Intelligent agriculture remote cooperative scheduling method and system

Abstract

The invention relates to the technical field of data processing, and provides an intelligent agricultural remote collaborative scheduling method and system, which realize all-weather and all-directional real-time monitoring of an agricultural production environment. The system utilizes various sensors to monitor farmland environment parameters in real time, and transmits the parameters to a cloud or a local control system through a wireless network for analysis and processing. Farmers or agricultural managers can remotely monitor farmland conditions through terminal equipment, and scientific and accurate decision support is obtained. The system also has an automatic control function, can adjust an agricultural production plan and a management strategy according to real-time data, and optimizes a crop growth environment. The agricultural production efficiency and the management level are remarkably improved, the production cost and the labor intensity are reduced, the anti-risk capacity of agricultural production is enhanced, and the method has wide application prospects and practical value.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a smart agriculture remote collaborative scheduling method and system. Background Art

[0002] With the rapid development of science and technology, agricultural management systems have been widely used in various fields. Traditional agricultural management systems mainly rely on manual monitoring and simple automatic control, which can realize basic energy data collection and monitoring functions. However, with the rise of Internet of Things technology, agricultural management systems have gradually incorporated more intelligent elements. By deploying various sensors and intelligent devices, the system can collect a large amount of energy usage data in real time and transmit it to the data center through the network for analysis and processing. At the same time, the use of big data technology enables the system to explore the potential laws in energy use and provide more powerful data support for agricultural management. In addition, the introduction of artificial intelligence algorithms has further improved the intelligence level of the system, enabling the system to make more accurate energy scheduling and management decisions based on real-time data.

[0003] Although the existing intelligent agricultural management system has been significantly improved in terms of functionality, there are still some technical pain points that need to be addressed. On the one hand, due to the massiveness and complexity of energy usage data, traditional data processing and analysis methods often fail to meet the requirements of real-time and accuracy, resulting in possible delays or misjudgments in the system in actual applications. On the other hand, the existing agricultural management system still has certain limitations in terms of intelligent scheduling and management decision-making, and it is difficult to fully consider the mutual influence of various complex factors, so as to formulate the optimal energy usage plan.

[0004] In existing agricultural production, the monitoring of the production environment often relies on manual observation and empirical judgment, which is not only time-consuming and laborious, but also difficult to achieve all-weather and all-round real-time monitoring. Therefore, it is particularly important to develop a more efficient, intelligent and scalable agricultural management system. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention provides a remote collaborative scheduling method and system for smart agriculture to solve the problem that in existing agricultural production, the monitoring of the production environment often relies on manual observation and empirical judgment. This method is not only time-consuming and labor-intensive, but also difficult to achieve all-weather and all-round real-time monitoring.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows: In a first aspect, the invention provides a remote collaborative scheduling method for smart agriculture, comprising: Step S101, acquiring crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; Step S102, training a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; Step S103, receiving expected crop growth data, substituting the expected crop growth data into a crop breeding prediction model, the crop breeding prediction model outputs the expected crop growth data, collecting real-time crop image data, substituting the real-time crop image data into a preset digital twin model, and generating a real-time crop digital twin image; Step S104, copying the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, hashing the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and storing the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in the blockchain; Step S105, the first real-time crop digital twin image data and the real-time crop image data are substituted into the preset metaverse platform to generate a real-time crop growth metaverse image, the real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data, the first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data, the first metaverse comparison data is verified, if the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm, an optimized agricultural remote collaborative scheduling command is generated, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

[0007] Furthermore, the smart agriculture remote collaborative scheduling method, step S102, includes: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

[0008] Furthermore, the smart agriculture remote collaborative scheduling method, step S103, includes: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

[0009] Furthermore, the smart agriculture remote collaborative scheduling method, step S104, includes: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

[0010] Furthermore, the smart agriculture remote collaborative scheduling method, step S105, includes: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

[0011] In a second aspect, the smart agriculture remote collaborative scheduling system provided by the present invention is applied to the smart agriculture remote collaborative scheduling method as described above, including: A data acquisition module, used to acquire crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; A model building module is used to train a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; An image generation module is used to receive expected crop growth data, substitute the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model outputs the expected crop growth data, collects real-time crop image data, substitutes the real-time crop image data into a preset digital twin model, and generates a real-time crop digital twin image; A data processing module, used to copy the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, perform hash processing on the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and store the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in a blockchain; The data optimization module is used to substitute the first real-time crop digital twin image data and the real-time crop image data into the preset metaverse platform to generate a real-time crop growth metaverse image. The real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data. The first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data. The first metaverse comparison data is verified. If the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, and the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm to generate an optimized agricultural remote collaborative scheduling command, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

[0012] Furthermore, in the smart agriculture remote collaborative scheduling system of the present invention, the model building module is also used for: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

[0013] Furthermore, in the smart agriculture remote collaborative scheduling system of the present invention, the image generation module is also used for: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

[0014] Furthermore, in the smart agriculture remote collaborative scheduling system of the present invention, the data processing module is also used for: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

[0015] Furthermore, in the smart agriculture remote collaborative scheduling system of the present invention, the data optimization module is also used for: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

[0016] Beneficial effects of the present invention: The present invention integrates Internet of Things technology, uses various sensors to monitor farmland environmental parameters in real time, and transmits data to the cloud or local control system in real time through wireless networks, thus achieving all-weather, all-round real-time monitoring of the agricultural production environment. This greatly improves the accuracy and timeliness of monitoring and avoids the limitations of manual observation and empirical judgment.

[0017] Real-time monitoring and data analysis provide scientific and accurate decision-making support for agricultural production. Farmers or agricultural managers can adjust production plans and management strategies based on real-time data, optimize crop growth environments, and improve crop yields and quality. At the same time, the application of automated control technology also improves the efficiency and accuracy of agricultural operations.

[0018] Real-time monitoring and automated control reduce the need for manual monitoring and operations, and reduce production costs and labor intensity. Farmers or agricultural managers can remotely monitor and manage farmland through terminal devices without having to frequently visit the site for inspection and operation.

[0019] Real-time monitoring and early warning mechanisms can promptly detect and respond to potential risks in agricultural production, such as pests and diseases, abnormal climate, etc. This helps farmers or agricultural managers take measures in advance to reduce losses and enhance the risk resistance of agricultural production.

[0020] To sum up, the technical solution of the present invention realizes all-weather, all-round real-time monitoring of the agricultural production environment by integrating advanced technologies, improves agricultural production efficiency and management level, reduces production costs and labor intensity, promotes sustainable agricultural development, and enhances the risk resistance of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0022] Figure 1 A schematic flow chart of an outdoor large-screen energy-saving control method based on real-time human body monitoring provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0024] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0025] like Figure 1 As shown, in the first aspect, the smart agriculture remote collaborative scheduling method provided by the present invention includes: Step S101, acquiring crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; Location data collection: Deploy GPS positioning devices or Beidou-based positioning devices in farmland to obtain accurate crop location data. These devices can send location information to the data center periodically or in real time.

[0026] Growth environment data collection: Install a variety of environmental sensors, such as temperature and humidity sensors, light sensors, soil moisture sensors, etc., to monitor the growth environment data of crops in real time. The sensors transmit data to the data center via wired or wireless methods (such as ZigBee, LoRa, etc.).

[0027] Growth data collection: Automated monitoring equipment, such as growth monitors, is used to monitor the growth status of crops, including parameters such as plant height, leaf area, and fruit size. These data are also transmitted to the data center via wired or wireless means.

[0028] Agricultural equipment status data collection: Install status monitoring sensors for agricultural equipment (such as irrigation equipment, fertilization equipment, temperature control equipment, etc.) to monitor the working status and performance parameters of the equipment in real time. The data is transmitted to the data center through the equipment's built-in communication module or an external communication gateway.

[0029] Energy usage data collection: Install meters at the energy supply end (such as solar panels, grid interfaces, etc.) to monitor energy usage, including electricity and water consumption. The data is transmitted to the data center through the communication interface of the meter.

[0030] Crop image data collection: Deploy high-definition cameras or high-definition camera equipment mounted on drones to collect images of crops regularly or in real time. Image data is transmitted to the data center via wireless networks (such as Wi-Fi, 4G / 5G, etc.).

[0031] Weather forecast data acquisition: Cooperate with meteorological departments or professional meteorological data providers to obtain real-time weather forecast data through network interfaces, including temperature, humidity, rainfall, wind speed, etc.

[0032] Acquisition of crop market price data: Obtain real-time price data of crops in the market through web crawler technology or interfaces with cooperative data providers.

[0033] The collected raw data is cleaned to remove noise, outliers and duplicate data.

[0034] Data format conversion: Convert data from different sources into a unified format to facilitate subsequent data analysis and processing.

[0035] Data integration: Integrate the cleaned and converted data according to dimensions such as crops and time to form a structured data set.

[0036] Data storage: Store the integrated data in a database for subsequent data query and analysis. The database can be a relational database (such as MySQL, PostgreSQL, etc.) or a non-relational database (such as MongoDB, Cassandra, etc.), depending on the characteristics of the data and query requirements.

[0037] Data transmission protocol: Select a secure and reliable data transmission protocol, such as HTTPS, MQTT, etc., to ensure the security and reliability of data during transmission.

[0038] Data encryption: Encrypt sensitive data (such as crop location data, market price data, etc.) to prevent the data from being stolen or tampered with during transmission and storage.

[0039] Access control: Set up strict access control policies to ensure that only authorized users and systems can access and use the data.

[0040] Step S102, training a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; Data screening: Data related to crop breeding prediction is screened out from the data set obtained in step S101, including crop location data, growth environment data, growth data, equipment status data, energy usage data and image data.

[0041] Data cleaning: Clean the screened data to remove missing values, outliers and noise data to ensure data quality and consistency.

[0042] Data normalization / standardization: Since the data dimensions and ranges from different data sources may be different, the data needs to be normalized or standardized for subsequent model training.

[0043] Data enhancement: For image data, data enhancement techniques (such as rotation, flipping, scaling, cropping, etc.) can be used to increase data diversity and improve the generalization ability of the model.

[0044] Feature selection: According to the needs of crop breeding forecast, key features are selected from the preprocessed data. These features may include temperature, humidity, light intensity, soil moisture, crop growth rate, growth amount, equipment operation status, etc.

[0045] Feature extraction: For image data, deep learning models such as convolutional neural networks (CNNs) can be used for feature extraction to convert high-dimensional image data into low-dimensional feature vectors.

[0046] Model selection: According to the specific tasks of crop breeding prediction (such as yield prediction, pest and disease prediction, etc.), select a suitable neural network model, such as fully connected neural network (FNN), convolutional neural network (CNN), recurrent neural network (RNN) or its variants (such as LSTM, GRU, etc.).

[0047] Model construction: According to the selected model type, build the corresponding neural network structure. For example, for the yield prediction task, you can choose FNN or CNN as the basic model; for the pest prediction task, since time series information needs to be considered, you can choose RNN or its variants.

[0048] Model parameter setting: Set the model's hyperparameters, such as learning rate, batch size, number of iterations, number of network layers, number of neurons, etc. The selection of these parameters needs to be adjusted according to the specific task and dataset.

[0049] Data partitioning: The preprocessed data set is divided into training set, validation set and test set. The training set is used for model training, the validation set is used for model tuning, and the test set is used to evaluate the performance of the model.

[0050] Model training: Use the training set data to train the neural network model, and adjust the model parameters through the back propagation algorithm to minimize the error between the model's predicted results and the actual results.

[0051] Model evaluation: Evaluate the trained model on the validation set and calculate performance indicators (such as accuracy, recall, F1 score, etc.). Optimize the model based on the evaluation results, such as adjusting hyperparameters and changing model structure.

[0052] Model tuning: Tune the model through cross-validation, grid search, random search and other methods to find the optimal model parameters and structure.

[0053] The trained and optimized crop breeding prediction model is deployed to the smart agriculture management system so that it can be applied in the actual production environment. With the continuous accumulation of data and the continuous development of technology, the model is regularly updated and optimized to improve its prediction accuracy and generalization ability.

[0054] Step S103, receiving expected crop growth data, substituting the expected crop growth data into a crop breeding prediction model, the crop breeding prediction model outputs the expected crop growth data, collecting real-time crop image data, substituting the real-time crop image data into a preset digital twin model, and generating a real-time crop digital twin image; Data reception: receiving the expected crop growth data input by the user through the user interface or other data interface. The data may include the expected crop yield, growth cycle, quality requirements, etc.

[0055] Data preprocessing: Preprocess the received expected data, such as format conversion, data verification, etc., to ensure the consistency and accuracy of the data.

[0056] Model substitution: Substitute the preprocessed crop growth expectation data into the trained crop breeding prediction model.

[0057] Expected data output: The model calculates and outputs expected crop growth data based on the input expected data. These expected data may include expected crop growth rate, growth amount, maturity time, quality prediction, etc.

[0058] Image acquisition: Real-time image data of crops can be collected through cameras or other image acquisition devices installed in the farmland.

[0059] Image preprocessing: Preprocess the collected image data, including denoising, enhancement, cropping, resizing and other operations, for subsequent digital twin model processing.

[0060] Model selection: Select or build a suitable digital twin model that can generate digital twin images of crops based on real-time image data.

[0061] Image generation: The pre-processed real-time crop image data is substituted into the preset digital twin model, and the model generates real-time digital twin images of crops based on the image data. These images may include visual information such as the three-dimensional model of the crops, growth status, and environmental parameters.

[0062] Image optimization: Optimize the generated digital twin images, such as improving the resolution of the images, enhancing the realism of the images, etc., so that users can better understand and analyze the growth of crops.

[0063] Image display: The generated real-time digital twin images of crops are displayed to users, who can view the growth of crops in real time through the user interface.

[0064] Decision support: Based on the generated digital twin images and expected crop growth data, agricultural managers can be provided with decision support, such as adjusting irrigation strategies, fertilization plans, and pest and disease control measures.

[0065] Step S104, copying the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, hashing the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and storing the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in the blockchain; Data replication: First, the crop digital twin image data generated in real time is replicated to generate three identical copies of the data, which are named the first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data. The purpose of this step is to improve the redundancy and integrity of the data and provide a basis for subsequent data verification and storage.

[0066] Hash algorithm selection: Select a suitable hash algorithm (such as SHA-256, MD5, etc.) to hash the real-time crop digital twin image data. The selection of the hash algorithm should take into account factors such as security, collision resistance, and computational efficiency.

[0067] Hash value generation: Substitute the first, second and third real-time crop digital twin image data into the selected hash algorithm, perform hash calculations, and generate corresponding hash values. These hash values ​​are named the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value, and the third real-time crop digital twin image data hash value.

[0068] Hash value verification: Verify the generated hash value to ensure the correctness of the hash algorithm and the uniqueness of the hash value. If the verification fails, the hashing process needs to be repeated.

[0069] Blockchain network selection: Choose a suitable blockchain network (such as public chain, private chain or alliance chain) to store hash values. The selection of blockchain network should take into account factors such as data security, privacy, scalability and cost.

[0070] Smart contract development: If the blockchain network supports smart contracts, a smart contract can be developed to automatically store and manage hash values. Smart contracts should include data storage rules, access rights control, data integrity verification and other functions.

[0071] Hash value storage: The generated first, second, and third real-time crop digital twin image data hash values ​​are transmitted to the selected blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain. During the storage process, the security and privacy of the data should be ensured to prevent data leakage and tampering.

[0072] Storage record management: Record relevant information of storage operations on the blockchain, such as storage time, storage address, storage user, etc., for subsequent data tracing and auditing.

[0073] Data integrity verification: Verify the integrity of the data by comparing the hash value stored on the blockchain with the hash value of the original data. If the hash values ​​do not match, it means that the data may have been tampered with during transmission or storage.

[0074] Data access control: Set up a data access control mechanism to ensure that only authorized users can access and modify data on the blockchain. This can be achieved through smart contracts or access control policies of the blockchain network.

[0075] Data encryption: During data transmission and storage, encryption technology can be used to encrypt data to improve data security. Encryption technology can be symmetric encryption, asymmetric encryption or hybrid encryption.

[0076] Step S105, the first real-time crop digital twin image data and the real-time crop image data are substituted into the preset metaverse platform to generate a real-time crop growth metaverse image, the real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data, the first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data, the first metaverse comparison data is verified, if the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm, an optimized agricultural remote collaborative scheduling command is generated, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

[0077] Metaverse platform selection: Select a Metaverse platform that supports 3D modeling, real-time rendering, and interaction. The platform should be able to provide sufficient computing resources and graphics processing capabilities to generate high-quality Metaverse images.

[0078] Data Substitution: Substitute the first real-time crop digital twin image data and real-time crop image data into the Metaverse platform. The first real-time crop digital twin image data provides accurate 3D models and growth status information of crops, while the real-time crop image data provides the actual appearance and growth environment information of crops.

[0079] Metaverse image synthesis: In the Metaverse platform, the first real-time crop digital twin image data is integrated with the real-time crop image data to generate a real-time crop growth Metaverse image. This image contains both the precise 3D model of the crop and its growth status in the actual environment.

[0080] Metaverse data extraction: Extract the first real-time crop metaverse data and crop image data metaverse data from the generated metaverse image. The first real-time crop metaverse data represents the virtual state of the crop in the metaverse, while the crop image data metaverse data represents the state of the crop in the actual environment mapped to the data in the metaverse.

[0081] Data comparison: Compare the first real-time crop metaverse data with the crop image data metaverse data to analyze the differences between the two. These differences may include differences in the growth rate, shape, color, etc. of the crops.

[0082] Comparative data generation: Based on the comparison results, the first metaverse comparative data is generated, which contains the difference information between crops in the metaverse and the actual environment.

[0083] Data verification: Verify the first metaverse comparison data to determine whether the difference is within an acceptable range. The verification process may involve manual review, automatic algorithm detection, or a combination of both.

[0084] Genetic algorithm preparation: Select genetic algorithm as the optimization tool because it is good at dealing with complex, nonlinear optimization problems. Set the parameters of the genetic algorithm, such as population size, crossover probability, mutation probability, etc.

[0085] Model optimization: The first metaverse error data in the first metaverse comparison data is used as the input of the genetic algorithm to optimize the preset agricultural remote collaborative scheduling model. The optimization goal is to reduce the difference between the metaverse and the actual environment and improve the accuracy and efficiency of agricultural remote collaborative scheduling.

[0086] Optimization command generation: Based on the optimization results of the genetic algorithm, optimized agricultural remote collaborative scheduling commands are generated. These commands may include irrigation amount, fertilizer type and quantity, pest control measures, etc.

[0087] Command sending: The optimized agricultural remote collaborative dispatching command is sent to the agricultural equipment end through the communication network. The communication network may include a wireless network, a wired network, or a combination thereof.

[0088] Equipment execution: The agricultural equipment receives and executes the optimized scheduling commands, adjusts the growth environment and management measures of crops to reduce the difference between the metaverse and the actual environment.

[0089] Execution monitoring: Real-time monitoring of the execution of agricultural equipment to ensure that dispatch orders are accurately executed.

[0090] Feedback mechanism: Establish a feedback mechanism to collect the execution results of agricultural equipment and real-time data on crop growth status, and provide data support for subsequent metaverse image generation and model optimization.

[0091] Specifically, the smart agriculture remote collaborative scheduling method, step S102, includes: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

[0092] Crop location data: obtained through GPS or other positioning technologies.

[0093] Crop growth environment data: including soil moisture, temperature, light intensity, CO2 concentration, etc., collected by environmental sensors.

[0094] Crop growth data: such as growth cycle, growth rate, yield, etc., are recorded by the agricultural management system.

[0095] Agricultural equipment status data: reflects the working status, usage time, etc. of agricultural machinery and equipment.

[0096] Energy usage data: Record energy consumption in agricultural production.

[0097] Crop image data: obtained by cameras or drones.

[0098] Weather forecast data: obtained from meteorological departments or third-party weather services.

[0099] Crop market price data: obtained from the market or relevant departments.

[0100] Data cleaning: Remove duplicate data, missing data or abnormal data. Standardize or normalize the data to ensure data consistency.

[0101] Data transformation: Convert non-numeric data (such as text, images) into numerical form for model processing. Perform timestamp alignment and missing value filling for time series data.

[0102] Feature Engineering: Extract features useful for crop growth prediction. Construct new features, such as derived indicators obtained through calculations.

[0103] Dataset division: Random partitioning: The preprocessed training data is randomly divided into training set, validation set and test set.

[0104] Proportional allocation: Usually, the training set accounts for a larger proportion (such as 70%), and the validation set and test set each account for a smaller proportion (such as 15% and 15%).

[0105] Stratified sampling: If the data is class imbalanced, use stratified sampling to ensure that each class is represented in the training set, validation set, and test set.

[0106] Neural network model training: Model selection: Choose an appropriate neural network model based on the task requirements, such as convolutional neural network (CNN) for image processing, recurrent neural network (RNN) for time series analysis, or a hybrid model.

[0107] Model initialization: Randomly initialize model parameters or use the parameters of a pre-trained model for initialization.

[0108] Training process: Use the training set data to train the model. Calculate the gradient of the loss function through the back-propagation algorithm and update the model parameters. Use optimization algorithms such as batch gradient descent or stochastic gradient descent.

[0109] Regularization technology: Apply L1 or L2 regularization to prevent overfitting. Use Dropout technology to randomly discard some neurons to improve the generalization ability of the model.

[0110] Validation set evaluation: Evaluate the performance of the model on the validation set, such as accuracy, recall, F1 score, etc. Adjust model parameters such as learning rate, batch size, number of network layers, etc. based on the validation results.

[0111] Hyperparameter tuning: Use grid search or random search methods to tune hyperparameters. Select the best hyperparameter combination through cross validation.

[0112] Model export: Export the trained model to a deployable format, such as TensorFlow's SavedModel or PyTorch's TorchScript. Integrate the exported model into the smart agriculture management system. Monitor the performance of the deployed model to ensure its stable performance in actual applications. Update the model regularly to adapt to changes in the crop growth environment and new data features.

[0113] Specifically, the smart agriculture remote collaborative scheduling method, step S103, includes: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

[0114] Receive user input: Design a user interface to allow users to input expected crop growth data, such as expected growth cycle, yield target, quality requirements, etc. Verify the data entered by the user to ensure the rationality and validity of the data.

[0115] Operation of crop breeding prediction model: Input the verified expected crop growth data into the crop breeding prediction model. The model calculates and outputs the expected crop growth data based on the input data and the learned crop growth laws.

[0116] The expected crop growth data include crop growth rate, growth amount (such as expected yield), and shape data (such as plant height, leaf shape, etc.).

[0117] Image data collection: Real-time crop image data is collected through cameras installed in farmland.

[0118] Image data preprocessing: Perform denoising, enhancement and other preprocessing operations on the collected image data to improve image quality.

[0119] The image is cropped, scaled, and processed to meet the input requirements of the digital twin model.

[0120] Digital twin model preparation: Preset the digital twin model, which can generate a three-dimensional model of crops, growth status and environmental parameters based on the input image data.

[0121] Substituting image data into the model: Input the preprocessed real-time crop image data into the preset digital twin model.

[0122] The model generates real-time digital twin images of crops based on information in the image data, such as the shape, color, and texture of the crops.

[0123] Digital twin image content: Real-time crop digital twin images include a 3D model of the crop, which accurately reflects the crop’s actual shape and structure. The image also includes the crop’s growth status, such as growth stage, health status, etc.

[0124] In addition, the images also include environmental parameters of crops, such as soil moisture, temperature, light intensity, etc., which have an important impact on the growth of crops.

[0125] Data output: Output expected crop growth data and real-time crop digital twin images to the user interface or smart agriculture management system, allowing users to view, analyze and compare these data to make more informed decisions.

[0126] Data application: Use crop growth forecast data to develop reasonable planting plans and management strategies. Monitor crop growth through real-time crop digital twin images to promptly identify and address potential problems. Combine this data with other agricultural data (such as weather forecasts, market prices, etc.) to provide more comprehensive support for remote collaborative scheduling of smart agriculture.

[0127] Specifically, the smart agriculture remote collaborative scheduling method, step S104, includes: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

[0128] Data acquisition: Acquire first real-time crop digital twin image data, second real-time crop digital twin image data, and third real-time crop digital twin image data from the smart agriculture management system.

[0129] Hash algorithm selection: Select a secure and efficient hash algorithm, such as SHA-256, to perform hash calculations on real-time crop digital twin image data. Perform hash calculations on the first real-time crop digital twin image data to generate a first hash value. Perform hash calculations on the second real-time crop digital twin image data to generate a second hash value. Perform hash calculations on the third real-time crop digital twin image data to generate a third hash value.

[0130] Communication protocol selection: Choose a secure communication protocol, such as HTTPS or TLS, to transmit the generated hash value to the blockchain network. Make the communication protocol have encryption and authentication functions to prevent data from being stolen or tampered with during transmission.

[0131] Data transmission: The three generated hash values ​​are transmitted to the nodes in the blockchain network through the selected communication protocol.

[0132] Blockchain network configuration: Configure the blockchain network, including node settings, consensus mechanism selection, smart contract deployment, etc.

[0133] Distributed ledger storage: The received hash value is stored in the distributed ledger of the blockchain. The consistency and accuracy of the ledger are ensured through the consensus mechanism to prevent data from being maliciously tampered with.

[0134] Data query: Provide an interface or tool that allows users or systems to query hash values ​​stored on the blockchain.

[0135] Data verification: When the integrity or authenticity of real-time crop digital twin image data needs to be verified, its hash value can be calculated and compared with the hash value stored on the blockchain. If the hash values ​​match, it proves that the data has not been tampered with and is authentic; otherwise, the data may have been tampered with or is not up to date.

[0136] Specifically, the smart agriculture remote collaborative scheduling method, step S105, includes: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

[0137] The system receives first real-time crop Metaverse data and crop image data Metaverse data from the Metaverse platform.

[0138] The data should contain necessary metadata, such as timestamp, geographic location, sensor information, etc., to ensure the traceability and accuracy of the data.

[0139] Data preprocessing: The received Metaverse data is preprocessed, including denoising, enhancement, and registration steps to improve the accuracy of subsequent comparisons. The registration step aligns the two sets of data in space to facilitate pixel-level comparison and feature point matching.

[0140] Metaverse data pixel-level comparison: Use pixel-level comparison algorithms to compare the first real-time crop Metaverse data and crop image data Metaverse data pixel by pixel. Generate a difference map in which the difference areas are highlighted with significant colors or markers for easy visual identification.

[0141] Feature point matching: Extract feature points from two sets of data, such as edges, textures, corners, etc., and use feature matching algorithms to match them. Calculate the matching rate and matching error of feature points as part of similarity evaluation.

[0142] Similarity calculation: Based on the results of pixel-level comparison and feature point matching, similarity scores of two sets of data are calculated using similarity measurement algorithms (such as cosine similarity, structural similarity, etc.). The similarity score is used to quantify the similarity between two sets of data, providing a quantitative basis for subsequent error extraction and model optimization.

[0143] Completeness check: Check whether the comparison data is complete, including difference graphs, similarity scores, and necessary metadata.

[0144] Range verification: Compare the similarity score with the preset threshold or range to determine whether the comparison result meets expectations. If the similarity score is within the preset range, the verification passes; otherwise, an alarm is triggered or the comparison is repeated.

[0145] Extract error data: If the comparison data is verified, extract the error data from the difference map, including timestamp, error value (such as the area of ​​the difference area, the number of pixel differences, etc.) and error source information (such as sensor drift, lighting changes, etc.).

[0146] Error data should be stored in a database for subsequent analysis and traceability.

[0147] Error analysis: Analyze the extracted error data to identify the main sources and distribution patterns of the errors. According to the analysis results, adjust the acquisition parameters or preprocessing process of the Metaverse data to reduce the occurrence of errors.

[0148] Prepare the optimization model: Select the preset agricultural remote collaborative scheduling model as the optimization object to ensure that the model is adaptable and adjustable. The model should be able to dynamically adjust the scheduling strategy to optimize the operating efficiency of agricultural equipment based on the input metaverse error data.

[0149] Genetic algorithm optimization: The first metaverse error data is used as input to optimize the agricultural remote collaborative scheduling model using a genetic algorithm. The genetic algorithm iteratively calculates and adjusts the model parameters through operations such as selection, crossover, and mutation to minimize the error or maximize the scheduling efficiency. During the optimization process, the performance indicators of the model, such as scheduling accuracy and response time, should be monitored to ensure the optimization effect.

[0150] Optimization result verification: Verify the optimized model, use historical data or simulated data for testing, and evaluate its performance in actual applications. If the verification passes, the optimized model will be deployed to the production environment to generate scheduling commands.

[0151] Generate dispatch commands: The command generation module generates dispatch commands for agricultural equipment based on the optimized agricultural remote collaborative dispatch model, combined with real-time metaverse data and agricultural equipment status. Dispatch commands should contain clear operation instructions, operation time, operation targets and other information to ensure that agricultural equipment can be executed accurately.

[0152] Sending dispatch commands: The command sending module sends the generated dispatch commands to the agricultural equipment end through the communication network. After receiving the dispatch commands, the agricultural equipment end should perform corresponding operations and feed back the execution results to the system for monitoring and evaluation.

[0153] In a second aspect, the smart agriculture remote collaborative scheduling system provided by the present invention is applied to the smart agriculture remote collaborative scheduling method as described above, including: A data acquisition module, used to acquire crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; A model building module is used to train a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; An image generation module is used to receive expected crop growth data, substitute the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model outputs the expected crop growth data, collects real-time crop image data, substitutes the real-time crop image data into a preset digital twin model, and generates a real-time crop digital twin image; A data processing module, used to copy the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, perform hash processing on the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and store the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in a blockchain; The data optimization module is used to substitute the first real-time crop digital twin image data and the real-time crop image data into the preset metaverse platform to generate a real-time crop growth metaverse image. The real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data. The first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data. The first metaverse comparison data is verified. If the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, and the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm to generate an optimized agricultural remote collaborative scheduling command, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

[0154] Specifically, in the smart agriculture remote collaborative scheduling system of the present invention, the model building module is also used for: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

[0155] Specifically, in the smart agriculture remote collaborative scheduling system of the present invention, the image generation module is also used for: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

[0156] Specifically, in the smart agriculture remote collaborative scheduling system of the present invention, the data processing module is also used for: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

[0157] Specifically, the data optimization module of the smart agriculture remote collaborative scheduling system of the present invention is also used for: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

[0158] The technical solution of the present invention integrates advanced technologies such as the Internet of Things, big data, cloud computing and artificial intelligence, effectively solving the problem of relying on manual observation and empirical judgment for production environment monitoring in existing agricultural production, and realizing all-weather, all-round real-time monitoring of the agricultural production environment. The specific solutions are as follows: The present invention uses various sensors (such as soil moisture sensors, temperature sensors, light intensity sensors, CO2 concentration sensors, etc.) to monitor farmland environmental parameters in real time. These sensors can accurately obtain data such as soil moisture, nutrient status, and climate conditions, providing a scientific basis for agricultural production. Sensor data is transmitted to a cloud data center or a local control system in real time via wireless networks (such as LoRa, Wi-Fi, 4G / 5G, etc.), improving the timeliness and accuracy of the data.

[0159] Farmers or agricultural managers can remotely monitor farmland environmental conditions through mobile phones, computers and other terminal devices, view sensor data in real time, and understand the growth environment and growth status of crops. The remote monitoring system supports data visualization, and intuitively reflects the changing trends of farmland environmental parameters through charts, maps and other forms, which is convenient for users to understand and make decisions.

[0160] Cloud data centers or local control systems store and manage the massive amounts of sensor data collected to form an agricultural production database. Big data analysis and machine learning algorithms are used to process and analyze sensor data to reveal key influencing factors and optimal management strategies in the crop growth cycle. For example, by analyzing historical meteorological data and crop growth records, the probability of future crop diseases and pests can be predicted, and preventive measures can be taken in advance.

[0161] Based on the results of big data analysis, crop growth models and climate models are constructed to provide accurate prediction and decision support for agricultural production. The agricultural remote collaborative scheduling model is optimized using optimization algorithms such as genetic algorithms, and the scheduling strategy is dynamically adjusted according to real-time sensor data and crop growth requirements to improve agricultural production efficiency.

[0162] Using artificial intelligence technologies, such as image recognition and voice recognition, crop image data is analyzed and processed to identify crop growth status, pests and diseases, etc. When abnormal data or potential risks are detected, the system can automatically trigger an alarm mechanism and notify farmers or agricultural managers through SMS, email, APP push, etc., so that timely measures can be taken.

[0163] According to sensor data and crop growth requirements, agricultural facilities such as smart irrigation systems and smart greenhouses can automatically adjust environmental factors such as irrigation volume, temperature, humidity, and light to create optimal growth conditions for crops. Drones and smart agricultural machinery can perform precise fertilization, sowing, harvesting, and other operations to improve operational efficiency and accuracy. All-weather, all-round real-time monitoring of the agricultural production environment is achieved, avoiding the limitations of manual observation and empirical judgment. The accuracy and timeliness of data collection are improved, and the cost and labor intensity of manual monitoring are reduced.

[0164] To sum up, the technical solution of the present invention effectively solves the problem of existing agricultural production that production environment monitoring relies on manual observation and experience judgment by integrating advanced technologies such as the Internet of Things, big data, cloud computing and artificial intelligence, and realizes all-weather, all-round real-time monitoring of the agricultural production environment, providing scientific and accurate decision-making support for agricultural production.

Claims

1. A remote collaborative scheduling method for smart agriculture, characterized in that: include: Step S101, acquiring crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; Step S102, training a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; Step S103, receiving expected crop growth data, substituting the expected crop growth data into a crop breeding prediction model, the crop breeding prediction model outputs the expected crop growth data, collecting real-time crop image data, substituting the real-time crop image data into a preset digital twin model, and generating a real-time crop digital twin image; Step S104, copying the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, hashing the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and storing the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in the blockchain; Step S105, the first real-time crop digital twin image data and the real-time crop image data are substituted into the preset metaverse platform to generate a real-time crop growth metaverse image, the real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data, the first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data, the first metaverse comparison data is verified, if the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm, an optimized agricultural remote collaborative scheduling command is generated, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

2. According to the smart agriculture remote collaborative scheduling method as claimed in claim 1, it is characterized in that: The step S102 includes: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

3. According to the smart agriculture remote collaborative scheduling method as claimed in claim 1, it is characterized in that: The step S103 includes: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

4. According to the smart agriculture remote collaborative scheduling method as claimed in claim 1, it is characterized in that: The step S104 includes: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

5. According to the smart agriculture remote collaborative scheduling method as claimed in claim 1, it is characterized in that: The step S105 includes: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

6. A smart agriculture remote collaborative scheduling system, applied to the smart agriculture remote collaborative scheduling method as claimed in any one of claims 1 to 5, characterized in that: include: A data acquisition module, used to acquire crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data; A model building module is used to train a neural network model based on crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, and crop image data to obtain a crop breeding prediction model; An image generation module is used to receive expected crop growth data, substitute the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model outputs the expected crop growth data, collects real-time crop image data, substitutes the real-time crop image data into a preset digital twin model, and generates a real-time crop digital twin image; A data processing module, used to copy the real-time crop digital twin image data to generate first real-time crop digital twin image data, second real-time crop digital twin image data and third real-time crop digital twin image data, perform hash processing on the first real-time crop digital twin image data, the second real-time crop digital twin image data and the third real-time crop digital twin image data in sequence to generate a first real-time crop digital twin image data hash value, a second real-time crop digital twin image data hash value and a third real-time crop digital twin image data hash value, and store the first real-time crop digital twin image data hash value, the second real-time crop digital twin image data hash value and the third real-time crop digital twin image data hash value in a blockchain; The data optimization module is used to substitute the first real-time crop digital twin image data and the real-time crop image data into the preset metaverse platform to generate a real-time crop growth metaverse image. The real-time crop growth metaverse image includes the first real-time crop metaverse data and the real-time crop image data metaverse data. The first real-time crop metaverse data is compared with the crop image data metaverse data to obtain the first metaverse comparison data. The first metaverse comparison data is verified. If the first metaverse comparison data is verified, the first metaverse error data in the first metaverse comparison data is retrieved, and the preset agricultural remote collaborative scheduling model is optimized using a genetic algorithm to generate an optimized agricultural remote collaborative scheduling command, and the optimized agricultural remote collaborative scheduling command is sent to the agricultural equipment end.

7. According to the smart agriculture remote collaborative scheduling system as claimed in claim 6, it is characterized in that: The model building module is also used to: Preprocessing crop location data, crop growth environment data, crop growth data, agricultural equipment status data, energy usage data, crop image data, weather forecast data, and crop market price data to obtain preprocessed data to be trained; The preprocessed training data is divided into training set, validation set and test set. The neural network model is trained using the training set data. The model parameters are adjusted using the back propagation algorithm. The model performance is evaluated on the validation set. The model is optimized based on the validation results. Deploy the trained crop breeding prediction model into the smart agriculture management system.

8. According to the smart agriculture remote collaborative scheduling system as claimed in claim 6, it is characterized in that: The image generation module is also used for: The crop breeding prediction model outputs expected crop growth data based on the input crop growth expectation data, collects real-time crop image data, preprocesses the real-time crop image data, substitutes the preprocessed real-time crop image data into the preset digital twin model, and generates real-time crop digital twin images; receiving expected crop growth data input by a user, inputting the expected crop growth data into a crop breeding prediction model, and the crop breeding prediction model calculating and outputting the expected crop growth data, the expected crop growth data including crop growth rate, crop growth amount and crop shape data; The camera collects crop image data in real time, preprocesses the image data, and inputs the preprocessed real-time crop image data into a preset digital twin model. The preset digital twin model generates real-time crop digital twin images based on the image data. The real-time crop digital twin images include three-dimensional models of crops, growth status of crops, and environmental parameters of crops.

9. According to the smart agriculture remote collaborative scheduling system as claimed in claim 6, it is characterized in that: The data processing module is further used for: The first real-time crop digital twin image data, the second real-time crop digital twin image data, and the third real-time crop digital twin image data are hashed, and hash calculations are performed respectively to generate hash values. The three generated hash values ​​are transmitted to the blockchain network through a secure communication protocol and stored in the distributed ledger of the blockchain.

10. According to the smart agriculture remote collaborative scheduling system as claimed in claim 6, it is characterized in that: The data optimization module is also used to: Receiving first real-time crop metaverse data and crop image data metaverse data as input, comparing the two sets of data using pixel-level comparison, feature point matching, and similarity calculation to generate first metaverse comparison data, the first metaverse comparison data including a difference map and a similarity score; Check the integrity of the comparison data and whether it is within the expected range. If the verification passes, the accuracy of the comparison data of the first metaverse is confirmed; If the verification is successful, the first metaverse error data is extracted from the first metaverse comparison data, and the first metaverse error data includes a timestamp, an error value, and error source information; The preset agricultural remote collaborative scheduling model and the first metaverse error data are used as input, and the model is optimized using a genetic algorithm. The parameters of the agricultural remote collaborative scheduling model are adjusted through iterative calculation of the genetic algorithm to obtain the optimized agricultural remote collaborative scheduling model.

Images