Land resource management method and system based on big data analysis

Abstract

The invention relates to the technical field of land resource management, in particular to a land resource management method and system based on big data analysis. The method comprises the following steps: acquiring multi-source land resource acquisition data; carrying out asynchronous parallel data acquisition on the multi-source land resource acquisition data so as to generate multi-source land resource real-time acquisition data; performing request-response optimization on the multi-source land resource real-time acquisition data to generate a self-optimization transmission parameter set; performing precision error correction on the standard multi-source land resource real-time data by using the self-optimization transmission parameter set to generate land resource correction data; performing nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhanced feature matrix; and performing dimension compression processing on the nonlinear enhanced feature matrix to generate a multi-dimensional spatial-temporal feature matrix. According to the invention, the comprehensiveness and accuracy of land resource management are improved by improving data acquisition, processing, prediction and planning methods.

Description

Technical Field

[0001] The present invention relates to the technical field of land resource management, and in particular to a land resource management method and system based on big data analysis. Background Art

[0002] Traditional land resource management methods mainly rely on manual mapping, data collection and empirical judgment. There are problems such as delayed information update, low spatial resolution, and difficulty in data integration, which cannot meet the needs of modern refined management. Entering the 21st century, with the widespread application of remote sensing technology, geographic information system (GIS) and global positioning system (GPS), the acquisition and processing capabilities of land resource data have been significantly improved, prompting land management to gradually develop in the direction of informatization and intelligence. In recent years, the rapid development of big data technology has brought new opportunities for land resource management. By collecting, storing and analyzing multi-source data such as massive spatial data, environmental data, and economic data, managers can achieve comprehensive evaluation and dynamic monitoring of factors such as land use patterns, soil quality, and climate change. Combined with machine learning and artificial intelligence technology, land resource management further realizes intelligent decision support and prediction, and improves land use efficiency. However, the current traditional land resource management data sources are wide-ranging, including satellite images, geographic information system (GIS) data, environmental sensors, etc., but the data acquisition process is complex and the frequency is low, resulting in delayed data updates and failure to reflect the current land use status in real time. At the same time, land resources have significant spatial heterogeneity. Existing big data analysis methods are difficult to fully capture the spatial laws of land use changes when dealing with complex factors such as different terrains, climates, and vegetation, which leads to low comprehensiveness and accuracy of land resource management. Summary of the invention

[0003] Based on this, it is necessary to provide a land resource management method and system based on big data analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a land resource management method based on big data analysis is provided, the method comprising the following steps:

[0005] Step S1: Acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data to generate multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data to generate a self-optimizing transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimizing transmission parameter set to generate land resource correction data;

[0006] Step S2: Perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data according to the land resource environment data to generate dynamic terrain modeling data;

[0007] Step S3: predicting future land use trends based on dynamic terrain modeling data to generate land use prediction data; visualizing dynamic prediction results of land use prediction data to generate a comprehensive land use prediction report; performing overall land resource planning and management on dynamic terrain modeling data according to the comprehensive land use prediction report to generate an overall land resource planning and management strategy;

[0008] Step S4: Virtually plan and manage the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual planning land resource management data; perform actual planning status analysis on the virtual sensor deployment data to generate actual planning status data; and perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data to generate land resource management feedback data to execute dynamic land resource management operations.

[0009] The present invention accelerates the data acquisition speed through asynchronous parallel data acquisition mode, ensures the real-time update of multi-source land resource data, and adapts to the rapidly changing land environment needs. Request-response optimization and self-optimization transmission parameter set improve the efficiency of data transmission, reduce data loss and delay, and ensure the stability and reliability of data transmission. Through precision error correction, the errors caused by transmission or acquisition process are corrected, and accurate land resource correction data is generated to support high-quality analysis and decision-making. Nonlinear feature enhancement processing improves the details and expressiveness of the data, so that the data can more accurately reflect the complex characteristics of land resources. Through dimensional compression processing, the data structure is simplified, the calculation and storage burden is reduced, and important spatial and temporal features are retained, thereby improving data processing efficiency. Terrain environment analysis based on multidimensional spatiotemporal feature matrix provides detailed environmental data and supports in-depth understanding and analysis of complex terrain. Multidimensional dynamic modeling enhances the responsiveness to terrain changes, generates dynamic terrain modeling data, and helps to more accurately predict land use changes and environmental impacts. Land use trend prediction based on dynamic terrain modeling data provides a scientific prediction of future land use changes and supports forward-looking planning decisions. The visualization of dynamic prediction results generates a comprehensive land use prediction report, making the prediction results more intuitive and easy to understand, providing strong support for decision-making. The global planning management strategy formulated according to the comprehensive prediction report optimizes the allocation and utilization of land resources and improves the overall efficiency of resource management. The comparative analysis of virtual planning management data and actual planning status data enhances the consistency and coordination between virtual planning and actual application. Through the feedback adjustment of virtual planning and actual planning status, dynamic optimization of land resource management is achieved, and the flexibility and adaptability of planning implementation are improved. Adaptive land resource management feedback adjustment provides timely management adjustment solutions, effectively responds to problems arising in actual management, and ensures the continuity and effectiveness of dynamic management operations. Therefore, the present invention improves the comprehensiveness and accuracy of land resource management by improving data acquisition, processing, prediction and planning methods.

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

[0011] Step S11: Acquire multi-source land resource collection data;

[0012] Step S12: constructing a distributed data storage channel for multi-source land resource collection data based on the IPFS protocol to obtain a multi-source data channel; using the multi-source data channel to perform asynchronous parallel data acquisition on the multi-source land resource collection data, thereby generating multi-source land resource real-time collection data;

[0013] Step S13: preprocessing the multi-source land resource real-time collected data to generate standard multi-source land resource real-time data, wherein the data preprocessing includes data format unification, data cleaning, data denoising and data standardization;

[0014] Step S14: streaming the standard multi-source land resource real-time data to generate a land resource streaming data packet; edge caching the land resource streaming data packet to generate a land resource distributed cache data set;

[0015] Step S15: Request-response optimization is performed on the land resource distributed cache data set to generate a self-optimizing transmission parameter set; the self-optimizing transmission parameter set is used to perform precision error correction on the standard multi-source land resource real-time data to generate land resource correction data.

[0016] The present invention can effectively disperse and manage the storage of multi-source land resource data through the distributed data storage channel constructed by the IPFS protocol. The asynchronous parallel data acquisition method improves the speed and efficiency of data acquisition and reduces the system load. The generated multi-source land resource real-time acquisition data can achieve efficient data transmission through streaming media transmission technology, so that the land resource data can be updated in a short time, supporting real-time monitoring and analysis. The data preprocessing steps include data format unification, cleaning, denoising and standardization, which ensures the consistency and accuracy of the data and improves the reliability of subsequent analysis and application. Edge caching technology can reduce the delay of data access, improve the access speed and responsiveness of data, especially in large-scale data processing and transmission scenarios. The self-optimizing transmission parameter set generated by request-response optimization helps to improve the accuracy and efficiency of data transmission, reduce errors in the transmission process, and ensure data quality. Accuracy error correction of standard multi-source land resource real-time data can effectively improve the accuracy of the data and support more accurate land resource analysis and decision-making.

[0017] Preferably, step S14 comprises the following steps:

[0018] Step S141: Perform streaming media segmentation processing on the standard multi-source land resource data to generate land resource streaming media segmentation data; perform transmission protocol encapsulation on the land resource streaming media segmentation data to generate a land resource transmission data packet:

[0019] Step S142: perform edge cache preprocessing on the land resource transmission data packet to generate edge cache data segments: perform consistency check on the edge cache data segments to generate cache consistency check data:

[0020] Step S143: Distributed aggregation processing is performed on the edge cache data fragments according to the cache consistency check data to generate a land resource distributed cache data set.

[0021] The present invention can effectively divide the data into small blocks suitable for transmission by performing streaming media slicing processing on standard multi-source land resource data, which reduces the complexity and delay of data transmission. The process of transmission protocol encapsulation ensures that the data can be transmitted safely and reliably in a predetermined format. Edge cache preprocessing can store data fragments in a location close to the user or application, thereby reducing the delay of data access and improving the response speed of the data. This is particularly important for land resource data that needs to be accessed frequently. Performing consistency checks can detect and repair data errors or inconsistencies that occur, ensuring the integrity and accuracy of edge cache data fragments. This is crucial for the reliability and accuracy of data. By performing distributed aggregation processing on cached data fragments, the storage and access strategies of the data can be optimized, making the land resource distributed cache data set more scalable and fault-tolerant. Distributed caching can effectively disperse the load and improve the overall performance of the system. Distributed aggregation processing helps to integrate data from different data sources and nodes, reduce data loss and errors, and improve data integrity and consistency. This processing method supports distributed management and storage of large-scale data, and can easily cope with the growth of data volume and system expansion requirements.

[0022] Preferably, step S2 comprises the following steps:

[0023] Step S21: performing multi-scale segmentation processing on the land resource correction data to generate multi-scale spatiotemporal data blocks, wherein the multi-scale segmentation includes geographic resolution segmentation and time scale segmentation; performing spatial feature extraction on the multi-scale spatiotemporal data blocks to generate a land resource spatial feature matrix;

[0024] Step S22: Perform time series correlation analysis on the land resource spatial feature matrix to generate spatiotemporal correlation feature data: Perform nonlinear feature enhancement processing on the spatiotemporal correlation feature data to generate a nonlinear enhancement feature matrix:

[0025] Step S23: Perform multi-scale fusion processing on the nonlinear enhancement feature matrix to generate a multi-scale feature fusion matrix: Perform dimension compression processing on the multi-scale feature fusion matrix to generate a multi-dimensional spatiotemporal feature matrix:

[0026] Step S24: performing terrain environment analysis on the land resource correction data based on the multi-dimensional spatio-temporal feature matrix to generate land resource environment data; performing multi-dimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data.

[0027] The present invention enables land resource data to be finely analyzed at different scales through geographic resolution segmentation and time scale segmentation. This method can identify features at different levels and time periods, and enhances the comprehensiveness and accuracy of data analysis. The generation of spatial feature matrix and time series association analysis provide a deep understanding of land resource data, and the generation of spatiotemporal association feature data can reveal the potential time and space relationship in the data. Nonlinear feature enhancement processing further improves the expression ability of the feature matrix and can better capture complex nonlinear patterns. Multi-scale feature fusion processing can integrate feature information at different scales and provide a richer perspective to understand land resource data. Dimension compression processing reduces the complexity of data, making the multidimensional spatiotemporal feature matrix easier to analyze and process. Terrain environment analysis based on multidimensional spatiotemporal feature matrix can generate detailed land resource environment data. This helps to understand the terrain, landform and environmental characteristics of the land, and supports more accurate land use and management decisions. Modeling land resource correction data through multidimensional dynamic modeling can generate dynamic terrain modeling data. This makes it possible to simulate and predict the dynamic changes of land resources, providing a basis for dynamic planning and real-time adjustment. The generation and dynamic modeling capabilities of multi-dimensional spatiotemporal feature matrices support comprehensive analysis of land resource data, including spatial and temporal trends, which helps achieve more comprehensive and scientific land resource management and planning.

[0028] Preferably, step S24 includes the following steps:

[0029] Step S241: extracting groundwater characteristics from the land resource correction data based on the multi-dimensional spatiotemporal characteristic matrix to obtain groundwater characteristic data; performing flow distribution analysis on the groundwater characteristic data to generate groundwater flow distribution data; performing erosion evolution on the groundwater characteristic data through the groundwater flow distribution data to generate groundwater erosion evolution data;

[0030] Step S242: Calculate the land population density of the land resource correction data to obtain land population density data; perform land degree change analysis on the land resource correction data based on the land population density data to generate human activity index data;

[0031] Step S243: extract dynamic terrain factors from the environmental feature analysis data to generate a terrain dynamic factor data set; perform land resource correction on the terrain dynamic factor data set using groundwater erosion evolution data and human activity index data to generate corrected terrain data;

[0032] Step S244: Perform multi-dimensional space-time modeling based on the corrected terrain data to generate a multi-dimensional space-time terrain simulation matrix; perform dynamic calculations on the multi-dimensional space-time terrain simulation matrix to generate dynamic terrain modeling data.

[0033] The present invention can deeply understand the characteristics and flow patterns of groundwater through groundwater feature extraction and flow distribution analysis. Erosion evolution analysis further reveals the long-term impact of groundwater on topography and provides detailed data on the impact of groundwater on land resources. Land population density calculation and human activity index analysis can evaluate the impact of population density on land use and environmental changes. This helps to understand the pressure and change trend of human activities on land resources. Through dynamic terrain factor extraction, the key factors affecting terrain changes can be captured. These data help to understand the dynamic characteristics of terrain and its changing laws, and provide a scientific basis for land management and planning. Correction based on groundwater erosion evolution data and human activity index data ensures the accuracy and reliability of terrain data. Multidimensional spatiotemporal modeling provides a comprehensive simulation of terrain changes and helps to predict future terrain change trends. Dynamic terrain modeling data generated by dynamic calculation can better simulate and predict terrain changes. This provides an important decision-making support tool for land resource management, environmental protection and disaster prevention. It integrates multiple factors such as groundwater, population density, and environmental characteristics, and provides comprehensive terrain analysis and prediction capabilities through multidimensional spatiotemporal modeling. This comprehensive analysis helps to formulate more scientific land management and planning strategies.

[0034] Preferably, step S3 comprises the following steps:

[0035] Step S31: constructing a parallelized deep learning network model based on dynamic terrain modeling data to generate a trained deep learning model; importing land resource correction data into the trained deep learning model to predict future land use trends and generate land use prediction data;

[0036] Step S32: performing deviation adjustment on the land use prediction data to generate high-precision land use trend data; performing dynamic prediction result visualization based on the high-precision land use trend data to generate land use trend visualization data; performing multi-dimensional analysis on the land use trend visualization data to generate a comprehensive land use prediction report;

[0037] Step S33: Performing overall planning and management of land resources on the dynamic terrain modeling data according to the comprehensive land use prediction report, and generating an overall planning and management strategy for land resources.

[0038] The present invention can use complex data patterns to predict future land use trends by constructing and training a parallelized deep learning network model. This deep learning model can capture complex relationships in the data and improve the accuracy and reliability of the prediction. The land use prediction data is adjusted for deviations, which can significantly improve the accuracy of the land use trend data. This ensures that the prediction results are closer to the actual situation and provides a more accurate basis for subsequent analysis and decision-making. The dynamic prediction results based on high-precision land use trend data are visualized, making complex prediction data intuitive and easy to understand. This helps relevant decision makers and managers to quickly understand the prediction results and make more scientific decisions. The land use trend visualization data is analyzed in multiple dimensions, and the generated comprehensive land use prediction report provides an in-depth understanding of the land use trend. This report takes into account a variety of factors and helps to identify potential problems and opportunities. The land resource global planning and management strategy formulated according to the comprehensive land use prediction report can effectively integrate and apply the prediction results, thereby formulating a reasonable land resource management strategy. This supports comprehensive and scientific land resource management and planning. By providing accurate prediction data and detailed analysis reports, step S3 provides strong support for land resource management and planning. This helps relevant departments formulate more scientific and reasonable policies and measures to improve the efficiency and sustainability of land resource utilization.

[0039] Preferably, step S31 includes the following steps:

[0040] Step S311: constructing a parallel network based on the convolutional neural network architecture and the long-short time neural network architecture to obtain parallel deep learning network architecture data;

[0041] Step S312: dividing the dynamic terrain modeling data into data sets to generate a model training set and a model test set; performing model training on the parallel deep learning network architecture data through the model training set to generate a parallel deep learning model;

[0042] Step S313: Use the model test set to iterate the model verification of the parallel deep learning model to generate a trained deep learning model; import the land resource correction data into the trained deep learning model to predict future land use trends and generate land use prediction data.

[0043] The present invention can make full use of the spatial feature extraction capability of CNN and the time series modeling capability of LSTM through a parallel deep learning network architecture based on convolutional neural network (CNN) and long short-term memory network (LSTM). This combination improves the performance of the model when processing complex space-time data. The dynamic terrain modeling data is divided into data sets to ensure that the training and testing process of the model can reflect the real characteristics of the data. This division method is conducive to the effective training and evaluation of the model and reduces the risk of overfitting. By training the parallel deep learning network with the model training set, the parameters and structure of the model can be optimized to improve its prediction accuracy. Model optimization during the training process enables the final deep learning model to better capture the patterns in the data. Using the model test set to verify and iterate the trained deep learning model can detect and correct problems in the model and ensure the generalization ability and prediction accuracy of the model. This step improves the reliability and effectiveness of the model. By importing land resource correction data into the trained deep learning model to predict future land use trends, the generated prediction data provides accurate information about land use changes. This helps decision makers to scientifically predict and plan future land use. Accurate land use trend forecast data provides a reliable basis for subsequent analysis and decision-making, and supports scientific and data-driven decision-making in land resource management and planning.

[0044] Preferably, step S33 includes the following steps:

[0045] Step S331: dividing the dynamic terrain modeling data into land planning areas according to the comprehensive land use prediction report to generate land planning area type data, wherein the land planning area type data includes land planning type area data within the city and land planning type area data outside the city;

[0046] Step S332: Perform infrastructure diversity analysis on the land planning type area data in the city to generate infrastructure diversity data; perform land use suitability assessment on the land planning type area data in the city through the infrastructure diversity data to generate urban land use suitability data; perform urban planning management on the land planning type area data in the city based on the urban land use suitability data to generate urban resource planning plot data;

[0047] Step S333: performing a regional land resource type ratio analysis on the land planning type area outside the city to generate land resource type ratio data; performing resource comparison difference calculation on the land resource type ratio data to obtain land resource comparison difference data;

[0048] Step S334: Compare the land resource comparison difference data with the preset land resource difference threshold. When the land resource comparison difference data is greater than or equal to the preset land resource difference threshold, perform mixed resource planning on the corresponding land planning type area outside the city to generate multi-resource planning plot data outside the city; when the land resource comparison difference data is less than the preset land resource difference threshold, perform single resource planning on the corresponding land planning type area outside the city to generate single resource planning plot data outside the city;

[0049] Step S335: Assign high priority to the resource planning plot data within the city, assign medium priority to the multi-resource planning plot data outside the city, and assign low priority to the single resource planning plot data outside the city to conduct comprehensive land resource global planning area allocation, thereby generating a global land resource planning management strategy.

[0050] The present invention can classify the land resources inside and outside the city in detail by dividing the land planning area according to the comprehensive land use forecast report. In this way, more accurate and targeted land planning can be achieved to ensure that different areas are properly managed and utilized. The infrastructure diversity analysis of the land planning type area in the city is helpful to identify and evaluate the diversity and distribution of infrastructure in the urban area. This helps to determine the suitability of land use and provide a scientific basis for urban planning. Based on the urban land use suitability data, the urban planning management can reasonably arrange urban resources and land use, and generate urban resource planning plot data with practical application value. This helps to improve the utilization efficiency of urban resources and the quality of urban development. The regional land resource type ratio analysis of the land planning type area outside the city can understand the distribution and proportion of resources. Resource comparison difference calculation can reveal the resource differences between different regions and provide a basis for resource allocation. According to the land resource comparison difference data and the preset threshold, mixed resource planning or single resource planning is carried out to ensure that the resource allocation of different regions meets the actual needs. This flexible planning method can effectively cope with the specific conditions of different regions. By giving high priority to resource planning plots within the city, medium priority to multi-resource planning plots outside the city, and low priority to single resource planning plots outside the city, resources can be effectively allocated. This ensures that the most critical areas are given priority, optimizes the overall use of resources, and the generated land resource global planning management strategy provides a systematic and scientific land resource management solution. This strategy comprehensively considers the specific conditions of different areas inside and outside the city, which helps to formulate more reasonable and efficient land resource management policies.

[0051] Preferably, step S4 comprises the following steps:

[0052] Step S41: virtual sensor deployment is performed on the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual sensor deployment data, and virtual planning management is performed on the dynamic terrain modeling data according to the virtual sensor deployment data to generate virtual planning land resource management data;

[0053] Step S42: performing real-world mapping on the virtual sensor deployment data to generate real-world sensor deployment data; performing distributed planning management monitoring on the real-world sensor deployment data to generate land resource utilization planning monitoring data;

[0054] Step S43: Analyze the actual planning status of the land resource utilization planning monitoring data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

[0055] The present invention can carry out detailed planning and management of dynamic terrain in a simulated environment through the deployment of virtual sensors and virtual planning management. This virtual management method helps to optimize resource allocation and planning strategies and provides a scientific basis for actual deployment. The virtual sensor deployment data is mapped to reality, and the planning scheme in the virtual environment can be effectively converted into a deployment scheme in the real world. This ensures the consistency between virtual planning and actual operation and improves the accuracy of deployment. The land resource utilization planning monitoring data generated by distributed planning management monitoring can monitor and evaluate the utilization of actual land resources in real time. This helps to timely discover and solve problems in the planning implementation process and ensure the realization of planning goals. The actual planning status data is analyzed to understand the gap between the actual implementation and the original plan. This analysis provides detailed information for evaluating the effect of planning implementation and helps to discover and correct deficiencies in actual operations. Adaptive feedback adjustment is carried out based on the actual planning status data and the virtual planning land resource management data, and the land resource management strategy can be adjusted and optimized according to the actual situation. This ensures that management measures can adapt to changes in real time, improves the flexibility and effect of management, and the generated land resource management feedback data supports dynamic management operations, so that management work can be adjusted in time according to actual conditions and feedback data. This helps improve the response speed and adaptability of land resource management. Through the effective combination of virtual and real, real-time monitoring and feedback adjustment, more efficient and scientific land resource management can be achieved. This not only improves resource utilization efficiency, but also optimizes management strategies and supports the sustainable use of resources.

[0056] In this specification, a land resource management system based on big data analysis is provided, which is used to execute the above-mentioned land resource management method based on big data analysis. The land resource management system based on big data analysis includes:

[0057] A real-time acquisition module is used to acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data, thereby generating multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data, and generate a self-optimized transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimized transmission parameter set, and generate land resource correction data;

[0058] The dynamic modeling module is used to perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data;

[0059] The planning and management module is used to predict future land use trends based on dynamic terrain modeling data and generate land use prediction data; visualize the dynamic prediction results of land use prediction data and generate a comprehensive land use prediction report; perform overall land resource planning and management on dynamic terrain modeling data based on the comprehensive land use prediction report and generate an overall land resource planning and management strategy;

[0060] The monitoring feedback module is used to perform virtual planning management on dynamic terrain modeling data through the global planning management strategy of land resources, and generate virtual planning land resource management data; perform actual planning status analysis on virtual sensor deployment data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on actual planning status data and virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

[0061] The beneficial effect of the present invention is that the real-time performance of multi-source land resource data is significantly improved through asynchronous parallel data acquisition and request-response optimization. The accuracy error of the data is corrected by using a self-optimizing transmission parameter set, which ensures the accuracy of the land resource data and reduces the error. The nonlinear feature enhancement processing improves the expressiveness of the data and generates a richer nonlinear enhancement feature matrix. The terrain environment analysis and dynamic modeling are carried out through a multidimensional spatiotemporal feature matrix, which provides an accurate description of the complex terrain environment and improves the dynamic adaptability of the modeling. The future land use trend is predicted based on the dynamic terrain modeling data, which improves the accuracy and foresight of the land use prediction. The comprehensive land use prediction report is used to formulate an overall planning management strategy, which realizes the effective planning and management of land resources. Through virtual planning management and actual planning status analysis, the flexibility and adaptability of land resource planning are enhanced. Based on the adaptive feedback adjustment of the actual planning status data and the virtual planning data, the dynamic management of land resources is realized, and the effectiveness and accuracy of management are improved. Therefore, the present invention improves the comprehensiveness and accuracy of land resource management by improving the data acquisition, processing, prediction and planning methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of the steps of a land resource management method based on big data analysis;

[0063] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0064] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0068] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0069] To achieve this, please refer to Figures 1 to 3 , a land resource management method based on big data analysis, the method comprising the following steps:

[0070] Step S1: Acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data to generate multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data to generate a self-optimizing transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimizing transmission parameter set to generate land resource correction data;

[0071] Step S2: Perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data according to the land resource environment data to generate dynamic terrain modeling data;

[0072] Step S3: predicting future land use trends based on dynamic terrain modeling data to generate land use prediction data; visualizing dynamic prediction results of land use prediction data to generate a comprehensive land use prediction report; performing overall land resource planning and management on dynamic terrain modeling data according to the comprehensive land use prediction report to generate an overall land resource planning and management strategy;

[0073] Step S4: Virtually plan and manage the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual planning land resource management data; perform actual planning status analysis on the virtual sensor deployment data to generate actual planning status data; and perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data to generate land resource management feedback data to execute dynamic land resource management operations.

[0074] The present invention accelerates the data acquisition speed through asynchronous parallel data acquisition mode, ensures the real-time update of multi-source land resource data, and adapts to the rapidly changing land environment needs. Request-response optimization and self-optimization transmission parameter set improve the efficiency of data transmission, reduce data loss and delay, and ensure the stability and reliability of data transmission. Through precision error correction, the errors caused by transmission or acquisition process are corrected, and accurate land resource correction data is generated to support high-quality analysis and decision-making. Nonlinear feature enhancement processing improves the details and expressiveness of the data, so that the data can more accurately reflect the complex characteristics of land resources. Through dimensional compression processing, the data structure is simplified, the calculation and storage burden is reduced, and important spatial and temporal features are retained, thereby improving data processing efficiency. Terrain environment analysis based on multidimensional spatiotemporal feature matrix provides detailed environmental data and supports in-depth understanding and analysis of complex terrain. Multidimensional dynamic modeling enhances the responsiveness to terrain changes, generates dynamic terrain modeling data, and helps to more accurately predict land use changes and environmental impacts. Land use trend prediction based on dynamic terrain modeling data provides a scientific prediction of future land use changes and supports forward-looking planning decisions. The visualization of dynamic prediction results generates a comprehensive land use prediction report, making the prediction results more intuitive and easy to understand, providing strong support for decision-making. The global planning management strategy formulated according to the comprehensive prediction report optimizes the allocation and utilization of land resources and improves the overall efficiency of resource management. The comparative analysis of virtual planning management data and actual planning status data enhances the consistency and coordination between virtual planning and actual application. Through the feedback adjustment of virtual planning and actual planning status, dynamic optimization of land resource management is achieved, and the flexibility and adaptability of planning implementation are improved. Adaptive land resource management feedback adjustment provides timely management adjustment solutions, effectively responds to problems arising in actual management, and ensures the continuity and effectiveness of dynamic management operations. Therefore, the present invention improves the comprehensiveness and accuracy of land resource management by improving data acquisition, processing, prediction and planning methods.

[0075] In the embodiment of the present invention, reference Figure 1The above is a schematic diagram of the steps of a land resource management method based on big data analysis of the present invention. In this example, the land resource management method based on big data analysis includes the following steps:

[0076] Step S1: Acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data to generate multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data to generate a self-optimizing transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimizing transmission parameter set to generate land resource correction data;

[0077] In an embodiment of the present invention, land resource data from multiple channels such as satellite images, drone photography, ground sensors, geographic information systems (GIS), and environmental monitoring equipment are integrated, including terrain data, land cover types, soil properties, water resource information, etc. The collected data is stored in a format suitable for subsequent processing, such as GeoTIFF, Shapefile, NetCDF, etc. A data warehouse or distributed storage system is used to manage multi-source data to ensure the integrity and accessibility of the data. Data is collected asynchronously at different times and in different spatial ranges to improve data collection efficiency and coverage. Multi-threading or multi-process technology is used to simultaneously obtain data from multiple data sources to reduce collection time. Timestamp alignment and spatial registration are performed on asynchronously collected data to ensure data consistency. The asynchronously collected data is integrated into a unified data set to generate multi-source land resource real-time collection data. The scheduling strategy of data requests is optimized according to the response time and network conditions of the data source to reduce delays. Multiple data requests are merged into one request to reduce network burden and improve efficiency. A cache mechanism is used in the request-response process to reduce repeated requests and improve response speed. The response data is compressed to reduce the amount of data transmission. Based on the results of request and response optimization, a self-optimized transmission parameter set is generated for the current network status and data requirements, including data compression ratio, transmission protocol settings, cache strategy, etc. Error assessment is performed using sample data and real measurement data to identify systematic errors and random errors. An error model is constructed to describe the accuracy error characteristics of the data. Correction algorithms (such as least squares method, Kalman filtering, deviation correction algorithm, etc.) are applied to correct errors. Data accuracy is adjusted according to the self-optimized transmission parameter set. Accuracy correction is performed on standard multi-source land resource real-time data to generate land resource correction data.

[0078] Step S2: Perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data according to the land resource environment data to generate dynamic terrain modeling data;

[0079] In an embodiment of the present invention, the land resource correction data is standardized, including denoising, data smoothing, etc. Preliminary features are selected from the correction data using feature selection techniques (such as principal component analysis (PCA) and independent component analysis (ICA)). Features are enhanced using nonlinear transformation methods (such as kernel principal component analysis (KPCA), autoencoders, and convolutional neural networks (CNN)). These methods can capture complex nonlinear relationships in the data. The nonlinear features of the data are enhanced by mapping to a high-dimensional space. Unsupervised learning is performed using a neural network to learn the nonlinear feature representation of the data. Convolution operations are performed on spatial data to extract high-level features. The nonlinear feature enhancement results are organized into a feature matrix to retain the intensity and relationship of each feature. High-dimensional data is projected into a low-dimensional space through linear transformation to retain the main features. A selected dimension compression technique is applied to generate a low-dimensional matrix with spatiotemporal features. The matrix retains important information in the data while reducing the dimension. Based on the multidimensional spatiotemporal feature matrix, environmental features related to the terrain are identified, such as terrain undulation, soil type, vegetation cover, etc. A classification algorithm (such as random forest, support vector machine (SVM)) is used to classify terrain features into different types. Integrate the extracted environmental features and classification results to generate land resource environmental data. Integrate the land resource correction data and land resource environmental data to form a comprehensive data set for modeling. Select specific dynamic modeling algorithms, such as dynamic system modeling (e.g., Kalman filtering), time series prediction models (e.g., long short-term memory network (LSTM)). Apply the selected dynamic modeling algorithm to model the integrated data. Consider terrain changes, environmental impacts, and time dynamics. Use training data to train the dynamic modeling model to capture the changing trends and dynamic features in the time series. Output the modeling results as dynamic terrain modeling data to describe the changing dynamics of land resources and their impact on the environment.

[0080] Step S3: predicting future land use trends based on dynamic terrain modeling data to generate land use prediction data; visualizing dynamic prediction results of land use prediction data to generate a comprehensive land use prediction report; performing overall land resource planning and management on dynamic terrain modeling data according to the comprehensive land use prediction report to generate an overall land resource planning and management strategy;

[0081] In an embodiment of the present invention, dynamic terrain modeling data is integrated with historical land use data, climate data, population growth data, etc. The integrated data is standardized and cleaned to ensure data quality. A specific prediction algorithm is selected, such as a time series prediction model (e.g., long short-term memory network (LSTM), gated recurrent units (GRU)), a regression model (e.g., linear regression, support vector regression), or a machine learning model (e.g., random forest, gradient boosting tree). The prediction model is trained using historical data, and the model parameters are adjusted to improve the prediction accuracy. The trained model is used to predict future land use trends and generate land use prediction data. These data should cover prediction results for different time periods, including short-term, medium-term, and long-term trends. Data visualization tools (e.g., Tableau, Power BI, Python's Matplotlib, Seaborn) are used to create charts and images, and different types of visualization charts, such as heat maps, trend charts, distribution charts, etc., are generated to show the dynamic changes of land use prediction data. A dynamic dashboard is created to view the changes and trends of land use prediction results in real time. Based on the visualization results, a comprehensive land use prediction report is written. The report should include trend analysis, key findings, charts and explanations of the prediction results. Format the report as a PDF or online report for decision makers and stakeholders to review and analyze. Analyze future land use trends based on the comprehensive land use forecast report to identify key areas and potential problems. Develop an overall planning strategy based on the forecast data, including land use zoning, resource allocation, infrastructure planning, etc. Optimize the strategy and consider planning adjustments under different scenarios to ensure the flexibility and feasibility of the plan. Develop an implementation plan, clarify the steps, timetable and responsible persons, and implement the overall planning management strategy into specific actions, and generate a detailed land resource overall planning management strategy document, including planning goals, implementation steps, monitoring and evaluation mechanisms.

[0082] Step S4: Virtually plan and manage the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual planning land resource management data; perform actual planning status analysis on the virtual sensor deployment data to generate actual planning status data; and perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data to generate land resource management feedback data to execute dynamic land resource management operations.

[0083] In the embodiment of the present invention, a deployment scheme of virtual sensors is designed according to the global planning and management strategy of land resources. Virtual sensors can simulate the functions of actual sensors, such as monitoring land use changes, environmental parameters, etc. Virtual planning and management of dynamic terrain modeling data is performed using virtual sensor deployment data. This includes simulating various land use scenarios and predicting resource changes under different planning schemes. Different land resource management strategies are run in a virtual environment to generate virtual planning land resource management data. These data reflect the impact of different planning schemes on land resources, including resource allocation, utilization efficiency, environmental impact, etc. According to the virtual sensor deployment data, sensors are actually installed to monitor land resources. Ensure that sensors can accurately collect data, such as land use status, environmental conditions, infrastructure conditions, etc. Collect data from actual sensors to monitor the use of land resources in real time. Analyze the collected sensor data to generate actual planning status data. This includes an assessment of the current land use situation, an analysis of resource utilization efficiency, and an assessment of environmental impact. Compare the actual planning status data with the virtual planning land resource management data to identify differences and problems. Adaptive adjustments are made based on the comparison results, including optimizing resource allocation, adjusting management strategies, and improving planning schemes. Formulate adjustment strategies based on the analysis results and generate feedback data for land resource management. The feedback data includes adjustment suggestions, optimization measures and implementation plans. Put the feedback adjustment strategies into practice and implement dynamic land resource management operations. This includes adjusting land use plans, optimizing resource allocation, implementing environmental protection measures, etc. During the implementation process, continuously monitor the land resource situation to ensure the effectiveness of the adjustment measures, and further adjust the management strategy according to the actual situation.

[0084] Preferably, step S1 comprises the following steps:

[0085] Step S11: Acquire multi-source land resource collection data;

[0086] Step S12: constructing a distributed data storage channel for multi-source land resource collection data based on the IPFS protocol to obtain a multi-source data channel; using the multi-source data channel to perform asynchronous parallel data acquisition on the multi-source land resource collection data, thereby generating multi-source land resource real-time collection data;

[0087] Step S13: preprocessing the multi-source land resource real-time collected data to generate standard multi-source land resource real-time data, wherein the data preprocessing includes data format unification, data cleaning, data denoising and data standardization;

[0088] Step S14: streaming the standard multi-source land resource real-time data to generate a land resource streaming data packet; edge caching the land resource streaming data packet to generate a land resource distributed cache data set;

[0089] Step S15: Request-response optimization is performed on the land resource distributed cache data set to generate a self-optimizing transmission parameter set; the self-optimizing transmission parameter set is used to perform precision error correction on the standard multi-source land resource real-time data to generate land resource correction data.

[0090] In an embodiment of the present invention, land resource acquisition data from multiple data sources are obtained, including geographic information system (GIS) data, remote sensing satellite data, drone aerial photography data, and field survey data. These data are aggregated into a preliminary multi-source data acquisition framework for subsequent processing and analysis. The IPFS (Interplanetary File System) protocol is used to store multi-source land resource acquisition data in segments through a distributed network to construct multiple data storage channels. These storage channels are responsible for asynchronous parallel acquisition of data, that is, acquiring land resource data from different data sources and nodes at the same time, thereby speeding up data acquisition and ensuring the real-time nature of the data. The generated multi-source data channel allows the system to dynamically acquire real-time updated data, thereby generating multi-source land resource real-time acquisition data. The multi-source land resource real-time acquisition data is preprocessed to ensure data consistency and quality, including the following sub-steps: standardizing the data formats from different sources into a unified data model for subsequent processing. Remove erroneous, redundant or missing data to ensure that the data is accurate. Eliminate the noise in the data to reduce the interference of useless information. Normalize the data to eliminate the scale differences between different data sources. After data preprocessing, the generated standard multi-source land resource real-time data can be used for subsequent steps. The standardized data is converted into land resource streaming data packets and transmitted in real time through the network. To ensure the stability and efficiency of the transmission process, the cache mechanism of distributed edge computing is used to cache these streaming data in real time to form a distributed cache data set of land resources. This can reduce delays during data transmission, improve user access speed, and ensure rapid data response. By analyzing parameters such as transmission delay and packet loss rate between network requests and data responses, a self-optimizing transmission parameter set is automatically generated. Using this parameter set, the accuracy error of real-time transmission and cached data is corrected to eliminate data deviations and errors introduced during the transmission process, thereby ensuring the accuracy of land resource data. The final generated land resource correction data can provide a reliable data basis for land resource management, planning, and decision-making.

[0091] Preferably, step S14 comprises the following steps:

[0092] Step S141: Perform streaming media segmentation processing on the standard multi-source land resource data to generate land resource streaming media segmentation data; perform transmission protocol encapsulation on the land resource streaming media segmentation data to generate a land resource transmission data packet:

[0093] Step S142: perform edge cache preprocessing on the land resource transmission data packet to generate edge cache data segments: perform consistency check on the edge cache data segments to generate cache consistency check data:

[0094] Step S143: Distributed aggregation processing is performed on the edge cache data fragments according to the cache consistency check data to generate a land resource distributed cache data set.

[0095] In an embodiment of the present invention, standard multi-source land resource data is obtained from a multi-source land resource database, and the data includes land use type, land resource distribution, resource utilization rate, etc. According to the spatial and temporal attributes of the land resource data, a slicing algorithm (such as a sliding window, a fixed time / space interval) is used to divide the data into multiple streaming media slices. Each slice represents land resource information of a specific area and time period. The land resource data is sliced ​​to generate land resource streaming media slice data, ensuring that each slice has a relatively small size for transmission. According to the network environment and data transmission requirements, a suitable transmission protocol, such as HTTP, WebRTC or a custom protocol, is selected to ensure stable and efficient data transmission. The land resource streaming media slice data is encapsulated according to the selected transmission protocol to generate a land resource transmission data packet, which contains metadata of the slice (such as timestamp, area identifier, slice size, etc.) for identification and management during the transmission process. The land resource transmission data packet is edge cache preprocessed by cache nodes distributed at the edge of the network. The edge node can be located on a server or gateway device close to the user. The land resource transmission data packet is decomposed into multiple cache data fragments, each of which represents a part of the transmission data and is stored in different edge nodes to reduce the latency and bandwidth occupation of data access. A verification algorithm (such as MD5, SHA-256) is used to perform consistency verification on each edge cache data fragment to ensure the integrity and consistency during data transmission. After verification, cache consistency verification data is generated, which contains the verification code of each data fragment for subsequent distributed processing and data verification. According to the cache consistency verification data, the cache data fragments distributed on different edge nodes are distributedly aggregated. The cache fragments scattered on multiple nodes are integrated by using consistent hashing or distributed file systems (such as HDFS). During the aggregation process, the integrity of each cache fragment is verified by consistency verification data to ensure that the data is not lost or tampered with during aggregation. After the aggregation process is completed, a complete land resource distributed cache data set is generated. This data set can be accessed in a distributed environment and can provide fast data retrieval and analysis capabilities. According to the new land resource data changes, the distributed cache data set is updated in real time to ensure the timeliness and accuracy of the data.

[0096] As an example of the present invention, refer to Figure 2As shown, in this example, step S2 includes:

[0097] Step S21: performing multi-scale segmentation processing on the land resource correction data to generate multi-scale spatiotemporal data blocks, wherein the multi-scale segmentation includes geographic resolution segmentation and time scale segmentation; performing spatial feature extraction on the multi-scale spatiotemporal data blocks to generate a land resource spatial feature matrix;

[0098] Step S22: Perform time series correlation analysis on the land resource spatial feature matrix to generate spatiotemporal correlation feature data: Perform nonlinear feature enhancement processing on the spatiotemporal correlation feature data to generate a nonlinear enhancement feature matrix:

[0099] Step S23: Perform multi-scale fusion processing on the nonlinear enhancement feature matrix to generate a multi-scale feature fusion matrix: Perform dimension compression processing on the multi-scale feature fusion matrix to generate a multi-dimensional spatiotemporal feature matrix:

[0100] Step S24: performing terrain environment analysis on the land resource correction data based on the multi-dimensional spatio-temporal feature matrix to generate land resource environment data; performing multi-dimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data.

[0101] In an embodiment of the present invention, land resource correction data is obtained, including geographic information and resource distribution data. The land resource data is divided according to the geographic resolution, and different spatial scales are selected, such as segmentation by region, plot or administrative division. Through geographic resolution segmentation, the land resource data is divided into multiple spatiotemporal data blocks with different spatial resolutions. According to the time dimension, it is divided by time units such as year, quarter, month, day, etc. to ensure the temporal nature of the land resource data. Combined with time scale segmentation and geographic resolution segmentation, multi-scale spatiotemporal data blocks are generated. Spatial features are extracted from the multi-scale spatiotemporal data blocks, using edge detection, morphological analysis, spatial statistics and other methods. The extracted spatial features include land use type, terrain features, resource distribution pattern, etc., to generate a land resource spatial feature matrix. Based on the land resource spatial feature matrix generated in step S21, the time series changes therein are analyzed. The time series data is analyzed using correlation analysis (such as Granger causal analysis, mutual information analysis) to identify the association patterns at different time scales. Based on the time series association analysis, spatiotemporal association feature data is generated to capture the changing laws of land resources in different time and space. Nonlinear processing techniques, such as neural networks, support vector machines or other nonlinear regression methods, are used to perform nonlinear enhancement on spatiotemporal correlation feature data. Through nonlinear feature enhancement processing, a nonlinear enhanced feature matrix is ​​generated to improve the ability to capture complex terrain and environmental features. The nonlinear enhanced feature matrix is ​​subjected to multi-scale fusion processing, and features at different scales are unified and integrated using methods such as wavelet transform and convolutional neural network (CNN). The fused matrix can characterize the characteristics of land resources at different geographical scales and time scales, and generate a multi-scale feature fusion matrix. Methods such as principal component analysis (PCA), linear discriminant analysis (LDA) or t-SNE are used to perform dimensional compression on the multi-scale feature fusion matrix to reduce redundant data. Through dimensional compression, a streamlined multi-dimensional spatiotemporal feature matrix is ​​obtained, which retains important spatiotemporal features, reduces data dimensions, and facilitates subsequent modeling and analysis. Based on the multi-dimensional spatiotemporal feature matrix, terrain data (such as elevation, slope, land cover, etc.) are combined to perform terrain environment analysis. The terrain environment of land resources is analyzed using methods such as digital elevation model (DEM) analysis, terrain statistical analysis, and environmental factor calculation. The analysis results generate land resource and environmental data to characterize the terrain and environmental characteristics of the area where the land resources are located. Based on the land resource and environmental data, dynamic modeling techniques (such as dynamic Bayesian networks and spatiotemporal dynamic models) are used to model the land resource correction data. In the dynamic modeling process, multi-dimensional factors such as space, time, and environment are comprehensively considered to simulate the dynamic changes of land resources. The modeling results generate dynamic terrain modeling data to provide predictions and assessments of land resource changes over time and the environment.

[0102] Preferably, step S24 includes the following steps:

[0103] Step S241: extracting groundwater characteristics from the land resource correction data based on the multi-dimensional spatiotemporal characteristic matrix to obtain groundwater characteristic data; performing flow distribution analysis on the groundwater characteristic data to generate groundwater flow distribution data; performing erosion evolution on the groundwater characteristic data through the groundwater flow distribution data to generate groundwater erosion evolution data;

[0104] Step S242: Calculate the land population density of the land resource correction data to obtain land population density data; perform land degree change analysis on the land resource correction data based on the land population density data to generate human activity index data;

[0105] Step S243: extract dynamic terrain factors from the environmental feature analysis data to generate a terrain dynamic factor data set; perform land resource correction on the terrain dynamic factor data set using groundwater erosion evolution data and human activity index data to generate corrected terrain data;

[0106] Step S244: Perform multi-dimensional space-time modeling based on the corrected terrain data to generate a multi-dimensional space-time terrain simulation matrix; perform dynamic calculations on the multi-dimensional space-time terrain simulation matrix to generate dynamic terrain modeling data.

[0107] In the embodiment of the present invention, land resource correction data and related groundwater data are obtained based on a multidimensional spatiotemporal feature matrix. Hydrological features are extracted from the land resource correction data using a hydrological model (such as a groundwater flow model and a permeability model), including groundwater velocity, groundwater content, permeability coefficient, etc. After processing, groundwater characteristic data reflecting groundwater characteristics are obtained. Finite difference method or finite element method is used to analyze the flow distribution of groundwater characteristic data in combination with a hydrological model to simulate the flow path and distribution pattern of groundwater. After analysis, groundwater flow distribution data is generated to show the flow law and distribution characteristics of groundwater in space. Through groundwater flow distribution data, the erosion of groundwater on soil and landforms is simulated. An erosion model (such as a RUSLE model and a USLE model) is used to analyze the impact of groundwater erosion on the surface and underground topography. The simulation results generate groundwater erosion evolution data to show the dynamic change process of erosion and the long-term impact on land resources. Data such as population distribution, infrastructure, and buildings in the land resource correction data are used. Through the geographic information system (GIS) and remote sensing technology, combined with demographic data, the land population density of different regions is calculated to reflect the density of population distribution in each region. After calculation, the land population density data is obtained for further analysis. Based on the land population density data, combined with the impact of human activities on land use, the change analysis of land resources is carried out, including changes in different land use types such as cultivated land, construction land, and forests. Through the land use change analysis, index data reflecting the intensity and impact of human activities are generated to characterize the degree of impact of human activities on land resources. Based on the environmental characteristic analysis data, dynamic terrain factors are extracted. The extracted terrain factors include dynamic change factors such as slope, aspect, and elevation. Through the dynamic extraction process, a set of dynamic factor data sets representing terrain changes are generated for subsequent correction. Combined with the groundwater erosion evolution data and the human activity index data, the extracted terrain dynamic factor data set is corrected for land resources, and the terrain data is adjusted to reflect the real terrain changes. After correction, more accurate corrected terrain data is generated, taking into account the dual impact of groundwater erosion and human activities. Based on the corrected terrain data, a multidimensional space-time model is constructed, taking into account spatial, temporal and environmental factors, and modeling is performed using multidimensional space-time modeling techniques (such as space-time weighted regression model and dynamic Bayesian model). Through modeling, a multidimensional space-time terrain simulation matrix is ​​generated to reflect the terrain changes and evolution trends at different space-time scales. The multidimensional space-time terrain simulation matrix is ​​dynamically calculated, and simulation algorithms (such as numerical simulation and dynamic system analysis) are used to simulate terrain changes in the future. The calculation results generate dynamic terrain modeling data, which provides predictions of future terrain change trends to assist decision-making and management.

[0108] As an example of the present invention, refer to Figure 3As shown, in this example, step S3 includes:

[0109] Step S31: constructing a parallelized deep learning network model based on dynamic terrain modeling data to generate a trained deep learning model; importing land resource correction data into the trained deep learning model to predict future land use trends and generate land use prediction data;

[0110] Step S32: performing deviation adjustment on the land use prediction data to generate high-precision land use trend data; performing dynamic prediction result visualization based on the high-precision land use trend data to generate land use trend visualization data; performing multi-dimensional analysis on the land use trend visualization data to generate a comprehensive land use prediction report;

[0111] Step S33: Performing overall planning and management of land resources on the dynamic terrain modeling data according to the comprehensive land use prediction report, and generating an overall planning and management strategy for land resources.

[0112] In the embodiment of the present invention, the dynamic terrain modeling data generated based on step S24 includes multi-dimensional spatiotemporal features such as terrain change, groundwater erosion, and human activity index. A deep learning model such as a convolutional neural network (CNN), a long short-term memory network (LSTM), or a graph neural network (GNN) is used to consider the spatiotemporal characteristics of land use data for future prediction. Distributed parallel computing technology (such as GPU parallel computing, MPI distributed computing) is used to accelerate the training of the deep learning network model to improve the computing efficiency and training speed of the model. Through parallel training, a deep learning model suitable for land use trend prediction is generated. The corrected land resource data generated in step S24 is imported into the trained deep learning model. The model predicts the future land use change trend based on the input correction data and the learned spatiotemporal features, and the prediction content includes the utilization of different types of land such as agricultural land, industrial land, and residential areas. After the prediction, preliminary land use prediction data is generated, including the quantification and distribution of land use in the future period of time. The preliminary prediction data is subjected to deviation analysis, and the historical data and the prediction results are compared to find the errors and deviations in the prediction process. Use statistical methods (such as Kalman filtering, Bayesian updating) or rule-based algorithms to adjust the prediction deviation and improve the accuracy of the prediction data. The adjusted prediction data is closer to the actual land use situation, generating high-precision land use trend data. Use GIS visualization tools or custom visualization platforms (such as D3.js, Leaflet) to visualize high-precision land use trend data and generate dynamic maps reflecting future land use trends. The dimensions displayed in the visualization process include time changes, spatial distribution, and changing trends of different land types, ensuring that the visualized data can dynamically reflect the changes in land use trends. The visualization results are displayed in various forms such as maps, charts, and animations to generate comprehensive land use trend visualization data. Based on the visualization data, multidimensional analysis is performed, including spatial distribution feature analysis, time series analysis, land use classification analysis, etc., to gain an in-depth understanding of the changing laws of land use. Combined with the results of multidimensional analysis, a comprehensive land use forecast report is generated, which includes future land use trend forecasts, risk assessments, potential land resource use conflicts, etc. According to the analysis results in the comprehensive land use forecast report, combined with dynamic terrain modeling data, a global planning and management strategy for land resources is formulated. By comprehensively considering the current status of land resources, future utilization trends, potential environmental risks and other factors, land use planning strategies for different regions are formulated. The planning content may include the allocation of land use, the division of development areas, the establishment of protection areas, the transformation of land use, etc., to ensure the rational use of land resources.

[0113] Preferably, step S31 includes the following steps:

[0114] Step S311: constructing a parallel network based on the convolutional neural network architecture and the long-short time neural network architecture to obtain parallel deep learning network architecture data;

[0115] Step S312: dividing the dynamic terrain modeling data into data sets to generate a model training set and a model test set; performing model training on the parallel deep learning network architecture data through the model training set to generate a parallel deep learning model;

[0116] Step S313: Use the model test set to iterate the model verification of the parallel deep learning model to generate a trained deep learning model; import the land resource correction data into the trained deep learning model to predict future land use trends and generate land use prediction data.

[0117] In an embodiment of the present invention, a convolutional neural network architecture (CNN) is used to extract spatial features from land resource correction data. CNN can process input data through multiple convolutional layers and pooling layers to learn the spatial distribution characteristics of terrain data. Long short-term memory network architecture (LSTM) is used to process time series features of land use. LSTM can capture the long-term dependence of land use changes and is suitable for processing future trend predictions. The CNN architecture and the LSTM architecture are parallelized, and the advantages of the two are combined to construct a parallel deep learning network. The parallel network can capture features in spatial and temporal dimensions at the same time, thereby improving the prediction performance of the model. By setting the hyperparameters of the network (such as convolution kernel size, number of layers, number of hidden layer units, etc.), network connection relationship, loss function, optimizer parameters, etc., the parallel deep learning network architecture data is finally generated. The dynamic terrain modeling data generated in step S24 is divided into a data set, and the data is divided into a model training set and a model test set according to a certain ratio. The training set is used to train the model, and the test set is used to verify the performance of the model. The data set should contain multi-dimensional data such as terrain features, groundwater features, and land use features to ensure the diversity and comprehensiveness of model training and testing. Import the divided training set data into the parallel deep learning network architecture to start model training. Use optimization algorithms such as Adam optimizer or RMSProp, combined with gradient descent method, adjust the model weights through multiple rounds of iterations to minimize the prediction error. According to the land use prediction task, select a suitable loss function (such as mean square error MSE, cross entropy loss, etc.) to evaluate the performance of the model. After multiple iterative training, generate a parallel deep learning model that can accurately capture spatial and temporal characteristics. Import the divided test set data into the parallel deep learning model for model verification. Through the verification process, evaluate the prediction accuracy and generalization ability of the model. Use indicators such as mean square error (MSE), mean absolute error (MAE), accuracy, etc. to evaluate the performance of the model on the test set. Through multiple iterations and adjustment of hyperparameters, the best performance model is finally obtained. According to the verification results, iteratively optimize the model parameters until the model converges and reaches the expected performance standard. The corrected land resource data generated in step S24 is input into the verified parallel deep learning model, and the model predicts future land use trends through spatial and temporal dimension features. The model predicts the future distribution of land use based on the input data, including the changes in the use of different types of land such as agricultural land, urban development land, and ecological protection areas, and generates land use prediction data.

[0118] Preferably, step S33 includes the following steps:

[0119] Step S331: dividing the dynamic terrain modeling data into land planning areas according to the comprehensive land use prediction report to generate land planning area type data, wherein the land planning area type data includes land planning type area data within the city and land planning type area data outside the city;

[0120] Step S332: Perform infrastructure diversity analysis on the land planning type area data in the city to generate infrastructure diversity data; perform land use suitability assessment on the land planning type area data in the city through the infrastructure diversity data to generate urban land use suitability data; perform urban planning management on the land planning type area data in the city based on the urban land use suitability data to generate urban resource planning plot data;

[0121] Step S333: performing a regional land resource type ratio analysis on the land planning type area outside the city to generate land resource type ratio data; performing resource comparison difference calculation on the land resource type ratio data to obtain land resource comparison difference data;

[0122] Step S334: Compare the land resource comparison difference data with the preset land resource difference threshold. When the land resource comparison difference data is greater than or equal to the preset land resource difference threshold, perform mixed resource planning on the corresponding land planning type area outside the city to generate multi-resource planning plot data outside the city; when the land resource comparison difference data is less than the preset land resource difference threshold, perform single resource planning on the corresponding land planning type area outside the city to generate single resource planning plot data outside the city;

[0123] Step S335: Assign high priority to the resource planning plot data within the city, assign medium priority to the multi-resource planning plot data outside the city, and assign low priority to the single resource planning plot data outside the city to conduct comprehensive land resource global planning area allocation, thereby generating a global land resource planning management strategy.

[0124] In the embodiment of the present invention, the land planning area is divided by using the previously generated dynamic terrain modeling data and combining the land use trend prediction data. According to the comprehensive land use prediction report, the different land planning types in the city and outside the city are defined. The land planning areas in the city mainly include residential areas, commercial areas, industrial areas, etc., and the land planning areas outside the city include agricultural land, ecological protection areas, forest land, etc. The land planning type area data in the city and the land planning type area data outside the city are generated by division. The data of infrastructure is extracted from the different land planning type areas in the city, including the distribution of infrastructure such as roads, water supply, power supply, and transportation. The distribution of various types of infrastructure in different areas of the city is analyzed by using the diversity analysis method to generate infrastructure diversity data. By comprehensively evaluating the infrastructure diversity data, the suitability of land use in various areas of the city is measured, such as suitability for residence, business or industry, and the urban land use suitability data is generated according to the evaluation results. Based on the urban land use suitability data, the land planning of various areas in the city is further optimized, the use and development strategies of different plots are adjusted, and the resource planning plot data in the city is generated. Classify the land resources in the land planning type area outside the city, and calculate the proportion of each type of resources, including agricultural land, forest, pasture, etc. According to the calculation results, generate the land resource type proportion data. By comparing the proportion of different types of land resources, calculate the difference between resources, and obtain the land resource comparison difference data. This difference calculation helps to judge the utilization potential and optimization direction of land resources outside the city. Compare the land resource comparison difference data with the preset difference threshold. If the comparison difference is greater than or equal to the threshold, it means that the resource distribution is relatively balanced and mixed resource planning can be implemented; if the difference is small, it means that the resource is relatively single, and single resource planning is inclined. When the difference is greater than or equal to the threshold, mixed resource planning is carried out in the land planning type area outside the city to generate multi-resource planning plot data outside the city. When the difference is less than the threshold, single resource planning is carried out to generate single resource planning plot data outside the city. Different priorities are given to different types of planning areas. Due to its importance and resource density, the resource planning plot data within the city is usually given high priority; the multi-resource planning plot data outside the city is given medium priority; and the single resource planning plot data outside the city is given low priority. Based on the priority division, global planning is carried out for different land resources to ensure that high-priority areas obtain more planning resources and development support, and ultimately generate a global planning and management strategy for land resources.

[0125] Preferably, step S4 comprises the following steps:

[0126] Step S41: virtual sensor deployment is performed on the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual sensor deployment data, and virtual planning management is performed on the dynamic terrain modeling data according to the virtual sensor deployment data to generate virtual planning land resource management data;

[0127] Step S42: performing real-world mapping on the virtual sensor deployment data to generate real-world sensor deployment data; performing distributed planning management monitoring on the real-world sensor deployment data to generate land resource utilization planning monitoring data;

[0128] Step S43: Analyze the actual planning status of the land resource utilization planning monitoring data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

[0129] In the embodiment of the present invention, the deployment position of the virtual sensor in the dynamic terrain modeling data is determined according to the previously generated global planning management strategy of land resources. Virtual sensors (such as geological sensors, hydrological sensors, environmental monitoring sensors, etc.) with specific different land types and resource requirements are selected, and virtual sensor deployment data is generated. The virtual sensor deployment data is combined with the dynamic terrain modeling data to perform comprehensive management and simulation planning of the land resources in the virtual planning area, and virtual planning land resource management data is generated. According to the virtual sensor deployment data, the real sensor is deployed, the actual location is selected for installing the sensor, and the real sensor deployment data is generated. The real sensor includes environmental monitoring equipment, soil sensors, air quality monitoring sensors, etc. A distributed monitoring system is constructed using the real sensor deployment data to monitor the utilization of different land resources in real time. The monitoring system collects relevant data on land resource utilization (such as terrain changes, water resource consumption, vegetation cover changes, etc.) to generate land resource utilization planning monitoring data. The land resource utilization planning monitoring data collected from the sensor monitoring system is compared with the previous plan to generate actual planning status data to analyze whether the land resource utilization is consistent with the expected plan. Based on the actual planning status data and virtual planning land resource management data, the adaptive feedback mechanism is used to dynamically adjust the land resource management to ensure that the planning is in line with the actual situation. The land resource management feedback data is generated according to the adjusted results to optimize the subsequent dynamic land resource management operations.

[0130] In this specification, a land resource management system based on big data analysis is provided, which is used to execute the above-mentioned land resource management method based on big data analysis. The land resource management system based on big data analysis includes:

[0131] A real-time acquisition module is used to acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data, thereby generating multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data, and generate a self-optimized transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimized transmission parameter set, and generate land resource correction data;

[0132] The dynamic modeling module is used to perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data;

[0133] The planning and management module is used to predict future land use trends based on dynamic terrain modeling data and generate land use prediction data; visualize the dynamic prediction results of land use prediction data and generate a comprehensive land use prediction report; perform overall land resource planning and management on dynamic terrain modeling data based on the comprehensive land use prediction report and generate an overall land resource planning and management strategy;

[0134] The monitoring feedback module is used to perform virtual planning management on dynamic terrain modeling data through the global planning management strategy of land resources, and generate virtual planning land resource management data; perform actual planning status analysis on virtual sensor deployment data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on actual planning status data and virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

[0135] The beneficial effect of the present invention is that the real-time performance of multi-source land resource data is significantly improved through asynchronous parallel data acquisition and request-response optimization. The accuracy error of the data is corrected by using a self-optimizing transmission parameter set, which ensures the accuracy of the land resource data and reduces the error. The nonlinear feature enhancement processing improves the expressiveness of the data and generates a richer nonlinear enhancement feature matrix. The terrain environment analysis and dynamic modeling are carried out through a multidimensional spatiotemporal feature matrix, which provides an accurate description of the complex terrain environment and improves the dynamic adaptability of the modeling. The future land use trend is predicted based on the dynamic terrain modeling data, which improves the accuracy and foresight of the land use prediction. The comprehensive land use prediction report is used to formulate an overall planning management strategy, which realizes the effective planning and management of land resources. Through virtual planning management and actual planning status analysis, the flexibility and adaptability of land resource planning are enhanced. Based on the adaptive feedback adjustment of the actual planning status data and the virtual planning data, the dynamic management of land resources is realized, and the effectiveness and accuracy of management are improved. Therefore, the present invention improves the comprehensiveness and accuracy of land resource management by improving the data acquisition, processing, prediction and planning methods.

[0136] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0137] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A land resource management method based on big data analysis, characterized in that: The following steps are involved: Step S1: Acquire multi-source land resource collection data; Asynchronous parallel data acquisition is performed on multi-source land resource collection data, thereby generating multi-source land resource real-time collection data; Perform request-response optimization on multi-source land resource real-time collection data to generate a self-optimized transmission parameter set; use the self-optimized transmission parameter set to perform precision error correction on standard multi-source land resource real-time data to generate land resource correction data; Step S2: Perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data according to the land resource environment data to generate dynamic terrain modeling data; Step S3: predicting future land use trends based on dynamic terrain modeling data to generate land use prediction data; visualizing dynamic prediction results of land use prediction data to generate a comprehensive land use prediction report; performing overall land resource planning and management on dynamic terrain modeling data according to the comprehensive land use prediction report to generate an overall land resource planning and management strategy; Step S4: virtual planning management is performed on the dynamic terrain modeling data through the land resource global planning management strategy to generate virtual planning land resource management data; Performing actual planning status analysis on the virtual sensor deployment data to generate actual planning status data; Adaptive land resource management feedback adjustment is performed based on actual planning status data and virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

2. The land resource management method based on big data analysis according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Acquire multi-source land resource collection data; Step S12: constructing a distributed data storage channel for multi-source land resource collection data based on the IPFS protocol to obtain a multi-source data channel; using the multi-source data channel to perform asynchronous parallel data acquisition on the multi-source land resource collection data, thereby generating multi-source land resource real-time collection data; Step S13: preprocessing the multi-source land resource real-time collected data to generate standard multi-source land resource real-time data, wherein the data preprocessing includes data format unification, data cleaning, data denoising and data standardization; Step S14: streaming the standard multi-source land resource real-time data to generate a land resource streaming data packet; edge caching the land resource streaming data packet to generate a land resource distributed cache data set; Step S15: Request-response optimization is performed on the land resource distributed cache data set to generate a self-optimizing transmission parameter set; the self-optimizing transmission parameter set is used to perform precision error correction on the standard multi-source land resource real-time data to generate land resource correction data.

3. The land resource management method based on big data analysis according to claim 2 is characterized in that: Step S14 includes the following steps: Step S141: Perform streaming media segmentation processing on the standard multi-source land resource data to generate land resource streaming media segmentation data; perform transmission protocol encapsulation on the land resource streaming media segmentation data to generate a land resource transmission data packet: Step S142: perform edge cache preprocessing on the land resource transmission data packet to generate edge cache data segments: perform consistency check on the edge cache data segments to generate cache consistency check data: Step S143: Distributed aggregation processing is performed on the edge cache data fragments according to the cache consistency check data to generate a land resource distributed cache data set.

4. The land resource management method based on big data analysis according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: performing multi-scale segmentation processing on the land resource correction data to generate multi-scale spatiotemporal data blocks, wherein the multi-scale segmentation includes geographic resolution segmentation and time scale segmentation; performing spatial feature extraction on the multi-scale spatiotemporal data blocks to generate a land resource spatial feature matrix; Step S22: Perform time series correlation analysis on the land resource spatial feature matrix to generate spatiotemporal correlation feature data: Perform nonlinear feature enhancement processing on the spatiotemporal correlation feature data to generate a nonlinear enhancement feature matrix: Step S23: Perform multi-scale fusion processing on the nonlinear enhancement feature matrix to generate a multi-scale feature fusion matrix: Perform dimension compression processing on the multi-scale feature fusion matrix to generate a multi-dimensional spatiotemporal feature matrix: Step S24: performing terrain environment analysis on the land resource correction data based on the multi-dimensional spatio-temporal feature matrix to generate land resource environment data; performing multi-dimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data.

5. The land resource management method based on big data analysis according to claim 4 is characterized in that: Step S24 includes the following steps: Step S241: extracting groundwater characteristics from the land resource correction data based on the multi-dimensional spatiotemporal characteristic matrix to obtain groundwater characteristic data; performing flow distribution analysis on the groundwater characteristic data to generate groundwater flow distribution data; performing erosion evolution on the groundwater characteristic data through the groundwater flow distribution data to generate groundwater erosion evolution data; Step S242: Calculate the land population density of the land resource correction data to obtain land population density data; perform land degree change analysis on the land resource correction data based on the land population density data to generate human activity index data; Step S243: extract dynamic terrain factors from the environmental feature analysis data to generate a terrain dynamic factor data set; perform land resource correction on the terrain dynamic factor data set using groundwater erosion evolution data and human activity index data to generate corrected terrain data; Step S244: Perform multi-dimensional space-time modeling based on the corrected terrain data to generate a multi-dimensional space-time terrain simulation matrix; perform dynamic calculations on the multi-dimensional space-time terrain simulation matrix to generate dynamic terrain modeling data.

6. The land resource management method based on big data analysis according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: constructing a parallelized deep learning network model based on dynamic terrain modeling data to generate a trained deep learning model; importing land resource correction data into the trained deep learning model to predict future land use trends and generate land use prediction data; Step S32: performing deviation adjustment on the land use prediction data to generate high-precision land use trend data; performing dynamic prediction result visualization based on the high-precision land use trend data to generate land use trend visualization data; performing multi-dimensional analysis on the land use trend visualization data to generate a comprehensive land use prediction report; Step S33: Performing overall planning and management of land resources on the dynamic terrain modeling data according to the comprehensive land use prediction report, and generating an overall planning and management strategy for land resources.

7. The land resource management method based on big data analysis according to claim 6 is characterized in that: Step S31 The following steps are involved: Step S311: constructing a parallel network based on the convolutional neural network architecture and the long-short time neural network architecture to obtain parallel deep learning network architecture data; Step S312: dividing the dynamic terrain modeling data into data sets to generate a model training set and a model test set; performing model training on the parallel deep learning network architecture data through the model training set to generate a parallel deep learning model; Step S313: using the model test set to perform model verification iteration on the parallel deep learning model, thereby generating a trained deep learning model; The land resource correction data is imported into the trained deep learning model to predict future land use trends and generate land use prediction data.

8. The land resource management method based on big data analysis according to claim 6 is characterized in that: Step S33 includes the following steps: Step S331: dividing the dynamic terrain modeling data into land planning areas according to the comprehensive land use prediction report to generate land planning area type data, wherein the land planning area type data includes land planning type area data within the city and land planning type area data outside the city; Step S332: Perform infrastructure diversity analysis on the land planning type area data in the city to generate infrastructure diversity data; perform land use suitability assessment on the land planning type area data in the city through the infrastructure diversity data to generate urban land use suitability data; perform urban planning management on the land planning type area data in the city based on the urban land use suitability data to generate urban resource planning plot data; Step S333: performing a regional land resource type ratio analysis on the land planning type area outside the city to generate land resource type ratio data; performing a resource comparison difference calculation on the land resource type ratio data to obtain land resource comparison difference data; Step S334: Compare the land resource comparison difference data with the preset land resource difference threshold. When the land resource comparison difference data is greater than or equal to the preset land resource difference threshold, perform mixed resource planning on the corresponding land planning type area outside the city to generate multi-resource planning plot data outside the city; when the land resource comparison difference data is less than the preset land resource difference threshold, perform single resource planning on the corresponding land planning type area outside the city to generate single resource planning plot data outside the city; Step S335: Assign high priority to the resource planning plot data within the city, assign medium priority to the multi-resource planning plot data outside the city, and assign low priority to the single resource planning plot data outside the city to conduct comprehensive land resource global planning area allocation, thereby generating a global land resource planning management strategy.

9. The land resource management method based on big data analysis according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: virtual sensor deployment is performed on the dynamic terrain modeling data through the global land resource planning management strategy to generate virtual sensor deployment data, and virtual planning management is performed on the dynamic terrain modeling data according to the virtual sensor deployment data to generate virtual planning land resource management data; Step S42: performing real-world mapping on the virtual sensor deployment data to generate real-world sensor deployment data; performing distributed planning management monitoring on the real-world sensor deployment data to generate land resource utilization planning monitoring data; Step S43: Analyze the actual planning status of the land resource utilization planning monitoring data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on the actual planning status data and the virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

10. A land resource management system based on big data analysis, characterized in that: Used to execute the land resource management method based on big data analysis as claimed in claim 1, the land resource management system based on big data analysis comprises: A real-time acquisition module is used to acquire multi-source land resource acquisition data; perform asynchronous parallel data acquisition on the multi-source land resource acquisition data, thereby generating multi-source land resource real-time acquisition data; perform request-response optimization on the multi-source land resource real-time acquisition data, and generate a self-optimized transmission parameter set; perform precision error correction on the standard multi-source land resource real-time data using the self-optimized transmission parameter set, and generate land resource correction data; The dynamic modeling module is used to perform nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhancement feature matrix; perform dimension compression processing on the nonlinear enhancement feature matrix to generate a multidimensional spatiotemporal feature matrix; perform terrain environment analysis on the land resource correction data based on the multidimensional spatiotemporal feature matrix to generate land resource environment data; perform multidimensional dynamic modeling on the land resource correction data based on the land resource environment data to generate dynamic terrain modeling data; The planning and management module is used to predict future land use trends based on dynamic terrain modeling data and generate land use prediction data; visualize the dynamic prediction results of land use prediction data and generate a comprehensive land use prediction report; perform overall land resource planning and management on dynamic terrain modeling data based on the comprehensive land use prediction report and generate an overall land resource planning and management strategy; The monitoring feedback module is used to perform virtual planning management on dynamic terrain modeling data through the global planning management strategy of land resources, and generate virtual planning land resource management data; perform actual planning status analysis on virtual sensor deployment data to generate actual planning status data; perform adaptive land resource management feedback adjustment based on actual planning status data and virtual planning land resource management data, thereby generating land resource management feedback data to execute dynamic land resource management operations.

Images