Land monitoring method, system and equipment based on knowledge graph and medium

Abstract

The invention provides a land monitoring method, system and device based on a knowledge graph, and a medium. The method comprises the steps of obtaining land characteristic data of a to-be-monitored area; performing semantic marking on the land characteristic data by using a pre-constructed knowledge graph to obtain semantic label information of the land characteristic data, and performing block chain storage on the land characteristic data and the semantic label information thereof; based on the land characteristic data and the semantic tag information thereof, performing disaster prediction on the to-be-monitored area by adopting a dynamic game algorithm to obtain a disaster occurrence probability of the to-be-monitored area; according to the disaster occurrence probability of the to-be-monitored area, performing emergency response on the to-be-monitored area through the smart contract on the block chain, and generating a multi-mode response instruction of the to-be-monitored area; by means of the method, the real-time performance and accuracy of land monitoring can be improved, efficient disaster prediction and an automatic emergency response mechanism can be achieved, and therefore the prevention and coping capacity to natural disasters is remarkably enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of land monitoring, and particularly relates to a land monitoring method, system, device and medium based on a knowledge graph. Background Art

[0002] Currently, with the increasing trend of frequent natural disasters, the importance of land monitoring has become more prominent. Traditional land monitoring methods have played an important role in early land management with their stability and the ability to obtain basic data. However, in the face of the rapidly changing impacts of natural disasters, these methods have shown significant deficiencies, especially in their ability to respond urgently to changes in land characteristics.

[0003] Traditional land monitoring methods mostly rely on periodic ground surveys and the acquisition of remote sensing data. These methods usually can only provide static and historical land characteristic information, and cannot dynamically reflect the sudden changes in land during natural disasters. For example, within a short period of time when floods, debris flows or other disasters occur, characteristics such as soil moisture, erosion conditions, and crop growth status will change rapidly. Due to the long data collection cycle and low update frequency of traditional monitoring means, it is often difficult to capture these changes in the first place, resulting in untimely disaster response and increasing the risk of losses caused by disasters. Summary of the Invention

[0004] In order to solve the problem that when encountering natural disasters, the land characteristics will change rapidly, and the traditional land monitoring methods cannot keep up in a timely manner, resulting in untimely disaster response and inability to effectively prevent and reduce losses, the present invention proposes a land monitoring method based on a knowledge graph, including:

[0005] Obtain land characteristic data of the area to be monitored;

[0006] Use the pre-constructed knowledge graph to perform semantic tagging on the land characteristic data to obtain semantic tag information of the land characteristic data, and store the land characteristic data and the semantic tag information of the land characteristic data on the blockchain;

[0007] Based on the land characteristic data and the semantic tag information of the land characteristic data, use a dynamic game algorithm to predict disasters in the area to be monitored, and obtain the probability of disasters occurring in the area to be monitored;

[0008] According to the probability of disasters occurring in the area to be monitored, perform emergency response on the area to be monitored through the smart contract on the blockchain, and generate multi-mode response instructions for the area to be monitored.

[0009] Optionally, the obtaining of the land characteristic data of the area to be monitored includes:

[0010] Collect the ground-air collaborative data of the area to be monitored by using a hybrid collaborative network;

[0011] Extract features from the ground-air collaborative data through an environmental perception algorithm to obtain the land characteristic data of the area to be monitored;

[0012] Among them, the hybrid collaborative network is collaboratively constructed based on an unmanned aerial vehicle (UAV) platform and ground sensors.

[0013] Optionally, the ground-air collaborative data includes: high-altitude acquisition data and ground acquisition data;

[0014] The high-altitude acquisition data includes one or more of the following: field area distribution data, three-dimensional terrain data, temperature gradient data, gas concentration gradient data, crop type data, disaster characteristic data;

[0015] The disaster characteristic data includes one or more of the following: surface deformation data, crack width data, and thermal anomaly data;

[0016] The ground acquisition data includes one or more of the following: soil moisture data, meteorological data, water level change data, soil temperature data, soil pH value, soil conductivity data, water quality data, crop status data;

[0017] The meteorological data includes one or more of the following: temperature data, humidity data, wind speed data, and rainfall data.

[0018] Optionally, the pre-construction process of the knowledge graph includes:

[0019] Obtain the historical land characteristic data of the area to be monitored and a Geographic Information System (GIS) database;

[0020] Perform data extraction on the historical land characteristic data to obtain the entity information and semantic label information of the historical land characteristic data;

[0021] Perform embedding processing on the entity information and semantic label information of the historical land characteristic data to generate the representation features of the historical land characteristic data;

[0022] Extract geographical features from the Geographic Information System (GIS) database, and perform weighted splicing on the representation features and the geographical features to obtain a comprehensive vector representation;

[0023] Construct a knowledge graph according to the comprehensive vector representation.

[0024] Optionally, the blockchain storage of the land characteristic data and the semantic label information of the land characteristic data includes:

[0025] Encrypt the land characteristic data and the semantic tag information of the land characteristic data in layers respectively, and output encrypted data blocks;

[0026] Based on the encrypted data blocks, construct a spatio-temporal Merkle forest structure, and output a spatio-temporal Merkle root set;

[0027] Through a smart contract, anchor the elements in the Merkle root set to the target block in the blockchain for storage according to the preset index tags.

[0028] Optionally, based on the land characteristic data and the semantic tag information of the land characteristic data, use a dynamic game algorithm to predict disasters in the area to be monitored, and obtain the disaster occurrence probability of the area to be monitored, including:

[0029] Perform non-linear feature mapping on the land characteristic data and the semantic tag information of the land characteristic data to generate a high-dimensional feature space vector;

[0030] Regard the stakeholders in the area to be monitored as participants in the game; the stakeholders include: the head of the field, farmers, and enterprises;

[0031] Based on the high-dimensional feature vector, determine the target selection strategies of the participants;

[0032] Based on each participant, calculate the payoff value of the target selection strategy of the participant;

[0033] According to the payoff values of the participants, determine the final selection strategy through regression analysis;

[0034] According to the final selection strategy, output the disaster occurrence probability of the area to be monitored through Bayesian inference.

[0035] Optionally, according to the disaster occurrence probability of the area to be monitored, perform an emergency response on the area to be monitored through the smart contract on the blockchain, and generate a multi-mode response instruction for the area to be monitored, including:

[0036] According to the disaster occurrence probability of the area to be monitored, determine the response priorities for different disaster types;

[0037] According to the response priorities for different disaster types, call the smart contract on the blockchain to output an emergency response plan;

[0038] Utilize the event listening mechanism of the smart contract to generate a multi-mode response instruction for the area to be monitored;

[0039] Among them, the multi-mode response instructions include one or more of the following: SMS warning, UAV dispatching and inspection path, broadcast information sending, and emergency channel response instruction.

[0040] Based on the same inventive concept, the present invention also provides a land monitoring system based on a knowledge graph, including:

[0041] A data acquisition module, configured to acquire land characteristic data of an area to be monitored;

[0042] A data storage module, configured to perform semantic tagging on the land characteristic data by using a pre-constructed knowledge graph to obtain semantic tag information of the land characteristic data, and perform blockchain storage on the land characteristic data and the semantic tag information of the land characteristic data;

[0043] A disaster prediction module, configured to perform disaster prediction on the area to be monitored by using a dynamic game algorithm based on the land characteristic data and the semantic tag information of the land characteristic data to obtain the disaster occurrence probability of the area to be monitored;

[0044] An emergency response module, configured to perform emergency response on the area to be monitored through a smart contract on the blockchain according to the disaster occurrence probability of the area to be monitored, and generate multi-mode response instructions for the area to be monitored.

[0045] Optionally, the data acquisition module includes:

[0046] A data collection sub-module, configured to collect ground-air collaborative data of the area to be monitored by using a hybrid collaborative network;

[0047] A feature extraction sub-module, configured to perform feature extraction on the ground-air collaborative data by using an environment perception algorithm to obtain land characteristic data of the area to be monitored;

[0048] Among them, the hybrid collaborative network is constructed by collaborating an unmanned aerial vehicle platform and ground sensors.

[0049] Optionally, the ground-air collaborative data includes: high-altitude acquisition data and ground acquisition data;

[0050] The high-altitude acquisition data includes one or more of the following: field area distribution data, three-dimensional terrain data, temperature gradient data, gas concentration gradient data, crop type data, disaster feature data;

[0051] The disaster feature data includes one or more of the following: surface deformation data, crack width data, and thermal anomaly data;

[0052] The ground acquisition data includes one or more of the following: soil humidity data, meteorological data, water level change data, soil temperature data, soil pH, soil conductivity data, water quality data, and crop status data;

[0053] The meteorological data includes one or more of the following: temperature data, humidity data, wind speed data, and rainfall data.

[0054] Optionally, the land monitoring system further includes: a map construction module, including:

[0055] An associated data acquisition sub-module for acquiring historical land characteristic data and a Geographic Information System (GIS) database of the area to be monitored;

[0056] A data extraction sub-module for extracting data from the historical land characteristic data to obtain entity information and semantic label information of the historical land characteristic data;

[0057] An embedding processing sub-module for performing embedding processing on the entity information and semantic label information to generate representation features of the historical land characteristic data;

[0058] A feature splicing sub-module for extracting geographic features from the GIS database and performing weighted splicing on the representation features and the geographic features to obtain a comprehensive vector representation;

[0059] A map output sub-module for constructing a knowledge map based on the comprehensive vector representation.

[0060] Optionally, the data storage module includes:

[0061] A hierarchical encryption sub-module for respectively performing hierarchical encryption on the land characteristic data and the semantic label information of the land characteristic data, and outputting encrypted data blocks;

[0062] A spatio-temporal construction sub-module for constructing a spatio-temporal Merkle forest structure based on the encrypted data blocks and outputting a spatio-temporal Merkle root set;

[0063] A target locking sub-module for anchoring the elements in the Merkle root set to a target block in the blockchain for storage according to a pre-set index label through a smart contract.

[0064] Optionally, the disaster prediction module includes:

[0065] A feature mapping sub-module for performing non-linear feature mapping on the land characteristic data and the semantic label information of the land characteristic data to generate a high-dimensional feature space vector;

[0066] A game setting sub-module for treating the stakeholders in the area to be monitored as participants in the game; the stakeholders include: the field chiefs, farmers, and enterprises;

[0067] A strategy selection sub-module for determining the target selection strategies of the respective participants based on the high-dimensional feature vector;

[0068] A revenue calculation sub-module for calculating the revenue values of the target selection strategies of the participants based on the respective participants;

[0069] A strategy determination sub-module for determining the final selection strategy through regression analysis according to the revenue values of the respective participants;

[0070] A probability output sub-module for outputting the disaster occurrence probability of the area to be monitored through Bayesian inference according to the final selection strategy.

[0071] Optionally, the emergency response module includes:

[0072] A priority response sub-module for determining the response priorities of different disaster types according to the disaster occurrence probability of the area to be monitored;

[0073] A contract invocation sub-module for outputting an emergency response plan by invoking the smart contract on the blockchain according to the response priorities of different disaster types;

[0074] An event listening sub-module for generating multi-mode response instructions for the area to be monitored by using the event listening mechanism of the smart contract;

[0075] Wherein, the multi-mode response instructions include one or more of the following: SMS warning, UAV dispatching and inspection path, broadcast information sending, and emergency channel response instruction.

[0076] On the other hand, the present invention also provides an electronic device, including: at least one processor and a memory; the memory and the processor are connected by a bus;

[0077] The memory is used for storing one or more programs;

[0078] When the one or more programs are executed by the at least one processor, the above-mentioned land monitoring method based on a knowledge graph is implemented.

[0079] On the other hand, the present invention also provides a computer-readable storage medium with an execution program stored thereon, and when the execution program is executed, the above-mentioned land monitoring method based on a knowledge graph is implemented.

[0080] Compared with the prior art, the beneficial effects of the present invention are:

[0081] The present invention provides a land monitoring method, system, device and medium based on a knowledge graph, including: obtaining land characteristic data of an area to be monitored; using the pre-constructed knowledge graph to perform semantic tagging on the land characteristic data to obtain semantic tag information of the land characteristic data, and storing the land characteristic data and the semantic tag information of the land characteristic data on a blockchain; based on the land characteristic data and the semantic tag information of the land characteristic data, using a dynamic game algorithm to perform disaster prediction on the area to be monitored to obtain the disaster occurrence probability of the area to be monitored; according to the disaster occurrence probability of the area to be monitored, through the smart contract on the blockchain, performing an emergency response on the area to be monitored to generate a multi-mode response instruction for the area to be monitored; by using the knowledge graph to perform semantic tagging on the land characteristic data, the present application can convert complex geographical and environmental information into clear structured data, which is beneficial to identifying key features and potential risks; by using the dynamic game algorithm for disaster prediction, the prediction can timely reflect environmental changes and improve the ability to identify potential disaster risks; therefore, through the method of the present invention, not only the real-time performance and accuracy of land monitoring are improved, but also efficient disaster prediction and an automated emergency response mechanism are realized, thereby significantly enhancing the ability to prevent and respond to natural disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a schematic flowchart of a land monitoring method based on a knowledge graph provided by the present invention;

[0083] Figure 2 It is a schematic flowchart of blockchain storage in a land monitoring method based on a knowledge graph provided by the present invention;

[0084] Figure 3 It is a schematic flowchart of disaster prediction in a land monitoring method based on a knowledge graph provided by the present invention;

[0085] Figure 4 It is a schematic diagram of the structural composition of a land monitoring system based on a knowledge graph provided by the present invention;

[0086] Figure 5 It is a schematic diagram of the structure of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0087] The present invention proposes a land monitoring method, system, device and medium based on a knowledge graph. The following further details the specific embodiments of the present invention with reference to the accompanying drawings.

[0088] Embodiment 1:

[0089] The present invention provides a land monitoring method based on a knowledge graph. The schematic flowchart is asFigure 1 As shown in, it includes:

[0090] Step 1: Obtain the land characteristic data of the area to be monitored;

[0091] Step 2: Use the pre-constructed knowledge graph to perform semantic tagging on the land characteristic data to obtain the semantic tag information of the land characteristic data, and store the land characteristic data and the semantic tag information of the land characteristic data on the blockchain;

[0092] Step 3: Based on the land characteristic data and the semantic tag information of the land characteristic data, adopt a dynamic game algorithm to predict disasters in the area to be monitored, and obtain the disaster occurrence probability of the area to be monitored;

[0093] Step 4: According to the disaster occurrence probability of the area to be monitored, perform an emergency response on the area to be monitored through the smart contract on the blockchain, and generate multi-mode response instructions for the area to be monitored.

[0094] Generally, in the prior art, the monitoring of land usually relies on multiple data sources, including satellite remote sensing data, data collected by ground sensors, historical land use records, climate and environmental data, etc. These data provide basic information for the evaluation and change analysis of land characteristics. However, there are many problems in dealing with sudden natural disasters. First, data acquisition often has a time delay and insufficient real-time performance, which makes the relevant data not updated in time when natural disasters occur, resulting in lagging decision-making information. Second, the heterogeneity and fragmentation of data make it difficult to effectively integrate and analyze data from different sources during disasters, reducing the accuracy and efficiency of dealing with emergencies. In response to this problem, this application adopts a brand-new method to obtain the land characteristic data of the area to be monitored. Specifically:

[0095] In one implementation, the process of obtaining the land characteristic data of the area to be monitored in the above Step 1 may include:

[0096] Collect the ground-air collaborative data of the area to be monitored by using a hybrid collaborative network;

[0097] Extract the features of the ground-air collaborative data through an environmental perception algorithm to obtain the land characteristic data of the area to be monitored;

[0098] Among them, the hybrid collaborative network is constructed by collaborating with an unmanned aerial vehicle platform and ground sensors.

[0099] Exemplarily, the above-mentioned ground-air collaborative data may include: high-altitude acquisition data and ground acquisition data (which can be collected by ground sensors or Internet of Things sensing devices);

[0100] High-altitude collected data may include one or more of the following: field area distribution data, three-dimensional terrain data, temperature gradient data, gas concentration gradient data, crop type data, disaster characteristic data;

[0101] Disaster characteristic data may include one or more of the following: surface deformation data, crack width data, and thermal anomaly data;

[0102] Ground collected data may include one or more of the following: soil moisture data, meteorological data, water level change data, soil temperature data, soil pH, soil conductivity data, water quality data, crop status data;

[0103] Meteorological data may include one or more of the following: temperature data, humidity data, wind speed data, and rainfall data;

[0104] In this implementation, by constructing a hybrid collaborative network to collect ground-air collaborative data and combining it with an environmental perception algorithm to extract land characteristic data, the accuracy and efficiency of land monitoring can be significantly improved. The hybrid collaborative network is based on the collaborative work of an unmanned aerial vehicle (UAV) platform and ground sensors, enabling multi-dimensional and multi-level data collection of the monitoring area, covering all-round information from high altitude to the ground. High-altitude collected data such as field area distribution, three-dimensional terrain, and temperature gradient can provide macroscopic environmental information, while ground-collected data such as soil moisture, meteorological data, and water quality provide microscopic environmental details. This multi-level data fusion makes the acquisition of land characteristic data more comprehensive and accurate. At the same time, the introduction of the environmental perception algorithm can extract key land characteristics from complex ground-air collaborative data, such as disaster characteristics and crop status, providing a scientific basis for land management and disaster warning. This implementation method can not only monitor land changes in real time but also predict potential environmental risks, thus providing strong technical support for agricultural production, disaster prevention, and ecological protection. In this implementation, although the UAV platform, ground sensors, and environmental perception algorithm used are applied in the existing technology, the organic combination and application of these technical means in a specific land monitoring scenario are not obvious. Moreover, the collaborative work of the UAV platform and ground sensors is not a simple technical superposition but requires solving a series of technical problems such as data synchronization, transmission, and processing. For example, there are differences in time, space, and accuracy between the data collected by the UAV at high altitude and the data collected by the ground sensors. How to effectively fuse and collaboratively analyze these data is a complex technical problem. In addition, when the environmental perception algorithm extracts land characteristic data, it needs to be optimized and adjusted for specific ground-air collaborative data to adapt to the land characteristics of different regions. This targeted algorithm optimization is not a general solution in the existing technology but requires customized development according to specific monitoring scenarios. Therefore, in the specific implementation process of this implementation, it involves the collaborative work of multiple technical links, solves the problems that are difficult to overcome in the actual application of the existing technology, and thus achieves significant technological progress in the field of land monitoring. Therefore, this implementation method not only improves the accuracy and efficiency of land monitoring through multi-level data fusion and algorithm optimization but also realizes efficient application in a specific scenario by solving technical problems such as data collaboration and algorithm adaptation. In addition to being collected through the UAV platform, the above-mentioned high-altitude collected data can also accurately obtain the land characteristic data of the area to be monitored through satellite positioning technology and realize transparent recording and full-process traceability of the data in combination with blockchain technology, which can effectively ensure the accuracy and credibility of the data.

[0105] Through the application of a hybrid collaborative network, rich ground-air collaborative data of the area to be monitored can be obtained in real time, and these data can be deeply analyzed using environmental perception algorithms to extract key land feature information. Based on the extracted land feature information, historical land feature data and GIS databases can be further considered. By means of data extraction and feature embedding, a comprehensive knowledge graph can be constructed to support deeper analysis and intelligent decision-making for the area to be monitored. Specifically:

[0106] In one implementation manner, the process of pre-constructing the knowledge graph in step 2 above may include:

[0107] Obtain historical land feature data and a Geographic Information System (GIS) database of the area to be monitored;

[0108] Perform data extraction on the historical land feature data to obtain entity information and semantic label information of the historical land feature data;

[0109] Perform embedding processing on the entity information and semantic label information of the historical land feature data to generate representation features of the historical land feature data;

[0110] Extract geographical features from the GIS database, and perform weighted splicing on the representation features and geographical features to obtain a comprehensive vector representation;

[0111] Construct a knowledge graph based on the comprehensive vector representation;

[0112] In this implementation, by obtaining the historical land characteristic data and GIS database of the area to be monitored, and combining technical means such as data extraction, embedding processing, and feature weighted splicing, a knowledge graph is constructed, which can significantly improve the intelligence and accuracy of land monitoring. For example, by extracting and embedding the entity information and semantic label information of historical land characteristic data, complex land characteristic data can be transformed into a structured representation feature, providing a high-quality data basis for subsequent analysis and applications. For example, extracting geographical features from the GIS database and performing weighted splicing with the representation features of land characteristic data to generate a comprehensive vector representation. This process not only integrates multi-dimensional information of land characteristics but also combines geospatial data, making the construction of the knowledge graph more comprehensive and accurate. For example, the knowledge graph constructed based on the comprehensive vector representation can intuitively display the correlation between land characteristics and the geographical environment, providing a scientific basis and decision-making support for application scenarios such as land management, disaster warning, and ecological protection. Although the technical means such as data extraction, embedding processing, and feature weighted splicing used in this implementation are not the latest in the existing technology, it is not obvious to organically combine and apply these technical means to the construction of the knowledge graph in the field of land monitoring. For example, the extraction of entity information and semantic label information of historical land characteristic data requires customized design according to the characteristics of the land monitoring field to ensure the accuracy and practicality of the extraction results. At the same time, performing weighted splicing of the representation features of land characteristic data and geographical features in the GIS database requires solving the problems of feature alignment and weight allocation between different data sources, which involves complex algorithm design and parameter optimization. In addition, the construction of the knowledge graph needs to consider the dynamic change relationship between land characteristics and the geographical environment, which puts higher requirements on the real-time nature of data and the adaptability of the model. Therefore, in the specific implementation process of this implementation, it is not only necessary to solve the problems that are difficult to overcome in the actual application of the existing technology but also to optimize the technology according to the specific needs of the land monitoring field, thus achieving efficient application in a specific scenario.

[0113] Completing the construction of the knowledge graph through the above steps will provide more comprehensive and accurate data support for disaster prediction and emergency response. In this regard, the land characteristic data and its semantic label information will be stored on the blockchain to ensure the security and reliability of the data. Specifically:

[0114] Such as Figure 2 shown, in one implementation, the process of storing the land characteristic data and the semantic label information of the land characteristic data on the blockchain can include:

[0115] Encrypt the land characteristic data and the semantic label information of the land characteristic data layer by layer and output encrypted data blocks;

[0116] Construct a spatio-temporal Merkle forest structure based on encrypted data blocks and output a set of spatio-temporal Merkle roots;

[0117] Through a smart contract, anchor the elements in the Merkle root set to the target block in the blockchain for storage according to the pre-set index tags;

[0118] In this implementation method, by performing hierarchical encryption on land characteristic data and its semantic tag information, constructing a spatio-temporal Merkle forest structure based on the encrypted data blocks, and finally anchoring the Merkle root set to the target block for storage through a smart contract, it can significantly improve the security, traceability, and management efficiency of data storage. The application of the hierarchical encryption technology in this implementation method ensures the security of land characteristic data and its semantic tag information during storage and transmission, preventing data from being tampered with or leaked; the construction of the spatio-temporal Merkle forest structure can not only efficiently organize and verify large-scale data, but also achieve spatio-temporal correlated storage of data through the Merkle root set, thus providing reliable technical support for the traceability and analysis of land monitoring data. Finally, anchoring the Merkle root set to the blockchain through a smart contract can further enhance the immutability and transparency of the data, while realizing the automated and intelligent management of data storage; this implementation method can effectively solve the security and credibility problems faced by land monitoring data during storage and management, providing a solid data foundation for the scientific management and decision-making of land resources. Although technical means such as hierarchical encryption, Merkle tree structure, and smart contract used in this implementation method have been applied in the prior art, it is not obvious to organically combine and apply these technical means to the blockchain storage scenario of land monitoring data. For example, the hierarchical encryption technology needs to design targeted encryption strategies according to the characteristics of land characteristic data and its semantic tag information to ensure the security and availability of data at different levels. At the same time, the construction of the spatio-temporal Merkle forest structure needs to solve the spatio-temporal correlation problem of large-scale data, which puts higher requirements on the optimization of data structure and the efficiency of algorithms. And when the smart contract anchors the Merkle root set to the blockchain, it needs to design reasonable index tags and storage rules according to the specific requirements of land monitoring data to achieve efficient management and rapid retrieval of data. The combination and optimization of these technical links are customized development for the specific requirements of the prior art in the land monitoring field, thus realizing efficient application in a specific scenario. Therefore, this implementation method improves the security, traceability, and management efficiency of land monitoring data storage through the combination of technical means such as hierarchical encryption, spatio-temporal Merkle forest structure, and smart contract.

[0119] Regarding the blockchain storage process of land characteristic data and its semantic tag information, it can effectively guarantee the security and credibility of the data. Based on these data, a dynamic game algorithm can be further adopted to predict disasters in the area to be monitored to evaluate the probability of disasters occurring. Specifically:

[0120] As Figure 3 shown, in one implementation, in the above step 3, based on the land characteristic data and the semantic label information of the land characteristic data, the process of using the dynamic game algorithm to predict disasters in the area to be monitored and obtaining the disaster occurrence probability of the area to be monitored may include:

[0121] Perform a non-linear feature mapping on the land characteristic data and the semantic label information of the land characteristic data to generate a high-dimensional feature space vector;

[0122] Regard the stakeholders in the area to be monitored as the participants in the game; the stakeholders include: the field chiefs, farmers, and enterprises;

[0123] Based on the high-dimensional feature vector, determine the target selection strategies of each participant;

[0124] Based on each participant, calculate the payoff value of the target selection strategy of the participant;

[0125] According to the payoff values of each participant, determine the final selection strategy through regression analysis;

[0126] According to the final selection strategy, through Bayesian inference, output the disaster occurrence probability of the area to be monitored;

[0127] Exemplarily, the calculation formula of the payoff value of the above target selection strategy may be as follows:

[0128] R l (S l ) = U(S l ) - C(S l ) + γ·Risk(S l ) + δ·Innovation(S l );

[0129] Wherein, R l (S l ) represents the payoff value of participant l when selecting strategy S l ; U(S l ) represents the utility value of strategy S l , indicating the direct economic benefits brought by this strategy (such as reducing losses and increasing production); C(S l ) represents the cost of implementing strategy S l (which may include direct expenditures and potential opportunity costs); Risk(S l ) represents the risk assessment of strategy S l (such as financial risk and reputation risk under natural disaster conditions); γ represents the risk weight; Innovation(S l ) represents the implementation of strategy S lThe possible innovative value; δ represents the value weight; in this example, by introducing the calculation of risk assessment and innovative value, not only can the comprehensiveness of revenue calculation be improved, but also participants can be motivated to take more proactive response measures under the influence of rapidly changing natural disasters. This framework not only covers direct economic benefits, but also helps to form long-term strategic thinking, promotes the application of new technologies and new management strategies in the industry, thereby enhancing the resilience and efficiency of the overall system.

[0130] For example, the expression of the above-mentioned disaster occurrence probability can be as follows:

[0131]

[0132] Among them, P(D|X) represents the posterior probability of the occurrence of disaster D given the feature set X; P(x i |D) represents the conditional probability of the i-th feature x i (such as soil humidity) when the disaster D occurs; i = 1…n; n represents the total number of features; ω i represents the weight of the feature x i , indicating the relative importance of this feature in the prediction; P(D) represents the prior probability of disaster D, that is, the probability of the occurrence of this disaster without any feature information; P(X) represents the total probability of observing the feature set X, which can be deduced from the distribution of all features; in this example, introducing the feature weight ω i makes important features (such as rainfall) account for a larger proportion in the model, thereby improving the accuracy of prediction. This makes the prediction more able to reflect the actual situation; and by combining the conditional probabilities of multiple features through the multiplication rule, this method allows handling the interdependence between features. Improve the adaptability of the model to complex environmental changes and enhance the ability to identify potential disasters.

[0133] In this implementation, by performing non-linear feature mapping on land characteristic data and its semantic label information to generate high-dimensional feature space vectors, and combining with a dynamic game algorithm to predict the disaster occurrence probability of the area to be monitored, the accuracy and practicality of disaster prediction can be significantly improved. For example, through non-linear feature mapping technology, high-dimensional features can be extracted from complex land characteristic data, capturing the non-linear relationships hidden in the data, thereby providing richer and more accurate feature information for subsequent disaster prediction. Secondly, regarding the stakeholders in the area to be monitored (such as field chiefs, farmers, and enterprises) as game participants, and determining the target selection strategies of each participant based on high-dimensional feature vectors, can fully consider the behaviors and decision-making impacts of different stakeholders in disaster prediction, making the prediction results closer to the actual scenario. By calculating the payoff values of each participant and combining regression analysis to determine the final selection strategy, the interests of all parties can be effectively balanced and the decision-making process optimized. Finally, based on Bayesian inference to output the disaster occurrence probability can not only quantify the prediction results but also dynamically update the prediction model to adapt to changes in land characteristic data. This implementation can provide a scientific basis for disaster early warning and risk management, helping stakeholders formulate more reasonable disaster prevention and mitigation strategies. Although the technical means such as non-linear feature mapping, game algorithms, and Bayesian inference used in this implementation are not new technical contents in the existing technology, it is not common to organically combine these technical means and apply them to the disaster prediction scenario in the land monitoring field. For example, non-linear feature mapping requires designing a specific mapping function according to the characteristics of land characteristic data to capture the key features in the data, and this process requires in-depth association with the data characteristics in the land monitoring field. At the same time, introducing stakeholders into the game model and determining their target selection strategies requires comprehensively considering the behavior patterns of different participants and their impacts on disaster prediction, which poses higher requirements for model construction and parameter optimization. In addition, calculating the disaster occurrence probability based on Bayesian inference requires dynamic updating in combination with real-time data to adapt to changes in land characteristic data, which poses challenges to the efficiency and adaptability of the algorithm. Therefore, this implementation improves the accuracy and practicality of disaster prediction through the combination of technical means such as non-linear feature mapping, dynamic game algorithm, and Bayesian inference.

[0134] By performing disaster prediction on the area to be monitored through the above steps, the probability of disaster occurrence can be effectively calculated. On this basis, further emergency response operations can be considered. For example, according to the disaster occurrence probability of the area to be monitored, multi-mode response instructions for this area can be generated through smart contracts on the blockchain. Specifically:

[0135] In one implementation, the process of generating multi-mode response instructions for the area to be monitored through smart contracts on the blockchain according to the disaster occurrence probability of the area to be monitored in step 4 above may include:

[0136] Determine the response priorities for different disaster types according to the disaster occurrence probabilities in the area to be monitored;

[0137] Output an emergency response plan by calling the smart contract on the blockchain according to the response priorities for different disaster types;

[0138] Generate multi-mode response instructions for the area to be monitored by using the event listening mechanism of the smart contract;

[0139] Among them, the multi-mode response instructions may include one or more of the following: SMS warning, UAV dispatching and inspection path, broadcast information sending, and emergency channel response instructions;

[0140] In this implementation, by determining the response priority according to the disaster occurrence probability of the area to be monitored and using smart contracts on the blockchain to generate multi-mode response instructions, the efficiency and accuracy of emergency response can be significantly improved. For example, determining the response priorities for different disaster types based on the disaster occurrence probability can ensure that limited emergency resources are preferentially allocated to the areas most in need of response, thereby enhancing the pertinence and effectiveness of emergency response. By invoking smart contracts on the blockchain to output an emergency response plan, not only can the automation and transparency of the response process be achieved, but also the immutability and traceability of the response plan can be ensured, enhancing the credibility of emergency response. In addition, by using the event listening mechanism of smart contracts to generate multi-mode response instructions (such as SMS warnings, drone dispatching and inspection paths, broadcast message sending, and emergency channel response instructions), the response strategy can be dynamically adjusted according to the disaster type and priority, enabling multi-dimensional and multi-level emergency response. This implementation can significantly shorten the emergency response time, reduce disaster losses, and provide a scientific and intelligent solution for disaster management. The organic combination and application of technical means such as priority determination, smart contract invocation, and event listening mechanism in this implementation in the emergency response scenario of land monitoring can demonstrate unique technical effects. For example, determining the disaster response priority requires comprehensive consideration of various factors such as disaster occurrence probability, regional vulnerability, and emergency resource distribution, which poses higher requirements for algorithm design and parameter optimization. At the same time, when a smart contract outputs an emergency response plan, reasonable contract logic needs to be designed according to the specific disaster type and priority to ensure the accuracy and executability of the response plan. In addition, using the event listening mechanism of smart contracts to generate multi-mode response instructions requires solving the coordination problem between different response modes. For example, how to ensure the coordination of SMS warnings, drone dispatching, and broadcast message sending in terms of time and space. The combination and optimization of these technical links achieve efficient application in specific scenarios. Therefore, this implementation improves the efficiency and accuracy of emergency response through the combination of technical means such as disaster priority determination, smart contract invocation, and event listening mechanism, and at the same time provides a more efficient and intelligent emergency response solution for the land monitoring field by solving problems such as priority assessment, contract logic design, and multi-mode coordination.

[0141] In summary, the present invention addresses the problems of insufficient real-time performance and difficult data integration in the current land monitoring and emergency management processes. Especially when natural disasters occur, the land characteristics change rapidly, and traditional land monitoring methods cannot keep up in a timely manner, resulting in a lag in disaster response and the inability to effectively prevent and reduce losses. The present invention proposes a land monitoring method based on a knowledge graph. By systematically acquiring land characteristic data of the area to be monitored and using the constructed knowledge graph for semantic tagging, in-depth understanding and effective processing of the data are achieved. Further, a dynamic game algorithm is adopted for disaster prediction, which can not only accurately evaluate the probability of a disaster occurring but also consider the decision-making behaviors of different stakeholders, thereby improving the scientificity and feasibility of decision-making. Combining with smart contracts on the blockchain to generate multi-mode emergency response instructions enables rapid response and flexible adjustment of emergency management measures. Therefore, through the combination of modern data technologies and intelligent algorithms, this application provides a brand-new solution for land monitoring and disaster response, significantly improving the prediction ability of natural disasters and the emergency response efficiency, facilitating the reduction of potential losses, and enhancing the society's ability to prevent disaster events.

[0142] Embodiment 2:

[0143] Based on the same inventive concept, the present invention also provides a land monitoring system based on a knowledge graph. The schematic structural composition diagram is as Figure 4 shown, including:

[0144] A data acquisition module for acquiring land characteristic data of the area to be monitored;

[0145] A data storage module for semantically tagging the land characteristic data by using a pre-constructed knowledge graph to obtain semantic tag information of the land characteristic data, and storing the land characteristic data and the semantic tag information of the land characteristic data on the blockchain;

[0146] A disaster prediction module for predicting disasters in the area to be monitored by using a dynamic game algorithm based on the land characteristic data and the semantic tag information of the land characteristic data to obtain the probability of a disaster occurring in the area to be monitored;

[0147] An emergency response module for performing an emergency response on the area to be monitored through a smart contract on the blockchain according to the probability of a disaster occurring in the area to be monitored, and generating multi-mode response instructions for the area to be monitored.

[0148] In one implementation, the above data acquisition module specifically includes:

[0149] A data collection sub-module for collecting ground-air collaborative data of the area to be monitored by using a hybrid collaborative network;

[0150] A feature extraction sub-module, which is used to extract features from the ground-air collaborative data through an environmental perception algorithm to obtain the land characteristic data of the area to be monitored;

[0151] Among them, the hybrid collaborative network is collaboratively constructed based on an unmanned aerial vehicle platform and ground sensors.

[0152] Exemplarily, the above-mentioned ground-air collaborative data may include: high-altitude acquisition data and ground acquisition data;

[0153] The high-altitude acquisition data may include one or more of the following: field area distribution data, three-dimensional terrain data, temperature gradient data, gas concentration gradient data, crop type data, disaster characteristic data;

[0154] The disaster characteristic data may include one or more of the following: surface deformation data, crack width data, and thermal anomaly data;

[0155] The ground acquisition data may include one or more of the following: soil moisture data, meteorological data, water level change data, soil temperature data, soil pH value, soil conductivity data, water quality data, crop status data;

[0156] The meteorological data may include one or more of the following: temperature data, humidity data, wind speed data, and rainfall data.

[0157] In one implementation, the above-mentioned land monitoring system may further include: a map construction module, specifically including:

[0158] An associated data acquisition sub-module, which is used to acquire the historical land characteristic data of the area to be monitored and a geographic information system (GIS) database;

[0159] A data extraction sub-module, which is used to extract data from the historical land characteristic data to obtain the entity information and semantic label information of the historical land characteristic data;

[0160] An embedding processing sub-module, which is used to perform embedding processing on the entity information and semantic label information of the historical land characteristic data to generate the representation features of the historical land characteristic data;

[0161] A feature splicing sub-module, which is used to extract geographic features from the geographic information system (GIS) database and perform weighted splicing on the representation features and the geographic features to obtain a comprehensive vector representation;

[0162] A map output sub-module, which is used to construct a knowledge map according to the comprehensive vector representation.

[0163] In one implementation, the above-mentioned data storage module may include:

[0164] The hierarchical encryption sub-module is used to perform hierarchical encryption on the land characteristic data and the semantic tag information of the land characteristic data respectively, and output encrypted data blocks;

[0165] The spatio-temporal construction sub-module is used to construct a spatio-temporal Merkle forest structure based on the encrypted data blocks, and output a spatio-temporal Merkle root set;

[0166] The target locking sub-module is used to anchor the elements in the Merkle root set to the target block in the blockchain for storage according to the pre-set index tags through a smart contract.

[0167] In one implementation, the above disaster prediction module may include:

[0168] The feature mapping sub-module is used to perform non-linear feature mapping on the land characteristic data and the semantic tag information of the land characteristic data, and generate high-dimensional feature space vectors;

[0169] The game setting sub-module is used to regard the stakeholders in the area to be monitored as participants in the game; the stakeholders include: the field chiefs, farmers and enterprises;

[0170] The strategy selection sub-module is used to determine the target selection strategies of each participant based on the high-dimensional feature vectors;

[0171] The revenue calculation sub-module is used to calculate the revenue values of the target selection strategies of the participants based on each participant;

[0172] The strategy determination sub-module is used to determine the final selection strategy through regression analysis according to the revenue values of each participant;

[0173] The probability output sub-module is used to output the disaster occurrence probability of the area to be monitored through Bayesian inference according to the final selection strategy.

[0174] In one implementation, the above emergency response module may include:

[0175] The priority response sub-module is used to determine the response priorities of different disaster types according to the disaster occurrence probability of the area to be monitored;

[0176] The contract call sub-module is used to output an emergency response plan by calling the smart contract on the blockchain according to the response priorities of different disaster types;

[0177] The event listening sub-module is used to generate multi-mode response instructions for the area to be monitored by using the event listening mechanism of the smart contract;

[0178] Among them, the multi-mode response instructions include one or more of the following: SMS warning, UAV dispatching and inspection path, broadcast information sending, and emergency channel response instruction.

[0179] Example 3:

[0180] As Figure 5 shown, the present invention also provides an electronic device, which may be a computer device, a single-chip microcomputer device, a smart mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected through a bus; the memory can be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, and this data can be called and / or modified when the instructions are executed.

[0181] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a land monitoring method based on a knowledge graph in the above embodiment.

[0182] Example 4:

[0183] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device-readable storage medium (Memory). The electronic device-readable storage medium is a memory device in the electronic device and is used to store programs and data. It can be understood that the storage medium here can include both the built-in storage medium in the electronic device and, of course, the extended storage medium supported by the electronic device. The storage medium provides a storage space, and the operating system of the terminal is stored in this storage space. And, one or more instructions suitable for being loaded and executed by the processor are also stored in this storage space, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. By loading and executing one or more instructions stored in the storage medium by the processor, the steps of a land monitoring method based on a knowledge graph in the above embodiment can be implemented.

[0184] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0185] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0186] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realize the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0187] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that after reading the present invention, various changes, modifications, or equivalent replacements can still be made to the specific implementation manners of the application. However, these changes, modifications, or equivalent replacements are all within the scope of the protection of the claims pending for approval of the application.

Claims

1. A land monitoring method based on a knowledge graph, characterized in that, Including: Obtain the land characteristic data of the area to be monitored; Use the pre-constructed knowledge graph to perform semantic tagging on the land characteristic data, obtain the semantic tag information of the land characteristic data, and store the land characteristic data and the semantic tag information of the land characteristic data on the blockchain; Based on the land characteristic data and the semantic tag information of the land characteristic data, adopt a dynamic game algorithm to predict disasters in the area to be monitored, and obtain the disaster occurrence probability of the area to be monitored; According to the disaster occurrence probability of the area to be monitored, perform an emergency response on the area to be monitored through the smart contract on the blockchain, and generate a multi-mode response instruction for the area to be monitored.

2. The method according to claim 1, wherein The obtaining of the land characteristic data of the area to be monitored includes: Use a hybrid collaborative network to collect the ground-air collaborative data of the area to be monitored; Extract features from the ground-air collaborative data through an environmental perception algorithm to obtain the land characteristic data of the area to be monitored; Among them, the hybrid collaborative network is constructed by collaborating an unmanned aerial vehicle platform and ground sensors.

3. The method according to claim 2, characterized in that, The ground-air collaborative data includes: high-altitude acquisition data and ground acquisition data; The high-altitude acquisition data includes one or more of the following: field area distribution data, three-dimensional terrain data, temperature gradient data, gas concentration gradient data, crop type data, disaster characteristic data; The disaster characteristic data includes one or more of the following: surface deformation data, crack width data, and thermal anomaly data; The ground acquisition data includes one or more of the following: soil humidity data, meteorological data, water level change data, soil temperature data, soil pH, soil conductivity data, water quality data, crop status data; The meteorological data includes one or more of the following: temperature data, humidity data, wind speed data, and rainfall data.

4. The method according to any one of claims 1 to 3, characterized in that The pre-construction process of the knowledge graph includes: Obtain the historical land characteristic data of the area to be monitored and the Geographic Information System (GIS) database; Perform data extraction on the historical land characteristic data to obtain the entity information and semantic tag information of the historical land characteristic data; Perform embedding processing on the entity information and semantic tag information of the historical land characteristic data to generate the representation features of the historical land characteristic data; Extract geographical features from the Geographic Information System (GIS) database, and perform weighted splicing on the representation features and the geographical features to obtain a comprehensive vector representation; Construct a knowledge graph according to the comprehensive vector representation.

5. The method according to claim 1, characterized in that, The storing of the land characteristic data and the semantic tag information of the land characteristic data on the blockchain includes: Perform hierarchical encryption on the land characteristic data and the semantic tag information of the land characteristic data respectively, and output encrypted data blocks; Based on the encrypted data blocks, construct a spatio-temporal Merkle forest structure, and output a spatio-temporal Merkle root set; Through a smart contract, anchor the elements in the Merkle root set to the target block in the blockchain for storage according to a pre-set index tag.

6. The method according to claim 1, characterized in that, Based on the land characteristic data and the semantic tag information of the land characteristic data, a dynamic game algorithm is used to predict disasters in the area to be monitored, and the disaster occurrence probability of the area to be monitored is obtained, including: Perform non-linear feature mapping on the land characteristic data and the semantic tag information of the land characteristic data to generate a high-dimensional feature space vector; Regard the stakeholders in the area to be monitored as participants in the game; the stakeholders include: the head of the field, farmers, and enterprises; Based on the high-dimensional feature vector, determine the target selection strategies of the participants; Based on each participant, calculate the payoff value of the target selection strategy of the participant; According to the payoff values of the participants, determine the final selection strategy through regression analysis; According to the final selection strategy, output the disaster occurrence probability of the area to be monitored through Bayesian inference.

7. The method according to claim 1, characterized in that According to the disaster occurrence probability of the area to be monitored, perform an emergency response on the area to be monitored through the smart contract on the blockchain, and generate a multi-mode response instruction for the area to be monitored, including: According to the disaster occurrence probability of the area to be monitored, determine the response priorities for different disaster types; According to the response priorities for different disaster types, call the smart contract on the blockchain to output an emergency response plan; Use the event listening mechanism of the smart contract to generate a multi-mode response instruction for the area to be monitored; Among them, the multi-mode response instruction includes one or more of the following: SMS warning, UAV dispatching inspection path, broadcast information sending, and emergency channel response instruction.

8. A land monitoring system based on a knowledge graph, characterized in that, Including: A data acquisition module for acquiring land characteristic data of the area to be monitored; A data storage module for semantically marking the land characteristic data by using a pre-constructed knowledge graph to obtain the semantic tag information of the land characteristic data, and storing the land characteristic data and the semantic tag information of the land characteristic data on the blockchain; A disaster prediction module for predicting disasters in the area to be monitored based on the land characteristic data and the semantic tag information of the land characteristic data by using a dynamic game algorithm, and obtaining the disaster occurrence probability of the area to be monitored; An emergency response module for performing an emergency response on the area to be monitored through the smart contract on the blockchain according to the disaster occurrence probability of the area to be monitored, and generating a multi-mode response instruction for the area to be monitored.

9. An electronic device, characterized in that, Including: At least one processor and a memory; The memory and the processor are connected by a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, a land monitoring method based on a knowledge graph as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that, There is an execution program stored thereon, and when the execution program is executed, a land monitoring method based on a knowledge graph as described in any one of claims 1 to 7 is implemented.

Images