to-shoot ratio of the crop can be obtained. The root-to-shoot ratio is an important indicator reflecting the growth balance of the crop, and too high or too low root-to-shoot ratio will affect the growth and development of the crop. By calculating and adjusting the root-to-shoot ratio, a more balanced growth environment can be provided for the crop, optimizing the resource allocation of the crop and improving its growth efficiency.

[0047] Preferably, the adjustment of the sowing simulation data based on the analysis result of the root-to-shoot ratio imbalance in step S3 includes:

[0048] Comparing the analysis result of the root-to-shoot ratio imbalance with a preset root-to-shoot ratio imbalance threshold. When the analysis result of the root-to-shoot ratio imbalance is greater than or equal to the root-to-shoot ratio imbalance threshold, abnormal seed sowing data is generated;

[0049] Analyzing the abnormal factors of the abnormal seed sowing data to obtain abnormal seed sowing factors, where the abnormal seed sowing factors are at least one of the factors of abnormal seed density, abnormal sowing depth, abnormal nutrient supply amount, and abnormal water;

[0050] Adjusting the sowing simulation data through the abnormal seed sowing factors:

[0051] When it is confirmed that the abnormal factor of seed sowing is abnormal seed density, the seed density in the sowing simulation data is increased or decreased by 10%-20%; when it is confirmed that the abnormal factor of seed sowing is abnormal sowing depth, the sowing depth in the sowing simulation data is increased or decreased by 2-5 cm; when it is confirmed that the abnormal factor of seed sowing is abnormal nutrient supply amount, the nitrogen fertilizer supply amount in the sowing simulation data is increased by 10%-15% kg / ha, the phosphorus fertilizer supply amount is decreased by 5%-10% kg / ha, and the potassium fertilizer supply amount is increased by 10%-15% kg / ha; when it is confirmed that the abnormal factor of seed sowing is abnormal moisture, the irrigation amount in the sowing simulation data is increased or decreased by 10%-20%, and the optimized sowing simulation data is integrated.

[0052] By comparing the analysis result of root-shoot ratio imbalance with the preset root-shoot ratio imbalance threshold, the present invention can accurately identify the abnormal factors in the sowing process. Adjusting according to the situation of root-shoot ratio imbalance can more precisely determine sowing parameters such as seed density, sowing depth, nutrients and water supply, which helps to improve the sowing accuracy, avoid problems such as uneven seed distribution, excessive or insufficient nutrient supply, and optimize the sowing effect. Root-shoot ratio imbalance usually leads to uneven crop growth and affects the resource allocation between roots and canopy. By timely detecting root-shoot ratio imbalance and adjusting the sowing simulation data, the problems of root and canopy imbalance in the planting process can be effectively solved, thereby enhancing the growth adaptability of crops in different environments. Adjusting the sowing data according to the analysis result of root-shoot ratio imbalance can provide more targeted improvements for sowing strategies. Specifically, adjusting the sowing density when abnormal seed density is found, adjusting the sowing depth when abnormal sowing depth is found, or correspondingly adjusting the fertilizer and irrigation amount when abnormal nutrient and water supply is found. This refined adjustment will greatly improve the planting success rate of crops and ensure that crops obtain the best conditions at each growth stage.

[0053] Preferably, the real-time sowing position correction of the sowing operation data in step S4 includes:

[0054] Importing the optimized sowing simulation data into the digital twin farmland for sowing path analysis to generate the theoretical sowing position;

[0055] Performing real-time positioning processing on the sowing operation data of the seeder to generate absolute coordinate data;

[0056] Performing attitude solution on the sowing operation data according to the absolute coordinate data to generate the seeder sowing position data;

[0057] Comparing the seeder sowing position data with the theoretical sowing position to generate spacing deviation data;

[0058] Use the spacing deviation data to perform sowing position motion compensation on the sowing operation data to execute intelligent agricultural sowing management operations.

[0059] Through real-time positioning processing and attitude solution of the seeder, the present invention can accurately obtain the absolute position and sowing position of the seeder in the field. After comparing with the theoretical sowing position, the deviation can be timely detected and corrected, so as to ensure that the seeder sows accurately at the predetermined position in each operation step, which helps to improve the sowing accuracy and avoid the influence of errors on crop growth. This process uses the digital twin farmland for sowing path analysis, combines real-time positioning and attitude solution, and improves the automation and intelligence level of sowing operations through the immediate processing and compensation of deviation data. The automatic compensation system can reduce manual intervention, improve sowing efficiency and accuracy, and support precision agriculture management. The accurate sowing position can ensure that each seed is in the best soil conditions, thus promoting the uniform germination and growth of seeds. By reducing the sowing position deviation, the growth environment of crops is more consistent, which helps to improve the uniformity of field crops, ensure the consistency of crop development stages, and is beneficial to improving crop yield and quality. Real-time sowing position correction can automatically identify the deviation generated during sowing, and then dynamically adjust the sowing path. This correction enables the seeder to complete the sowing task in a shorter time and avoid the phenomenon of re-sowing or missed sowing caused by deviation, improving operation efficiency and resource utilization rate.

[0060] In this specification, an Internet of Things-based intelligent agriculture management system is provided for executing the above-mentioned Internet of Things-based intelligent agriculture management method. The Internet of Things-based intelligent agriculture management system includes:

[0061] A three-dimensional modeling module for obtaining farmland geographical data and farmland environment images; extracting scene feature points of the farmland environment image, and combining the farmland geographical data to perform three-dimensional modeling of the farmland to generate a digital twin farmland;

[0062] A three-dimensional simulation module for obtaining data of seeds to be sown; analyzing the growth characteristics of the seeds to be sown to obtain growth characteristic data of the seeds to be sown; importing the growth characteristic data of the seeds to be sown into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data;

[0063] A sowing analysis module for performing sowing profile analysis on the sowing simulation data to obtain a sowing profile diagram; analyzing the root-shoot ratio imbalance of the sowing profile diagram based on the growth simulation data, and adjusting the sowing simulation data based on the analysis result of the root-shoot ratio imbalance to obtain optimized sowing simulation data;

[0064] A position correction module is used to import sowing optimization simulation data into an intelligent seeder through the Internet of Things for sowing operation execution, and synchronously collect sowing operation data; perform real-time sowing position correction on the sowing operation data to execute intelligent agricultural sowing management operations.

[0065] The beneficial effects of the present invention are as follows: By obtaining farmland geographical data and farmland environment images, a three-dimensional model of the farmland can be accurately reconstructed, which helps to provide detailed spatial data support for crop growth and sowing activities. Combining the scene feature points of the environment image and the geographical data, the generated digital twin farmland can accurately reflect the terrain, landform and environmental conditions of the actual farmland, which provides a more intuitive and detailed visualization tool for agricultural management. Through digital twin technology, the farmland state can be synchronized and updated in real time, providing high-precision data support for subsequent simulation, monitoring and management, thereby improving the intelligence and accuracy of agricultural production. By analyzing the development characteristics of the seeds to be sown, the sowing strategy can be adjusted according to the growth requirements of different seeds to ensure that the seeds germinate and grow under the best conditions. The generated sowing simulation data and growth simulation data provide a detailed crop growth model for agricultural managers. These simulation data can help predict each stage of crop growth and provide a basis for later management. Combining the seed development characteristic data with the model of the digital twin farmland makes the development characteristics of the seeds match the environmental conditions, thereby improving the sowing efficiency and crop yield. By performing a sowing profile analysis on the sowing simulation data, key factors such as the distribution of seeds in the soil, soil temperature and humidity can be revealed, which provides strong data support for optimizing the sowing depth and density. Analyzing the root-shoot ratio imbalance in the sowing profile can accurately evaluate the growth potential and space utilization efficiency of the seeds, providing a scientific basis for optimizing the sowing strategy. By adjusting the sowing simulation data, the problem of root-shoot ratio imbalance can be effectively solved, promoting the healthy growth of crops. Based on the adjustment of the root-shoot ratio imbalance, sowing optimization simulation data can be generated, thereby improving the sowing accuracy and the growth condition of the crops, and providing a more controllable management strategy for agricultural production. By using the Internet of Things technology to collect sowing operation data in real time and synchronously correct the position of the seeder, it can ensure that the seeder maintains an accurate sowing path during operation, avoiding uneven sowing caused by deviation. Real-time sowing position correction improves the automation and intelligence level of sowing operations, reduces manual intervention, and improves the sowing efficiency and accuracy. Precise sowing position and path control can maximize the utilization of land resources, ensure that the seeds are sown at the appropriate position and depth, and improve the spatial consistency and uniformity of crop growth. Therefore, the present invention improves the accuracy and efficiency in agricultural sowing management by combining digital twin, Internet of Things and intelligent simulation technologies. Description of the Drawings

[0066] Figure 1 It is a schematic diagram of the step flow of a smart agriculture management method based on the Internet of Things;

[0067] Figure 2 is Figure 1 a detailed implementation step flow diagram of step S1 in

[0068] Figure 3 is Figure 1 a detailed implementation step flow diagram of step S2 in

[0069] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0070] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0071] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0072] It should be understood that although the terms "first", "second", etc. may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be called the second unit, and similarly the second unit may be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0073] To achieve the above object, please refer to Figures 1 to 3 , a smart agriculture management method based on the Internet of Things, the method includes the following steps:

[0074] Step S1: Obtain farmland geographical data and farmland environmental images; extract the scene feature points of the farmland environmental images, and combine the farmland geographical data to perform three-dimensional modeling of the farmland to generate a digital twin farmland;

[0075] Step S2: Obtain the data of the seeds to be sown; analyze the developmental characteristics of the seeds to be sown to obtain the developmental characteristic data of the seeds to be sown; import the developmental characteristic data of the seeds to be sown into the digital twin farmland for sowing simulation and growth simulation, and generate sowing simulation data and growth simulation data;

[0076] Step S3: Conduct a sowing profile analysis on the sowing simulation data to obtain a sowing profile diagram; analyze the root-shoot ratio imbalance of the sowing profile diagram based on the growth simulation data, and adjust the sowing simulation data based on the analysis result of the root-shoot ratio imbalance to obtain optimized sowing simulation data;

[0077] Step S4: Use the Internet of Things to import the optimized sowing simulation data into the intelligent seeder for sowing operation execution, and synchronously collect sowing operation data; perform real-time sowing position correction on the sowing operation data to execute intelligent agricultural sowing management operations.

[0078] Through the combination of three-dimensional modeling and digital twin farmland, the present invention can accurately simulate the sowing and crop growth processes in a virtual environment. By updating and analyzing data in real time, fine adjustments can be made according to environmental changes and farmland conditions. Such an improvement in accuracy can ensure that crops grow in the most suitable environment, maximizing crop yields. By obtaining and analyzing the developmental characteristics of the seeds to be sown, customized sowing plans can be provided for different crop seeds, optimizing the growth conditions of different crops. This personalized management can improve the germination rate and growth efficiency of crops and reduce seed waste. The optimization of the root-shoot ratio is a key factor affecting plant health and yield. By analyzing the root-shoot ratio imbalance, parameters such as sowing depth and soil fertilization can be adjusted in a timely manner to avoid growth obstacles caused by an unbalanced root-shoot ratio, improve the root development and leaf growth of crops, and promote overall growth health. By continuously collecting and analyzing sowing simulation data, growth simulation data, and sowing operation data, real-time decision-making support can be provided for agricultural managers. Specifically, the growth trend, pest and disease probability, etc. of crops can be predicted based on the growth simulation data, so as to take measures in advance for intervention to improve yield and quality. The introduction of intelligent seeders not only realizes precise sowing but also can make dynamic adjustments according to real-time data. Such automated operations can significantly reduce manual intervention and operation errors, improve sowing efficiency, and ensure that crops grow at appropriate depths and spacings, helping to evenly distribute crops, reduce waste and resource waste. Therefore, the present invention improves the accuracy and efficiency in agricultural sowing management by combining digital twin, Internet of Things, and intelligent simulation technologies.

[0079] In the embodiment of the present invention, refer to Figure 1 As shown, it is a schematic diagram of the step flow of a smart agriculture management method based on the Internet of Things according to the present invention. In this example, the smart agriculture management method based on the Internet of Things includes the following steps:

[0080] Step S1: Obtain farmland geographical data and farmland environmental images; extract the scene feature points of the farmland environmental images, and combine the farmland geographical data to perform 3D modeling of the farmland to generate a digital twin farmland;

[0081] In the embodiment of the present invention, high-precision geographical coordinate data of the farmland is obtained through GNSS (Global Navigation Satellite System) and RTK (Real-Time Kinematic), and the measurement error is controlled within ±2 cm. At the same time, large-scale farmland distribution information is obtained by combining remote sensing images (resolution 0.3 m). Meanwhile, a drone is used to carry a multispectral camera (band range 400–1000 nm) to collect farmland environmental images at a height of 120 m, and information such as crop growth status and surface moisture is recorded at a resolution of 5 cm / pixel. Subsequently, the SIFT algorithm is used to extract the feature points in the environmental images, and about 5000 feature points are extracted on average for each image. More than 90% of the feature points are accurately aligned through the FLANN matching algorithm, and spatial registration is performed in combination with the geographical coordinate information. In the stage of 3D modeling of the farmland, the SfM (Structure from Motion) method is used to generate 3D point clouds, and the average point density reaches 1000 points / m 2 , and at the same time, the elevation accuracy is improved to ±5 cm by combining LiDAR scanning data. Subsequently, meshing is performed through the Poisson Surface Reconstruction algorithm, and finally a high-precision farmland model containing 1 million to 5 million patches is generated, and Texture Mapping technology is used for texture mapping to make the model more realistic. In addition, in combination with a soil moisture monitoring sensor (measurement range 0–60% water content, error ±1%), the dynamic update of the digital twin farmland is realized. Finally, the model is visualized by using WebGL and Cesium.js technologies, and users can view information such as farmland terrain and crop status at a 1:1 real scale to generate a digital twin farmland.

[0082] Step S2: Obtain the data of the seeds to be sown; analyze the development characteristics of the data of the seeds to be sown to obtain the development characteristic data of the seeds to be sown; import the development characteristic data of the seeds to be sown into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data;

[0083] In the embodiments of the present invention, by obtaining the basic data of the seeds to be sown, including the thousand-grain weight (such as 40±2 g for wheat and 320±10 g for corn), the seed moisture content (12%±1%), and the internal cavity rate (≤1%) and the germination embryo integrity rate (≥95%) detected by X-ray. Subsequently, the protein content (12.5±0.8% for wheat and 38±2% for soybeans) and the starch content (70±3% for corn) of the seeds are analyzed using near-infrared spectroscopy (900–2500 nm), and a 48-hour germination test is carried out at 25°C and 60% relative humidity. The germination rate of wheat is measured to be 92±3%, the root length growth rate is 3.2±0.5 mm / day, and the shoot length growth rate is 4.5±0.7 mm / day. Combining the temperature and humidity experimental data, a seed germination adaptability model is established, and the optimal growth temperature for wheat is calculated to be 18–25°C, with the critical temperature below 10°C or above 35°C. Then, in the digital twin farmland environment, the sowing process is simulated, the sowing depth of wheat is set to 3±0.5 cm, the plant spacing is 5 cm, the row spacing is 20 cm, and the coefficient of variation of sowing uniformity CV≤15%. Through three-dimensional physical simulation, the settlement and covering state of the seeds in the soil are simulated, and high-precision sowing point cloud data (resolution 2 cm) is generated. In the growth simulation stage, combining the WOFOST model and meteorological data (such as the average temperature and precipitation in the past 5 years), the growth rates of wheat at each stage during the 120±5-day growth cycle are predicted, such as the average daily growth height of 1.5±0.3 cm during the jointing stage. The leaf area index (LAI) is used to simulate the vegetation coverage, LAI = 0.3 in the seedling stage, LAI = 2.5 in the jointing stage, and LAI = 5.8 in the heading stage. At the same time, combining historical disease data, the incidence of scab is predicted to increase to 20% in the high-temperature and high-humidity environment. Finally, the system outputs sowing simulation data (sowing depth, seed distribution, sowing uniformity) and growth simulation data (growth cycle, biomass change, environmental adaptability), providing precise decision-making support for intelligent farmland management.

[0084] Step S3: Conduct a sowing profile analysis on the sowing simulation data to obtain a sowing profile diagram; analyze the root-shoot ratio imbalance of the sowing profile diagram based on the growth simulation data, and adjust the sowing simulation data based on the analysis results of the root-shoot ratio imbalance to obtain optimized sowing simulation data;

[0085] In the embodiments of the present invention, data such as the sowing depth of seeds, the surrounding soil structure, and the water content distribution are collected through three-dimensional laser scanning technology (resolution 1 mm) and high-resolution spectral imaging (spectral range 400–1000 nm). Combining the sowing point cloud data (resolution 2 cm), a farmland sowing profile diagram is generated, and key indicators such as the seed position, the initial growth direction of the root system, and the soil density are marked. The sowing depth distribution in the profile diagram is extracted. For example, the average sowing depth of wheat seeds is 3.2 ± 0.4 cm, and the coefficient of variation CV = 10%. The soil moisture content around the seeds is calculated (target range 40%–60%), and its influence on root germination is analyzed. Combining the soil profile data, the settlement stability of seeds in different regions is evaluated, and abnormal regions affected by soil compaction are screened out. According to the growth simulation data, the root growth length and canopy height in different growth periods are extracted. Specifically, the root length of wheat seedlings is 4.5 ± 0.5 cm, and the canopy height is 6.2 ± 0.7 cm, then the root-shoot ratio R / S = 0.73. The root-shoot ratio is calculated respectively at the tillering stage, jointing stage, and heading stage, and a growth dynamic curve is established. For example, at the jointing stage, R / S = 0.55, and at the heading stage, R / S = 0.48. If the root-shoot ratio is too low (R / S < 0.5), it indicates insufficient root development, which is caused by too shallow sowing, insufficient soil nutrients, or excessive water. If the root-shoot ratio is too high (R / S > 1.0), it indicates restricted growth of the above-ground part, which is caused by excessive soil compaction, insufficient seed vigor, or water deficit. Combining multi-dimensional data (root length, canopy height, soil environment), the imbalance of the root-shoot ratio in different regions is classified, and an imbalance distribution map is output. For the area with too shallow sowing (<2.5 cm), the depth is appropriately increased to 3.5 ± 0.3 cm to enhance the root development ability. For the area with too deep sowing (>4.5 cm), the depth is reduced to 3.0 ± 0.2 cm to avoid hindrance to seed emergence. Through sowing profile analysis, it is found that in the area with too high soil compaction degree (density > 1.6 g / cm 3 ), the pressure parameters of the sowing machine are adjusted to reduce the compaction effect and improve the root extension space of the seeds. In the area where the soil moisture content is lower than 40%, the local irrigation amount is increased to make the soil humidity reach 50% ± 5%. Combining the optimized data, the seed spacing is readjusted to avoid excessive competition. For example, the plant spacing of wheat is optimized to 6 cm to improve ventilation and light transmission. The sowing is retested in the optimized simulation environment to ensure that the coefficient of variation CV of the adjusted sowing uniformity ≤ 8%. The adjusted sowing depth distribution, soil moisture content, density and other parameters are recorded, and an optimized sowing profile diagram is generated. The optimized root-shoot ratio curve is calculated, with the goal of stabilizing R / S at 0.6–0.7 at the jointing stage and 0.45–0.55 at the heading stage to improve the growth balance of crops. The optimized sowing point cloud data, soil environment data, and growth model parameters are output to provide optimized references for subsequent precision farmland management.

[0086] Step S4: Use the Internet of Things to import the seeding optimization simulation data into the intelligent seeder for seeding operation execution, and synchronously collect seeding operation data; perform real-time seeding position correction on the seeding operation data to execute intelligent agricultural seeding management operations.

[0087] In the embodiments of the present invention, the optimized sowing simulation data (including sowing depth, plant spacing, row spacing, soil humidity, density, etc.) is transmitted to the intelligent seeder through the cloud platform, and standard Internet of Things protocols (such as MQTT or LoRa) are used to ensure efficient and low-latency data transmission. After the data is formatted, it is imported into the intelligent seeder control system, including sowing path planning, sowing depth setting, sowing speed and accuracy setting. The data reception rate supported by the system is 50 pieces of data per second to ensure real-time performance and accuracy. After receiving the optimized data, the intelligent seeder automatically adjusts the sowing depth (such as 3.2 ± 0.4 cm), plant spacing (such as 6 cm) and row spacing (such as 20 cm), and adjusts the sowing speed and seed delivery amount according to the soil environment (such as moisture content, density). Sensors in the seeder (including depth sensors, GPS modules, humidity sensors, temperature sensors) continuously monitor the operating environment and machine status, and return the data to the control system. During the execution of the sowing operation, the intelligent seeder continuously collects operation data through a variety of integrated sensors (GPS positioning, depth sensors, seed detectors, humidity monitors). The sowing data includes: real-time sowing position (GPS coordinates), actual sowing depth, soil humidity, operation speed, etc. The data collection frequency is 10 times per second to ensure that every change during the sowing process can be accurately captured. The operation data is synchronously uploaded to the cloud platform for real-time storage and analysis. Through the Internet of Things gateway (such as 4G / 5G module or LoRa communication), the real-time collected sowing data is transmitted to the central control platform, and the platform conducts data analysis and visualization. The system continuously monitors every link of the sowing operation, including seed delivery amount, depth change, soil humidity deviation, etc., to ensure the accuracy of the operation. During the operation, the GPS and ground RTK positioning systems are used to correct the sowing position in real time. Assuming the target accuracy is ±2 cm, if it is found that the sowing position deviation exceeds the threshold (such as ±3 cm), the system will automatically adjust the driving trajectory of the seeder. If it is detected that there is an error in the sowing depth (such as too shallow or too deep), through the feedback signal of the depth sensor, the system automatically controls the depth adjustment mechanism of the seeder to ensure that the depth is controlled within the target range. The system performs position correction every 5 seconds to ensure the accuracy of the entire sowing process. Based on the real-time position and operation data, the central platform can continuously monitor the progress of the sowing operation to ensure that each area meets the predetermined sowing standards. Through data analysis, the platform can identify problem areas in the operation (such as inconsistent sowing depth, abnormal seed delivery amount, etc.), and remind the operator through the operation interface or automatically adjust the seeder settings. During the operation, the system can automatically adjust the sowing depth and delivery amount according to environmental factors such as soil humidity and temperature. Specifically, in areas with higher humidity, the sowing depth is reduced to avoid seed waterlogging. After the operation is completed, the intelligent seeder feeds back the sowing data (including operation duration, sowing depth, sowing uniformity, operation area coverage, etc.) to the cloud platform for data analysis and storage.By analyzing the uniformity of seeding depth (e.g., the coefficient of variation CV of seeding depth ≤ 5%), the seeding quality is evaluated and the subsequent operation parameters are optimized. Based on the data feedback, the system can generate a seeding operation report and provide optimization suggestions. Specifically, if the seeding depth in some areas is too shallow, the system suggests increasing the seed covering depth or adjusting the seeding speed.

[0088] As an example of the present invention, refer to Figure 2 As shown, in this example, step S1 includes:

[0089] Step S11: Obtain the farmland geographical data through GIS technology; obtain the farmland environment image through a camera;

[0090] Step S12: Standardize the coordinates of the farmland geographical data to generate standardized farmland geographical data;

[0091] Step S13: Detect and match the feature points of the farmland environment image, and extract the scene feature points of the farmland environment image to obtain the scene feature points;

[0092] Step S14: Combine the standardized farmland geographical data and the scene feature points to reconstruct the farmland point cloud and generate the initial farmland point cloud data;

[0093] Step S15: Perform three-dimensional surface modeling on the initial farmland point cloud data to generate a digital twin farmland.

[0094] In the embodiments of the present invention, by using GIS (Geographic Information System) technology, precise geographical data of farmland is obtained through drones or satellite remote sensing. Commonly used geographical data includes boundary coordinates of farmland, terrain undulation, slope, land cover type, etc., with the accuracy generally reaching within 1 meter. High-resolution remote sensing images (such as satellite images with a resolution of 30 cm) are combined with ground measurement data (such as GNSS systems) to supplement the geographical data of farmland, ensuring the accuracy of the geographical data. A high-resolution camera (such as a digital camera with a resolution of 20 MP or above) is used to photograph the farmland environment. The camera needs to be equipped with a good optical zoom function to adapt to different weather and lighting conditions. By setting multiple cameras to capture images of the farmland environment from different angles to ensure full coverage, the shooting angle of the images is required to cover at least 30 degrees in each direction for subsequent feature point detection and matching. A unified coordinate reference system (such as WGS84 or UTM coordinate system) is adopted to convert and standardize the obtained farmland geographical data. The specific operation is to convert coordinate data from various different sources into a standard coordinate system through a projection transformation algorithm to ensure data consistency. If the farmland coverage area is large, a regional coordinate system (such as the geodetic coordinate system in geographical information standards) is considered to improve the accuracy and processing efficiency of geographical data. The data accuracy target reaches 1 - 2 meters. The standardized farmland geographical data includes information such as land boundaries, terrain elevations, crop distributions, soil types, etc., and these data will serve as the basic data for subsequent 3D reconstruction and environmental analysis. Feature point detection is performed on the farmland environment images using computer vision algorithms such as SIFT (Scale-Invariant Feature Transform), SURF (Speeded-Up Robust Features), or ORB (Oriented FAST and Rotated BRIEF). These algorithms can extract feature points with unique descriptors from the images and adapt to different lighting, viewing, and scale changes. Through a feature point matching algorithm (such as a descriptor-based matching method), the feature points of images from different perspectives are matched to generate a set of scene feature points of the farmland environment from different perspectives. Based on the matching results, key scene feature points in the farmland images (such as farmland roads, ridges, crop distributions, etc.) are extracted. The extraction accuracy of these feature points directly affects the effect of subsequent 3D point cloud reconstruction. Robust algorithms such as RANSAC (Random Sample Consensus algorithm) are used to further filter out mis-matched feature points to ensure that the finally extracted scene feature points have high accuracy. The standardized farmland geographical data is combined with the extracted scene feature points, and the multi-view stereo vision (MVS) algorithm or structured light technology is used to perform 3D reconstruction on the farmland environment. During the reconstruction process, the scene feature points in the farmland images are combined with the elevation information in the geographical data to generate the initial point cloud data of the farmland. This point cloud data includes the spatial coordinates (x, y, z) of each point, and the color and density attributes of the point cloud are generated according to the lighting and texture information of the images.The target point cloud density is 30 - 50 points per square meter to ensure that the reconstructed point cloud data has sufficient details and accuracy. Noise filtering and data sparsification are performed on the generated initial point cloud data to remove unnecessary noise points. Filters (such as median filtering, statistical filtering) are used to remove abnormal points caused by environmental factors (such as clouds, light reflections, etc.). Three-dimensional reconstruction software (such as MeshLab, CloudCompare, etc.) is used to perform surface modeling on the initial point cloud data of the farmland to generate a three-dimensional surface model of the farmland. Through surface reconstruction algorithms (such as Poisson surface reconstruction, Delaunay triangulation), a fine model of the farmland terrain is constructed. During this process, combined with the geographical characteristics of the farmland (such as ridges, roads, undulations, etc.), the boundaries and details of the model are optimized to make it as close as possible to the actual farmland environment. The accuracy requirement of the three-dimensional surface model reaches 5 - 10 cm to ensure its usability and accuracy in the digital twin system. Based on the three-dimensional surface modeling, a digital twin farmland system is generated. The digital twin farmland not only contains the three-dimensional geometric information of the farmland but also includes data such as crop types, soil information, and water source distribution, forming a highly restored virtual farmland model.

[0095] Preferably, step S15 includes the following steps:

[0096] Step S151: Perform point cloud denoising on the initial point cloud data of the farmland to generate purified point cloud data;

[0097] Step S152: Based on the purified point cloud data, perform point cloud surface reconstruction to generate three-dimensional surface data of the farmland;

[0098] Step S153: Extract the texture features of the farmland environment image and perform surface texture mapping on the three-dimensional surface data of the farmland to generate three-dimensional surface mapping data of the farmland;

[0099] Step S154: Integrate the three-dimensional surface mapping data of the farmland into a virtual scene to generate a digital twin farmland.

[0100] In the embodiments of the present invention, the initial point cloud data of the farmland is denoised by using point cloud denoising algorithms (such as statistical outlier removal, Voxel Grid filtering, Bilateral filtering, etc.). Point cloud data usually contains noise points, such as invalid data generated due to sensor errors, environmental interference, etc. The specific steps include: detecting and removing abnormal points by analyzing the local density of each point's neighborhood to ensure that the denoised point cloud data is more accurate. The denoising accuracy target is to remove more than 90% of the noise points and retain about 80%-90% of the valid point cloud. After denoising, the obtained purified point cloud data is more uniform in spatial distribution, and the point cloud density is generally controlled at about 40 points per square meter to ensure its accuracy in subsequent processing. After completing the denoising, the data is subjected to quality inspection to ensure that the purified point cloud data meets the requirements in terms of accuracy. The spatial distribution of the point cloud should be more uniform, and the accuracy should be maintained within 5 cm. The surface reconstruction algorithm (such as Poisson reconstruction, Delaunay triangulation reconstruction, Alpha shape, etc.) is used to perform surface reconstruction on the purified point cloud data. Through these algorithms, a three-dimensional surface model of the farmland is generated based on the spatial coordinates of the point cloud data. The model includes features such as the terrain of the farmland, undulations, and crop distribution. During the reconstruction process, the accuracy and integrity of the three-dimensional surface model are ensured, and the target accuracy is within 10 cm. To ensure the expression of model details, the target grid density after surface reconstruction is 10,000 triangular patches per square meter. Based on the surface reconstruction, the generated surface data is smoothed to eliminate the rough surface caused by sparse or uneven point cloud data. The grid smoothing algorithm (such as Laplacian smoothing or Taubin smoothing) is used to adjust the smoothness of the model while retaining the real features of the farmland terrain. The optimized surface model should have a good visual effect and be able to run stably in subsequent virtual scene integration. The texture features of the farmland environment image are extracted, and texture analysis algorithms (such as Gabor filtering, LBP (Local Binary Pattern), etc.) are used to extract the texture features in the image. Through the texture analysis algorithm, different texture regions in the farmland image are identified, such as ridges, crop regions, roads, etc. The extracted texture features are mapped onto the reconstructed three-dimensional surface data, and texture mapping algorithms (such as UV mapping, environment mapping, etc.) are used to attach the texture information of the farmland image to the three-dimensional surface. Through the texture mapping technology, the color, illumination, and details of the farmland image are accurately transferred to the three-dimensional surface model, enhancing the realism and visual effect of the farmland three-dimensional model. Finally, the generated farmland three-dimensional surface mapping data should contain the geometric data and surface texture data of the three-dimensional model. The farmland three-dimensional surface mapping data is integrated with other relevant data (such as soil data, moisture data, crop distribution data, etc.) to generate a complete farmland virtual scene. During the integration process, virtual reality technology and 3D visualization technology are used to synchronize the three-dimensional model with real-time data so that it can dynamically reflect the changes in the farmland environment.Finally, by integrating the three-dimensional surface model, environmental texture data, real-time sensor data, etc., a complete digital twin farmland system is generated.

[0101] As an example of the present invention, refer to Figure 3 As shown, in this example, step S2 includes:

[0102] Step S21: Obtain the data of the seeds to be sown;

[0103] Step S22: Extract the seed type, variety, and growth cycle of the seeds to be sown, classify the seeds to be sown based on the data, and generate seed classification data;

[0104] Step S23: Analyze the development characteristics of the seeds to be sown based on the seed classification data to obtain the development characteristic data of the seeds to be sown;

[0105] Step S24: Perform time series analysis on the development characteristic data of the seeds to be sown, extract the development parameters at different growth stages, and generate stage development characteristic data; import the stage development characteristic data into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data.

[0106] In the embodiments of the present invention, machine learning models (such as K-means clustering, decision trees, support vector machines, etc.) are used to classify the data of seeds to be sown, and the types, varieties of seeds, and the growth cycles of each type of seed are extracted. According to the growth cycles of different seeds, the seeds can be classified into early-maturing varieties, mid-maturing varieties, late-maturing varieties, etc. The growth cycle of each type of seed ranges from 60 days to 150 days, and the classification is based on the growth habits of different crops. Based on the above-extracted data, a seed classification data is generated, including the characteristics and growth cycle information of each type of seed. The data table records the classification labels, seed types, and corresponding growth cycles of each variety. Specifically, the growth cycle of some wheat varieties is 100 days, while the growth cycle of some corn varieties is 120 days. The generated seed classification data will provide a basis for the subsequent analysis of developmental characteristics. Standard models in the agricultural field (such as plant growth models, photosynthesis models, etc.) are used to analyze the developmental characteristics of different seeds. By analyzing the effects of environmental factors such as growth cycle, temperature, light, and moisture, the developmental speed, germination rate, emergence period, and other characteristics of each type of seed are determined. The developmental characteristics of different types and varieties of seeds are analyzed, and relevant data are extracted, such as the germination temperature range, optimal sowing depth, suitable soil moisture, etc. For each variety of seeds, according to its growth habits, developmental characteristic data are generated. The data table includes: the seed developmental characteristics of each variety (such as germination rate, emergence period, optimal growth environment, disease resistance, etc.) and the key developmental nodes in its growth process (such as heading, flowering, maturity, etc.). Specifically, the germination rate of a certain wheat variety reaches 90% under the condition of 25°C, while the corn variety requires the condition of 20 - 30°C to ensure a better germination rate and growth speed. Time series analysis methods (such as ARIMA models, periodic analysis, etc.) are used to perform time series modeling based on the developmental characteristic data to predict the developmental parameters at each growth stage. By analyzing the growth conditions of seeds at different time periods (such as 15 days, 30 days, 60 days after sowing, etc.), the developmental parameters at different growth stages are extracted, such as root length, stem height, leaf area index (LAI), chlorophyll content, etc. Based on the time series analysis results, the developmental characteristic data at different growth stages are generated. Specifically, the root system of some crops develops faster at 30 days, while the leaf growth speed is faster at 60 days. The generated stage-based developmental characteristic data are imported into the digital twin farmland system to ensure the accuracy and precision of the simulation. The digital twin farmland system can simulate the growth process of crops according to the developmental characteristics of the crops. This data includes the developmental parameters at each stage. Specifically, 30 days after sowing, the roots and stems of some crops reach a certain length, and they enter the flowering stage at 60 days. Sowing simulations are run in the digital twin farmland to simulate the effects of different sowing methods and times on crop growth. Through the sowing simulation data, the effects of the optimal sowing density, sowing depth, soil temperature and humidity, etc. on crop growth are analyzed. Based on the growth simulation data, the growth results of crops under different environmental conditions can be predicted.By dynamically adjusting seeding parameters (such as seeding density, planting row spacing, etc.), optimizing the farmland operation process, and increasing crop yields.

[0107] Preferably, the importing of the stage development characteristic data into the digital twin farmland for seeding simulation and growth simulation in step S24 includes:

[0108] Importing the stage development characteristic data into the digital twin farmland for seeding simulation parameter setting to obtain seeding simulation setting parameters: the seeding depth is set to 3 cm, the seed density is 100 seeds per square meter, the seed germination rate is 85%, the soil temperature is set to 20 °C, the soil humidity is 0.22 cm 3 / cm 3 , the initial growth rate of the seeds is 1.2 mm per day, the temperature adaptability range is 18 °C to 30 °C, the water demand is 5 mm per day, the light demand is 12 hours per day, the nutrient demand is 10 kg / ha of nitrogen, 5 kg / ha of phosphorus, and 8 kg / ha of potassium, the crop growth cycle is 120 days, the root growth rate is 0.8 cm per day, and the crop stress resistance score is 4 points;

[0109] Importing the stage development characteristic data into the digital twin farmland for growth simulation parameter setting to obtain growth simulation setting parameters: the initial growth rate is 1.5 mm per day, the nutrient demand is 15 kg / ha of nitrogen, 10 kg / ha of phosphorus, and 12 kg / ha of potassium, the maximum leaf area index is 3.5, the photosynthetically active radiation demand is 15 MJ / m 2 / day, the water evaporation rate is set to 4 mm per day, the root expansion rate is 2 cm per day, the growth cycle temperature requirement is between 15 °C and 30 °C, the growth stage distribution is 30 days of the initial growth period, 60 days of the rapid growth period, 30 days of the maturity period, the light response coefficient is 0.9, the pest and disease impact is 0.05, the environmental humidity requirement is 70% to 85%, and the crop maturity period is 90 days;

[0110] Using agricultural simulation software to perform simulation analysis on the seeding simulation setting parameters and the growth simulation setting parameters to generate seeding simulation data and growth simulation data.

[0111] Preferably, the seeding profile analysis of the seeding simulation data in step S3 includes:

[0112] Extracting the seed position information and soil temperature information of the seeding simulation data;

[0113] Analyzing the seeding simulation depth of the seeding simulation data based on the seed position information to obtain seed distribution data at different depths;

[0114] Extract the soil temperature and soil humidity from the soil temperature information, and perform soil temperature interpolation and soil humidity interpolation on the soil temperature and soil humidity respectively to generate continuous soil temperature profile data and continuous soil humidity profile data;

[0115] Visualize the soil temperature profile data and the soil humidity profile data according to the seed distribution data to generate a seeding profile diagram.

[0116] In the embodiments of the present invention, the position information of seeds is extracted from the sowing simulation data, including the longitude and latitude coordinates and depth information of each seed. The seed position information is usually recorded by the GPS device of the seeder, and the data format includes the longitude, latitude of the seed position, and the sowing depth of each seed. Specifically, the seed position is stored as "Seed 1: longitude 30.1234, latitude 45.6789, depth 2 cm", and the specific information of each seed is obtained in this way. The soil temperature information in the sowing simulation data is extracted, and this data can be obtained in real time through a sensor network. The soil temperature information includes the soil temperature at different depths, and usually records the soil temperature values at different time points and depths. Specifically, in the area where the soil depth is 0 - 10 cm, the soil temperature is 22 °C, while at a depth of 10 - 20 cm, the soil temperature is 18 °C. Through these data, the distribution of the soil temperature can be obtained. Based on the seed position information, the sowing depth of each seed is calculated. By counting the depth distribution of all seeds, the seed density at each depth level is analyzed. Specifically, the sowing depth can be stratified, and the number of seeds in depth segments such as 0 - 5 cm, 5 - 10 cm, 10 - 15 cm, etc. is calculated to obtain the seed distribution data at each depth level. Based on the analysis of the sowing depth, seed distribution data is generated. The data format is the number of seeds and their density in each depth segment. Specifically, the seed density in the 0 - 5 cm depth segment is 100 seeds per square meter, and the seed density in the 5 - 10 cm depth segment is 90 seeds per square meter. The extracted soil temperature data is interpolated using common interpolation methods (such as Kriging interpolation, bilinear interpolation, etc.) to generate continuous soil temperature profile data, which will help generate a smoother and more continuous soil temperature curve. Specifically, if the interval of soil temperature data is large at certain positions, continuous temperature data at this position can be generated through the interpolation method. The interpolated data can show the temperature change within the entire soil depth range. Specifically: the temperature at a depth of 0 - 10 cm is 22 °C, 10 - 20 cm is 20 °C, and 20 - 30 cm is 18 °C. Similarly, the extracted soil humidity data is interpolated to generate continuous soil humidity profile data. The interpolation method can be selected as Kriging interpolation or spline interpolation, etc., to ensure the continuity of the humidity data. Specifically, the soil humidity data shows that the humidity at a depth of 0 - 10 cm is 20%, and the humidity at a depth of 10 - 20 cm is 18%. Continuous humidity data is generated through interpolation, such as 20% for 0 - 10 cm, 19.5% for 10 - 20 cm, and 18.5% for 20 - 30 cm. The seed distribution data is integrated with the soil temperature profile data and the soil humidity profile data, and visual processing is performed according to different sowing depth levels. Through 3D modeling tools, the seed distribution, soil temperature, and humidity at different depths are intuitively displayed. A heat map or 3D solid graph is used to show the changes in soil temperature and humidity, and the distribution levels of seeds are marked on the graph.Specifically, it can display the seed distribution, soil temperature, and humidity changes in different depth segments such as 0 - 10 cm and 10 - 20 cm. Finally, a seeding profile diagram is generated to show the seed distribution at different depths, and at the same time, the corresponding soil temperature and humidity information is displayed.

[0117] Preferably, the imbalance of the root - shoot ratio in analyzing the seeding profile diagram according to the growth simulation data in step S3 includes:

[0118] Extract the morphological characteristics of the seeded crop from the growth simulation data, and calculate the growth rate of the growth simulation data according to the morphological characteristics to obtain dynamic growth data;

[0119] Based on the dynamic growth data, conduct growth stratification of the seeded crop for the growth simulation data to obtain upper - layer growth data and lower - layer growth data; conduct an analysis of the root spatial structure of the seeded crop for the lower - layer growth data to generate root characteristic data;

[0120] Conduct biomass calculation for the upper - layer growth data to obtain canopy characteristic data;

[0121] Calculate the ratio through the root characteristic data and the canopy characteristic data to obtain the root - shoot ratio of the crop;

[0122] Evaluate the imbalance of the root - shoot ratio of the crop according to the seeding profile diagram to generate the analysis result of the root - shoot ratio imbalance.

[0123] In the embodiments of the present invention, by extracting the morphological characteristics of the sown crops from the growth simulation data, including data such as the dimensions, volumes, distributions, etc. of parts such as the stems, leaves, and roots of the crops. The morphological characteristics usually come from crop growth models or dynamic monitoring through sensors (such as image recognition). Specifically, data such as crop height, leaf area index (LAI), root length, etc. can be extracted through imaging techniques. Calculate the growth rate of the crops based on the extracted morphological characteristics. Usually, time series data is used to calculate the growth change rate of the crops within a certain period of time. The growth rate can be deduced based on the height change or dry matter accumulation of the crops. Specifically, by measuring the crop height change every week, calculate the growth rate per week to obtain dynamic growth data. Specifically, if the crop height increases by 5 cm in the first week and 7 cm in the second week, the growth rate is 2 cm / week. Based on the dynamic growth data, divide the growth of the crops into upper and lower layers according to different growth stages of the crops. The upper layer usually refers to the canopy part of the crops (such as leaves and branches, etc.), and the lower layer refers to the root system and the surrounding soil area. Set the stratification criteria according to the characteristics of different crops. Specifically, for corn crops, define the canopy as the part from the ground to 1.5 meters, and the root system as the part from the soil surface to 1 meter deep. Extract the growth data of the upper and lower layers from the growth simulation data. The upper layer growth data includes the height, leaf area, number of branches, etc. of the canopy, while the lower layer growth data includes the depth, root length, root density, etc. of the root system. Specifically, assume that the upper layer leaf area of the crop is 2 square meters, the root system depth is 80 cm, and the root density is 0.4 g / cm³. Conduct a spatial structure analysis on the root system information in the lower layer growth data. The analysis content includes indicators such as the distribution range, root system depth, number of roots, etc. of the root system, as well as the three-dimensional spatial distribution of the root system. Use 3D modeling technology or computer vision technology to analyze the spatial structure of the root system and extract the geometric characteristics of the root system. The root system structure analysis helps to understand the growth potential of the root system and its interaction with the soil environment. Through the analysis, generate root system characteristic data, including information such as the density, depth, distribution range, etc. of the root system. Specifically, the depth of the root system in the soil is 80 cm, the total length of the root system is 10 meters, and the root density is 0.6 g / cm³. Based on the upper layer growth data (such as leaf area, branch length, etc.), use a biomass calculation model to calculate the total biomass of the canopy. Biomass is usually deduced by measuring indicators such as leaf area and dry matter accumulation. Specifically, by calculating data such as the leaf area index (LAI) and photosynthetic efficiency, obtain the biomass data of the canopy, such as the canopy biomass is 3 kg / m². Based on the biomass calculation result, generate canopy characteristic data. The canopy characteristic data can include the leaf area, canopy height, photosynthetic efficiency, etc. of the crops. Specifically, the average leaf area of the canopy is 2 square meters, and the canopy height is 1.5 meters. Based on the root system characteristic data and the canopy characteristic data, calculate the root-shoot ratio. The root-shoot ratio is usually the ratio of the root system biomass to the canopy biomass, reflecting the growth ratio of the root system and the canopy of the crops.Specifically, if the root biomass is 1.5 kg / m² and the canopy biomass is 3 kg / m², then the root-shoot ratio is 0.5. An imbalance assessment of the root-shoot ratio is carried out to analyze whether the root-shoot ratio of the crop conforms to the ideal growth ratio. Imbalance usually refers to a situation where the root system is too weak or the canopy is too large, resulting in problems in the absorption and growth of water and nutrients by the crop. According to the analysis results of the sowing profile, it is judged whether the root-shoot ratio of the crop is unreasonable, such as limited root growth due to too deep or too shallow sowing depth, or uneven soil temperature and humidity affecting root growth. Based on the analysis, an analysis report on the imbalance of the root-shoot ratio is generated, including the degree of imbalance, cause analysis, and optimization suggestions.

[0124] Preferably, the adjustment of the sowing simulation data based on the analysis result of the root-shoot ratio imbalance in step S3 includes:

[0125] Compare the analysis result of the root-shoot ratio imbalance with a preset root-shoot ratio imbalance threshold. When the analysis result of the root-shoot ratio imbalance is greater than or equal to the root-shoot ratio imbalance threshold, abnormal seed sowing data is generated;

[0126] Analyze the abnormal factors of the abnormal seed sowing data to obtain the abnormal seed sowing factors, where the abnormal seed sowing factors are at least one of the factors such as abnormal seed density, abnormal sowing depth, abnormal nutrient supply amount, and abnormal water.

[0127] Adjust the sowing simulation data through the abnormal seed sowing factors:

[0128] When it is confirmed that the abnormal seed sowing factor is abnormal seed density, the seed density in the sowing simulation data is increased or decreased by 10%-20%; when it is confirmed that the abnormal seed sowing factor is abnormal sowing depth, the sowing depth in the sowing simulation data is increased or decreased by 2-5 cm; when it is confirmed that the abnormal seed sowing factor is abnormal nutrient supply amount, the nitrogen fertilizer supply amount in the sowing simulation data is increased by 10%-15% kg / ha, the phosphorus fertilizer supply amount is decreased by 5%-10% kg / ha, and the potassium fertilizer supply amount is increased by 10%-15% kg / ha; when it is confirmed that the abnormal seed sowing factor is abnormal water, the irrigation amount in the sowing simulation data is increased or decreased by 10%-20%, and the sowing optimization simulation data is integrated.

[0129] In the embodiments of the present invention, by presetting a threshold for root-shoot ratio imbalance, which is usually set based on historical data and crop growth laws, this threshold can be obtained through experiments or adjusted according to the growth characteristics of different crops. Specifically, the threshold can be set such that when the root-shoot ratio imbalance is greater than 0.8, it is considered abnormal. Compare the root-shoot ratio imbalance analysis result obtained in step S35 with the preset root-shoot ratio imbalance threshold. If the analysis result of the root-shoot ratio imbalance is greater than or equal to the threshold, it indicates that there is a root-shoot ratio imbalance problem during the sowing process and adjustments must be made. Specifically, if the preset threshold is 0.7 and the analysis result is 0.75, abnormal seed sowing data is generated. Once the analysis result of the root-shoot ratio imbalance exceeds the threshold, it is considered that there is an abnormality in the sowing process, and the system will generate abnormal seed sowing data for subsequent adjustment and optimization. Analyze the generated abnormal seed sowing data to identify the abnormal factors that cause the root-shoot ratio imbalance. The main abnormal factors include: Abnormal seed density: The sowing density of seeds is too large or too small. Abnormal sowing depth: The sowing depth of seeds is too deep or too shallow. Abnormal nutrient supply: The soil nutrient supply (such as nitrogen, phosphorus, and potassium fertilizers) is insufficient or excessive. Abnormal water: The irrigation water volume is too much or too little. Based on the abnormal seed sowing data, the system automatically judges and determines the abnormal factors. Specifically, if it is found that the sowing density is too large, it is marked as "abnormal seed density"; if it is detected that the soil nutrient supply is insufficient or excessive, it is marked as "abnormal nutrient supply". Through analysis, the abnormal factors that cause the root-shoot ratio imbalance are finally obtained, specifically: "abnormal seed density", "abnormal sowing depth", "abnormal nutrient supply", or "abnormal water". According to the determined abnormal seed sowing factors, adjust the sowing simulation data. The adjustment methods are as follows: If the abnormal factor is abnormal seed density, adjust the sowing density. The proportion of increasing or decreasing the seed density can be set within the range of 10%-20%. The specific adjustment can be achieved by increasing or decreasing the number of seeds per hectare. Specifically, if the current sowing density is 100,000 plants per hectare, the sowing density after increasing by 10% is 110,000 plants per hectare, and the sowing density after decreasing by 10% is 90,000 plants per hectare. If the abnormal factor is abnormal sowing depth, adjust the sowing depth. The range of increasing or decreasing the sowing depth is 2-5 cm. The adjustment method can be determined according to the soil type, crop requirements, and climatic conditions. If the current sowing depth is 10 cm, the sowing depth after increasing by 5 cm is 15 cm, and the sowing depth after decreasing by 3 cm is 7 cm. If the abnormal factor is abnormal nutrient supply, adjust the soil nutrients (nitrogen, phosphorus, and potassium fertilizers). Increase the nitrogen fertilizer supply by 10%-15% kg / ha, decrease the phosphorus fertilizer supply by 5%-10% kg / ha, and increase the potassium fertilizer supply by 10%-15% kg / ha.Specifically, if the current nitrogen fertilizer supply is 50 kg / ha, the nitrogen fertilizer supply after a 15% increase is 57.5 kg / ha; if the current phosphorus fertilizer supply is 20 kg / ha, the phosphorus fertilizer supply after a 5% reduction is 19 kg / ha; if the current potassium fertilizer supply is 40 kg / ha, the potassium fertilizer supply after a 10% increase is 44 kg / ha. If the abnormal factor is water anomaly, adjust the irrigation amount. The proportion of increasing or decreasing the irrigation amount can be set within the range of 10% - 20%. Specifically, if the current irrigation amount is 500 mm, the irrigation amount after a 10% increase is 550 mm, and the irrigation amount after a 10% reduction is 450 mm. According to the above adjustments, the sowing optimization simulation data is integrated, and these data will be used as a new sowing simulation plan for subsequent sowing operations and growth monitoring. Integrate the adjusted data of seed density, sowing depth, nutrient supply, and irrigation amount into the optimized sowing simulation data to provide accurate guidance for actual sowing.

[0130] Preferably, the real-time sowing position correction of the sowing operation data in step S4 includes:

[0131] Import the sowing optimization simulation data into the digital twin farmland for sowing path analysis to generate the theoretical sowing position;

[0132] Perform real-time positioning processing of the sowing machine on the sowing operation data to generate absolute coordinate data;

[0133] Perform attitude calculation on the sowing operation data according to the absolute coordinate data to generate the sowing position data of the sowing machine;

[0134] Compare the sowing position data of the sowing machine with the theoretical sowing position to generate spacing deviation data;

[0135] Use the spacing deviation data to perform sowing position motion compensation on the sowing operation data to execute intelligent agricultural sowing management operations.

[0136] In the embodiments of the present invention, adjusted and optimized seeding simulation data (such as seeding density, seeding depth, nutrient and moisture adjustment, etc.) are imported into the digital twin farmland system. The digital twin farmland is a virtual environment that can simulate the operation conditions of real farmland. Based on the environmental data in the digital twin farmland (such as terrain, soil type, crop type, etc.), the system automatically conducts seeding path analysis. This analysis takes into account the actual situation of the farmland and generates an ideal seeding path and the theoretical seeding position of each seed. Through the seeding path analysis, the system generates the theoretical positions where each seed should be sown. These positions serve as reference points in the seeding operation for subsequent correction of the seeding machine's position. Positioning technologies such as the GPS system and inertial measurement unit (IMU) are used to real-time locate the seeding machine. Through the GPS and sensor systems, the real-time position of the seeding machine in the farmland is obtained. After calculation of the real-time positioning information of the seeding machine, absolute coordinate data of the seeding machine are generated (usually based on longitude, latitude, and altitude). This data represents the actual position of the seeding machine in the farmland. Attitude solution refers to determining the specific orientation and posture of the seeding machine during operation by analyzing the attitude of the seeding machine (such as pitch angle, roll angle, yaw angle). This step is usually calculated using the IMU sensor and related algorithms (such as Kalman filtering, etc.). Based on the absolute coordinates of the seeding machine and the result of attitude solution, precise seeding position data of the seeding machine are calculated. These data include the specific position of the seeding machine at each time point and its travel trajectory in the farmland. Compare the actual seeding position of the seeding machine (the seeding position data of the seeding machine generated in step S43) with the theoretical seeding position (the theoretical seeding position generated in step S41). The theoretical seeding position is the ideal position obtained based on seeding optimization simulation data and path analysis in the digital twin farmland, while the actual seeding position of the seeding machine is real-time obtained by the positioning system. According to the difference between the actual seeding position and the theoretical seeding position of the seeding machine, the deviation degree of the seeding position is calculated (i.e., the distance or angle difference between the actual seeding position and the theoretical position of the seeding machine), which can be calculated using methods such as Euclidean distance, Manhattan distance, or angle difference. The seeding position deviation degree is converted into spacing deviation data, which can reveal the displacement error or inaccuracy that occurs during the seeding process. Specifically, if the deviation between the seeding machine and the theoretical position exceeds the preset range, it indicates that there are problems in the seeding operation, affecting crop uniformity. Based on the spacing deviation data, the seeding operation data are dynamically adjusted. By compensating for the motion error of the seeding machine, the travel path and seeding position of the seeding machine are adjusted. The compensation methods include: correcting the forward direction or speed of the seeding machine according to the deviation data to ensure that the seeding machine travels along the predetermined seeding path. By adjusting the control system of the seeding machine, the seeding position is corrected in real-time to ensure that the seeding position of each seed meets the theoretical requirements. The compensated seeding operation data are used to perform subsequent seeding management tasks.Through the intelligent agriculture system, precise seeding is implemented to improve the uniformity of crop growth and optimize the allocation of resources (such as water, nutrients, seed density, etc.). The adjustment effect of the seeding operation is fed back in real time, and the seeding operation path and strategy are continuously optimized according to the actual results of each seeding, forming a closed-loop feedback mechanism for intelligent agriculture.

[0137] Therefore, in whatever aspects, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Thus, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be embraced within the present invention.

[0138] The above are only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A smart agriculture management method based on the Internet of Things, characterized in that, It includes the following steps: Step S1: Obtain farmland geographical data and farmland environment images; extract the scene feature points of the farmland environment images, and combine the farmland geographical data to perform 3D modeling of the farmland to generate a digital twin farmland; Step S2: Obtain the data of the seeds to be sown; analyze the growth characteristics of the seeds to be sown to obtain the growth characteristic data of the seeds to be sown; import the growth characteristic data of the seeds to be sown into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data; Step S3: Conduct a sowing profile analysis on the sowing simulation data to obtain a sowing profile diagram; analyze the root-shoot ratio imbalance of the sowing profile diagram based on the growth simulation data, and adjust the sowing simulation data based on the analysis result of the root-shoot ratio imbalance to obtain optimized sowing simulation data; Step S4: Use the Internet of Things to import the optimized sowing simulation data into an intelligent seeder for sowing operation execution, and simultaneously collect sowing operation data; perform real-time sowing position correction on the sowing operation data to execute intelligent agricultural sowing management operations.

2. The intelligent agriculture management method based on the Internet of Things according to claim 1, wherein, Step S1 includes the following steps: Step S11: Obtain farmland geographical data through GIS technology; obtain farmland environment images through a camera; Step S12: Standardize the coordinates of the farmland geographical data to generate standardized farmland geographical data; Step S13: Detect and match the feature points of the farmland environment image, and extract the scene feature points of the farmland environment image to obtain scene feature points; Step S14: Combine the standardized farmland geographical data and the scene feature points to perform farmland point cloud reconstruction to generate initial farmland point cloud data; Step S15: Perform 3D surface modeling on the initial farmland point cloud data to generate a digital twin farmland.

3. The smart agriculture management method based on the Internet of Things according to claim 2, characterized in that, Step S15 includes the following steps: Step S151: Denoise the initial farmland point cloud data to generate purified point cloud data; Step S152: Perform point cloud surface reconstruction based on the purified point cloud data to generate 3D surface data of the farmland; Step S153: Extract the texture features of the farmland environment image, and perform surface texture mapping on the 3D surface data of the farmland to generate 3D surface mapping data of the farmland; Step S154: Integrate the 3D surface mapping data of the farmland into a virtual scene to generate a digital twin farmland.

4. The intelligent agriculture management method based on the Internet of Things according to claim 1, characterized in that Step S2 includes the following steps: Step S21: Obtain the data of the seeds to be sown; Step S22: Extract the seed type, variety, and growth cycle of the seeds to be sown to classify the seeds to be sown and generate seed classification data; Step S23: Analyze the growth characteristics of the seeds to be sown based on the seed classification data to obtain the growth characteristic data of the seeds to be sown; Step S24: Perform time series analysis on the growth characteristic data of the seeds to be sown, extract the growth parameters at different growth stages to generate stage growth characteristic data; import the stage growth characteristic data into the digital twin farmland for sowing simulation and growth simulation to generate sowing simulation data and growth simulation data.

5. The intelligent agriculture management method based on the Internet of Things according to claim 4, wherein, The import of the stage growth characteristic data into the digital twin farmland for sowing simulation and growth simulation in Step S24 includes: Import the data of stage development characteristics into the digital twin farmland for setting sowing simulation parameters, and obtain the sowing simulation setting parameters: the sowing depth is set to 3 cm, the seed density is 100 seeds per square meter, the seed germination rate is 85%, the soil temperature is set to 20 °C, and the soil humidity is 0.22 cm 3 / cm 3 , the growth rate at the initial stage of seed is 1.2 mm per day, the temperature adaptability range is 18 °C to 30 °C, the water demand is 5 mm per day, the light demand is 12 hours per day, the nutrient demand is 10 kg / ha of nitrogen, 5 kg / ha of phosphorus, and 8 kg / ha of potassium, the crop growth cycle is 120 days, the root growth rate is 0.8 cm per day, and the crop stress resistance score is 4 points; Import the data of stage development characteristics into the digital twin farmland for growth simulation parameter setting, and obtain the growth simulation setting parameters: the initial growth rate is 1.5 mm / day, the nutrient requirements are 15 kg / ha of nitrogen, 10 kg / ha of phosphorus, and 12 kg / ha of potassium, the maximum leaf area index is 3.5, the photosynthetically active radiation requirement is 15 MJ / m 2 / day, the water evaporation rate is set to 4 mm / day, the root expansion rate is 2 cm / day, the growth period temperature requirement is between 15°C and 30°C, the growth stage distribution is 30 days for the initial growth period, 60 days for the rapid growth period, 30 days for the maturity period, the light reaction coefficient is 0.9, the pest and disease impact is 0.05, the environmental humidity requirement is 70% to 85%, and the crop maturity period is 90 days; Use agricultural simulation software to conduct simulation analysis on the parameters of seeding simulation and growth simulation, and generate seeding simulation data and growth simulation data.

6. The intelligent agriculture management method based on the Internet of Things according to claim 1, characterized in that, The seeding profile analysis of the seeding simulation data described in step S3 includes: Extract the seed position information and soil temperature information of the seeding simulation data; Analyze the seeding simulation depth of the seeding simulation data based on the seed position information to obtain the seed distribution data at different depths; Extract the soil temperature and soil humidity from the soil temperature information, and perform soil temperature interpolation and soil humidity interpolation on the soil temperature and soil humidity respectively to generate continuous soil temperature profile data and continuous soil humidity profile data; Perform visualization processing on the soil temperature profile data and soil humidity profile data according to the seed distribution data to generate a seeding profile diagram.

7. The intelligent agriculture management method based on the Internet of Things according to claim 1, characterized in that The analysis of root-shoot ratio imbalance of the seeding profile diagram according to the growth simulation data described in step S3 includes: Extract the morphological characteristics of the seeded crops in the growth simulation data, and calculate the growth rate of the growth simulation data according to the morphological characteristics to obtain dynamic growth data; Based on the dynamic growth data, conduct growth stratification of the seeded crops in the growth simulation data to obtain upper-layer growth data and lower-layer growth data; analyze the root spatial structure of the seeded crops for the lower-layer growth data to generate root characteristic data; Calculate the biomass of the upper-layer growth data to obtain canopy characteristic data; Calculate the ratio through the root characteristic data and the canopy characteristic data to obtain the root-shoot ratio of the crops; Evaluate the imbalance of the root-shoot ratio of the crops according to the seeding profile diagram to generate the analysis result of root-shoot ratio imbalance.

8. The method for intelligent agriculture management based on the Internet of Things according to claim 1, characterized in that, The adjustment of the seeding simulation data based on the analysis result of root-shoot ratio imbalance described in step S3 includes: Compare the analysis result of root-shoot ratio imbalance with the preset root-shoot ratio imbalance threshold. When the analysis result of root-shoot ratio imbalance is greater than or equal to the root-shoot ratio imbalance threshold, abnormal seed seeding data is generated; Analyze the abnormal factors of the abnormal seed seeding data to obtain the abnormal seed seeding factors, where the abnormal seed seeding factors are at least one of abnormal seed density, abnormal seeding depth, abnormal nutrient supply amount, and abnormal water; Adjust the seeding simulation data through the abnormal seed seeding factors: When it is confirmed that the abnormal seed seeding factor is abnormal seed density, increase / decrease the seed density of the seeding simulation data by 10%-20%; when it is confirmed that the abnormal seed seeding factor is abnormal seeding depth, increase / decrease the seeding depth of the seeding simulation data by 2-5 cm; when it is confirmed that the abnormal seed seeding factor is abnormal nutrient supply amount, increase the nitrogen fertilizer supply amount of the seeding simulation data by 10%-15% kg / ha, reduce the phosphorus fertilizer supply amount by 5%-10% kg / ha, and increase the potassium fertilizer supply amount by 10%-15% kg / ha; when it is confirmed that the abnormal seed seeding factor is abnormal water, increase / decrease the irrigation amount by 10%-20%, and integrate to obtain the optimized seeding simulation data.

9. The intelligent agriculture management method based on the Internet of Things according to claim 1, characterized in that, The real-time seeding position correction of the seeding operation data described in step S4 includes: Import the optimized seeding simulation data into the digital twin farmland for seeding path analysis to generate the theoretical seeding position; Perform real-time positioning processing on the seeding operation data to generate absolute coordinate data; Perform attitude calculation on the seeding operation data according to the absolute coordinate data to generate seeding position data of the seeder; Compare the seeding position data of the seeder with the theoretical seeding position to generate spacing deviation data; Use the spacing deviation data to perform seeding position motion compensation on the seeding operation data to execute intelligent agricultural seeding management operations.

10. A smart agriculture management system based on the Internet of Things, characterized in that, For performing the Internet of Things-based intelligent agriculture management method as described in claim 1, the Internet of Things-based intelligent agriculture management system includes: A three-dimensional modeling module, configured to obtain farmland geographical data and farmland environment images; extract scene feature points of the farmland environment images, and perform three-dimensional modeling of the farmland in combination with the farmland geographical data to generate a digital twin farmland; A three-dimensional simulation module, configured to obtain data of seeds to be sown; analyze the growth characteristics of the seeds to be sown to obtain growth characteristic data of the seeds to be sown; import the growth characteristic data of the seeds to be sown into the digital twin farmland for seeding simulation and growth simulation to generate seeding simulation data and growth simulation data; A seeding analysis module, configured to perform seeding profile analysis on the seeding simulation data to obtain a seeding profile diagram; analyze the root-shoot ratio imbalance of the seeding profile diagram according to the growth simulation data, and adjust the seeding simulation data based on the analysis result of the root-shoot ratio imbalance to obtain optimized seeding simulation data; A position correction module, configured to use the Internet of Things to import the optimized seeding simulation data into an intelligent seeder to execute seeding operations, and synchronously collect seeding operation data; perform real-time seeding position correction on the seeding operation data to execute intelligent agricultural seeding management operations.

Images