Method and system based on integrated territorial monitoring data

Abstract

The invention relates to the technical field of territorial monitoring data processing, in particular to a method and system based on territorial monitoring data integration. The method comprises the following steps: acquiring heterogeneous territorial monitoring data; performing heterogeneous feature dimension interaction data analysis based on the heterogeneous territorial monitoring data to generate heterogeneous feature multi-dimensional interaction data; performing territorial integrated three-dimensional model analysis according to the heterogeneous feature multi-dimensional interaction data and the heterogeneous territorial monitoring data to generate a territorial integrated three-dimensional model; performing self-adaptive territorial space recursive partitioning processing on the territorial integrated three-dimensional model to generate self-adaptive partitioned territorial space data; marking a landmark index of the self-adaptive partitioned territorial space data to obtain index partitioned territorial space data; and analyzing index partition adjacency characteristic land change transfer trend data through index partition territorial space data, and executing territorial change monitoring intelligent feedback operation. According to the invention, through integration of heterogeneous territorial monitoring, efficient territorial monitoring data feedback is realized.

Description

Technical Field

[0001] This invention relates to the field of land monitoring data processing technology, and in particular to a method and system based on the integration of land monitoring data. Background Technology

[0002] Land monitoring is a crucial means of achieving efficient management and sustainable utilization of land resources, and it is widely applied in land planning, ecological protection, resource management, and urban construction. With the continuous development of monitoring technologies, satellite remote sensing, UAV monitoring, and geographic information systems (GIS) are widely used, providing multi-source, multi-dimensional, and multi-scale data support for land monitoring. The data generated by these technologies exhibits significant heterogeneity and diversity, and effectively integrating and utilizing this heterogeneous data has become a significant challenge in the field of land monitoring. However, the multi-source heterogeneous nature of existing land monitoring data brings complex data fusion problems, directly leading to inefficiencies in traditional analysis and modeling methods when processing multi-dimensional heterogeneous data. Spatiotemporal alignment between different monitoring technologies is difficult, and in terms of spatial zoning and indexing, it is impossible to flexibly adjust zoning according to the attributes and characteristics of land space, affecting the accuracy of regional land change monitoring and prediction, failing to fully reveal the dynamic trends of land use change between regions, and consequently resulting in low accuracy in locating ecologically abnormal land areas. Summary of the Invention

[0003] Based on this, the present invention provides a method and system for integrating land monitoring data to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method based on integrated land monitoring data includes the following steps: Step S1: Acquire heterogeneous land monitoring data; perform heterogeneous feature dimension interaction data analysis based on the heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; perform land integration three-dimensional model analysis based on the heterogeneous feature multidimensional interaction data and heterogeneous land monitoring data to generate a land integration three-dimensional model. Step S2: Extract land attribute features from the integrated 3D land model to generate land attribute feature data; perform adaptive recursive zoning of the integrated 3D land model based on the land attribute feature data to generate adaptive zoning land spatial data. Step S3: Calculate the minimum bounding space of the adaptive partitioned territorial spatial data to generate minimum bounding space data; initialize the minimum leaf node landmark index based on the minimum bounding space data to generate minimum leaf node landmark index data; analyze the decision mapping relationship between territorial space and spatial index based on the minimum leaf node landmark index data to generate a territorial spatial index decision tree model; transfer the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space to generate indexed partitioned territorial spatial data. Step S4: Perform time-series data analysis on land change in the indexed partition land space data to generate time-series data on land change in the indexed partition land space; perform adjacency characteristic land change transfer trend analysis on the time-series data on land change in the indexed partition land space to generate adjacency characteristic land change transfer trend data. Step S5: Based on the land change transfer trend data of the adjacency characteristics of the index partition, perform ecological suitability analysis of land change in the index partition to generate ecological suitability data of land change in the index partition; perform land change monitoring weight analysis of the index partition based on the ecological suitability data of land change in the index partition to generate land change monitoring weight data of the index partition; execute intelligent feedback operation of land change monitoring for each index partition through the land change monitoring weight data of the index partition.

[0005] Furthermore, step S1 includes the following steps: Step S11: Use satellite remote sensing system, UAV monitoring system and geographic information system respectively to collect and process heterogeneous land monitoring data of the target area of ​​land monitoring to obtain heterogeneous land monitoring data; Step S12: Perform temporal and spatial sequence analysis on heterogeneous land monitoring data to obtain heterogeneous land spatiotemporal sequence data; perform heterogeneous land monitoring data spatiotemporal alignment mapping on heterogeneous land monitoring data through heterogeneous land spatiotemporal sequence data to generate mapped land monitoring data; Step S13: Perform heterogeneous feature dimension interaction data analysis based on heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; Step S14: Based on the heterogeneous feature multidimensional interactive data, perform heterogeneous dimensional feature integration on the mapped land monitoring data to generate a land integrated three-dimensional model.

[0006] Furthermore, step S13 includes the following steps: Step S131: Perform heterogeneous type characteristic analysis on heterogeneous land monitoring data to generate heterogeneous type characteristic data; Step S132: Based on the heterogeneous spatiotemporal sequence data of the mapped land monitoring data, use the heterogeneous type dimension modeling benchmark sequence, and perform heterogeneous feature dimension tensor modeling processing through the heterogeneous type dimension modeling benchmark sequence and heterogeneous type feature data to generate heterogeneous feature dimension tensor modeling data. Step S133: Perform heterogeneous feature dimension interaction data analysis on the heterogeneous feature dimension tensor modeling data to generate heterogeneous feature multidimensional interaction data.

[0007] Furthermore, step S2 includes the following steps: Step S21: Extract land attribute features based on the integrated 3D land model to generate land attribute feature data; Step S22: Use the preset land division ratio data to perform preliminary zoning of the land integrated 3D model to generate zoning land spatial data; Step S23: Calculate the regional land attribute feature information entropy based on the land attribute feature data to generate regional land attribute feature information entropy data; Step S24: When the entropy data of the zonal land attribute features of the zonal land spatial data is greater than the preset threshold of the land interval entropy, the zonal land spatial data is recursively divided again using the preset land division ratio data, and the recursive zonal land spatial data is output; when the entropy data of the zonal land attribute features corresponding to the recursive zonal land spatial data of each region is not greater than the preset threshold of the land interval entropy or the spatial range of the recursive zonal land spatial data of each region is less than the preset minimum spatial threshold of land division, the recursive zonal land spatial division operation is ended, and the final output recursive zonal land spatial data is marked as adaptive zonal land spatial data.

[0008] Furthermore, step S3 includes the following steps: Step S31: Perform minimum bounding space calculation on the adaptive partitioned territorial spatial data to generate minimum bounding space data for the territorial spatial data; Step S32: Initialize the minimum leaf node landmark index based on the minimum enclosing spatial data of the territory to generate minimum leaf node landmark index data; Step S33: Perform a higher-level spatial zoning relationship analysis based on the adaptive zoning territorial spatial data to generate higher-level spatial zoning relationship data; Step S34: Based on the higher-level spatial zoning relationship data, perform higher-level landmark index leaf node analysis on the minimum leaf node landmark index data to generate higher-level landmark index leaf node data; Step S35: Analyze the decision mapping relationship between land space and spatial index based on the upper-level landmark index leaf node data and the corresponding minimum leaf node landmark index data, and generate a land space index decision tree model; Step S36: Transmit the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space, and generate indexed partitioned territorial spatial data.

[0009] Furthermore, step S4 includes the following steps: Step S41: Design time series hierarchical window data for land change monitoring, wherein the land change monitoring time series hierarchical window data includes fine-grained window data and coarse-grained window data; Step S42: Utilize the time-series hierarchical window data of land change monitoring to perform time-series data analysis on the indexed partition land space data, and generate time-series data on land change in the indexed partition land space; Step S43: Perform adjacency analysis of the indexed partition land space based on the indexed partition land space data, and generate indexed partition land space adjacency data; Step S44: Based on the indexed partition land space adjacency relationship data, perform indexed partition adjacency characteristic land change transfer trend analysis on the indexed partition land space land change time series data to generate indexed partition adjacency characteristic land change transfer trend data, wherein the indexed partition adjacency characteristic land change transfer trend data includes indexed partition adjacency characteristic land environment change transfer trend data and indexed partition adjacency characteristic land use change transfer trend data.

[0010] Furthermore, step S44 includes the following steps: Step S441: Based on the indexed partition land space adjacency relationship data and the indexed partition land space land change time series data, perform land space adjacency land change transfer impact coefficient analysis to generate land space adjacency land change transfer impact coefficient; Step S442: Based on the land change and transfer impact coefficient of adjacent land in national land space and the time series data of land change in indexed partitions, establish the land change and transfer network of adjacent characteristics of each indexed partition, and generate the land change and transfer network of adjacent characteristics of indexed partitions; Step S443: Perform an analysis of the land change and transfer trends of the index partitions based on the land change and transfer network of the adjacency characteristics of the index partitions, and generate land change and transfer trend data of the adjacency characteristics of the index partitions.

[0011] Furthermore, step S441 includes the following steps: Based on the adjacency relationship data of the indexed partition land space, a collinearity analysis of the land space adjacency change time series data of the indexed partition land space is performed to generate collinearity data of the land space adjacency change. Based on the collinearity data of adjacent land changes in national land space, we conduct a dependency characteristic analysis of adjacent land changes in national land space to generate dependency characteristic data of adjacent land changes in national land space. Based on the data on the characteristics of adjacent land changes in national land space, an analysis of the transfer impact coefficient of adjacent land changes in national land space is conducted to generate the transfer impact coefficient of adjacent land changes in national land space.

[0012] Furthermore, step S5 includes the following steps: Step S51: Based on the land environment change and transfer trend data of the adjacency characteristics of the index partition, perform a return period analysis of the land environment change trend of the index partition to generate return period data of the land environment change trend of the index partition. Step S52: Perform land use type analysis of the indexed partition based on the indexed partition land space data, and generate land use type data of the indexed partition; Step S53: Design land use type index evaluation decision based on indexed partitioned land use type data; Step S54: Use land use type indicators to evaluate the land use change trend data of the index partition adjacency characteristics and generate index partition land use change indicator evaluation data. Step S55: Perform ecological sensitivity analysis of land use change in the indexed partitions based on the land use change index assessment data, and generate ecological sensitivity data of land use change in the indexed partitions. Step S56: Based on the return period data of land environment change trends in the indexed partitions and the ecological sensitivity data of land change in the indexed partitions, conduct an ecological suitability analysis of land change in the indexed partitions to generate ecological suitability data of land change in the indexed partitions. Step S57: Perform land change monitoring weight analysis on the indexed partition based on the ecological suitability data of land change in the indexed partition, and generate land change monitoring weight data for the indexed partition; Step S58: Execute intelligent feedback operations for land change monitoring in each indexed partition using the indexed partition land change monitoring weight data.

[0013] This invention provides a system based on integrated land monitoring data for performing the method described above, the system comprising: The heterogeneous land integrated monitoring module is used to acquire heterogeneous land monitoring data; perform heterogeneous feature dimension interaction data analysis based on the heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; and perform land integrated three-dimensional model analysis based on heterogeneous feature multidimensional interaction data and heterogeneous land monitoring data to generate a land integrated three-dimensional model. The adaptive zoning module for national land space is used to extract national land attribute features from the integrated 3D model of national land and generate national land attribute feature data; and to perform adaptive recursive zoning of the integrated 3D model of national land space based on the national land attribute feature data to generate adaptive zoning national land space data. The territorial spatial index partitioning module is used to calculate the minimum bounding space of the adaptive partitioned territorial spatial data, generating minimum bounding space data; to initialize the minimum leaf node landmark index based on the minimum bounding space data, generating minimum leaf node landmark index data; to analyze the decision mapping relationship between territorial space and spatial index based on the minimum leaf node landmark index data, generating a territorial spatial index decision tree model; and to transfer the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space, generating indexed partitioned territorial spatial data. The Index Partition Adjacency Characteristics Land Change and Transfer Trend Analysis Module is used to perform time-series data analysis on land changes in index partition land space data, generating time-series data on land changes in index partition land space; and to perform adjacency characteristics land change and transfer trend analysis on the time-series data on land changes in index partition land space, generating trend data on land changes in index partition adjacency characteristics. The index partition land change monitoring weight analysis module is used to perform ecological suitability analysis of index partition land change based on the adjacency characteristics of land change transfer trend data, and generate ecological suitability data of index partition land change; perform land change monitoring weight analysis of index partition land change based on the ecological suitability data of index partition land change, and generate land change monitoring weight data of index partition land change; and execute intelligent feedback operation of land change monitoring for each index partition through the land change monitoring weight data of index partition land change.

[0014] The beneficial effects of this application are as follows: by utilizing satellite remote sensing systems, UAV monitoring systems, and geographic information systems to collect data on the target area of ​​land monitoring, comprehensive and multidimensional land monitoring data can be obtained, covering different spatial scales, time spans, and perspectives from different monitoring platforms, providing richer and more comprehensive information. Temporal and spatial sequence analysis is performed on heterogeneous land monitoring data to identify spatiotemporal patterns and trends among different monitoring data. Simultaneously, these spatiotemporal sequence data are used to perform spatiotemporal alignment mapping on heterogeneous land monitoring data, eliminating spatiotemporal differences between various data sources. Based on heterogeneous land monitoring data, heterogeneous feature dimension interaction data analysis is conducted to deeply analyze the interaction relationships between different data dimensions, deeply explore the potential connections between data, and integrate heterogeneous feature multidimensional interaction data to discover the interaction effects of different monitoring data in spatial, temporal, and feature dimensions. This effectively integrates heterogeneous data from different monitoring platforms, accurately reflecting the land spatial characteristics and changing trends of the target area. Extracting land attribute features from the integrated 3D land model enables a comprehensive analysis of the spatial attributes of the target area from multiple perspectives, reflecting key information such as land type, ecological characteristics, and resource distribution in each region, providing data support for subsequent spatial zoning. Based on the entropy data of the zoning land attribute features, a recursive partitioning strategy is adopted to further subdivide the land space. This recursive partitioning mechanism exhibits strong flexibility and adaptability when dealing with complex regions. When certain regions have high attribute feature entropy or significant spatial distribution differences, recursive partitioning can refine the spatial zoning and accurately reflect the characteristic differences of each region. Minimum bounding space calculation is performed on the adaptive zoning land spatial data to obtain the minimum bounding space of each region. The minimum bounding space data is then used for minimum leaf node landmark index initialization, providing efficient support for subsequent spatial indexing and data retrieval. The landmark index can quickly identify and locate key areas of the land space, reducing time and computational costs in the data processing process and improving the efficiency of data access and analysis. This study analyzes the hierarchical zoning relationships of national land space based on adaptive zoning data, and performs leaf node analysis of the minimum leaf node landmark index data based on the hierarchical zoning relationship data. Within complex spatial structures, the data is hierarchically processed and organized, ensuring that the spatial index better aligns with the structural characteristics of the higher-level spatial regions and that the index data better meets spatial structure and practical needs. Furthermore, a decision mapping relationship analysis is conducted between the leaf node data of the higher-level landmark index and the minimum leaf node landmark index data, organically combining spatial data indexing with spatial relationships. This efficiently executes spatial data retrieval and management, achieving precise spatial data positioning and efficient searching. This makes data from different regions easier to manage and analyze, optimizing the spatial decision-making process.By designing hierarchical time-series window data for land change monitoring and adjusting the time and spatial scales of analysis according to different monitoring needs, the monitoring process becomes more accurate and controllable, while enhancing the ability to capture details of land change. Using this hierarchical time-series window data, land change time-series data analysis and adjacent spatial land change analysis are performed on indexed partition land spatial data. This identifies the patterns of land use change and their underlying driving factors, reveals the interrelationships between different spatial units, and demonstrates the geographical connections between different partitions, helping to analyze how land change expands from one region to another. The transmission mechanism of land change between adjacent regions is explored, revealing the transmission paths and trends of land environmental change and land use change. In particular, the analysis of the transfer trend data of environmental change and land use change predicts whether the change trend in one area will affect neighboring areas, providing accurate predictions for land planning and environmental protection. Based on the adjacent characteristics of the indexed partition land environmental change transfer trend data, a land environmental change trend return period analysis is performed, generating indexed partition land environmental change trend return period data to predict the land environmental change cycle and return period in different regions, thereby helping planners identify and respond to environmental changes. By combining land use change trends with established assessment indicator systems, systematic analysis and evaluation can identify potential risk areas and areas with significant sustainable development potential, providing precise decision support for regional development. Based on land use change indicator assessment data from index zones, ecological sensitivity analysis of land change is conducted, generating ecological sensitivity data to identify areas where land change has a significant impact on the ecological environment, thus aiding in the prioritization of ecological protection. A comprehensive assessment of the ecological impact of land change from both ecological protection and land development perspectives ensures that development activities conform to the carrying capacity of the ecological environment, providing a scientific basis for sustainable land use. Based on land change ecological suitability data, a weight analysis of national land change monitoring is conducted to clarify which areas have a greater impact on the overall national land space. This allows for the concentration of resources and efforts on monitoring high-risk areas, optimizing monitoring efficiency, improving the responsiveness of the monitoring system, and implementing intelligent feedback operations for national land change monitoring in each index zone. This enables dynamic monitoring of land change, improves the timeliness and accuracy of land change monitoring, and ensures the sustainability of national land use.

[0015] Therefore, the method based on integrated land monitoring data in this invention solves the problem of data fusion with multi-source heterogeneity, achieving the fusion of land monitoring data with different spatial resolutions, temporal resolutions, data formats, and semantic levels. This efficiently processes multi-dimensional heterogeneous land monitoring data. By fully exploring the inherent relationships between heterogeneous monitoring land data, it ensures spatiotemporal alignment between different monitoring technologies, improving the comprehensive utilization capability of monitoring data. Furthermore, in terms of spatial zoning and indexing, it can flexibly adjust zoning according to the attribute characteristics of land space, ensuring the accuracy and applicability of land space zoning. This reduces the amount of data processed for monitoring land space data in each zone, improving the accuracy of regional land change monitoring and prediction. Through the correlation analysis between the adjacency characteristics of land space and land change, it comprehensively reveals the dynamic trends of land use change between regions, thereby improving the accuracy of locating ecologically abnormal land areas. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of a method based on integrated land monitoring data according to the present invention. Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S5. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0019] It should be understood that although the terms "first," "second," etc., are used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit is referred to as the second unit, and similarly, the second unit is referred to as the first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0020] To achieve the above objectives, please refer to Figures 1 to 2 This invention provides a method based on integrated land monitoring data. In the embodiments of this invention, please refer to... Figure 1 The diagram shown is a flowchart illustrating the steps of the method based on integrated land monitoring data according to the present invention. The method based on integrated land monitoring data includes the following steps: Step S1: Acquire heterogeneous land monitoring data; perform heterogeneous feature dimension interaction data analysis based on the heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; perform land integration three-dimensional model analysis based on the heterogeneous feature multidimensional interaction data and heterogeneous land monitoring data to generate a land integration three-dimensional model. In this embodiment of the invention, a satellite remote sensing system is used to monitor a large area of ​​the target region. The satellite remote sensing system uses high-resolution remote sensing imagery for data acquisition, with an image resolution of at least 1 meter to ensure clear identification of ground objects such as buildings, vegetation, roads, and lakes. An unmanned aerial vehicle (UAV) monitoring system is used to supplement data that is difficult to obtain via satellite, especially in densely populated urban areas or mountainous regions. UAVs are used for low-altitude flight to collect detailed ground data, with a flight altitude not exceeding 500 meters and a data acquisition frequency of 10 frames per second to ensure high spatial resolution. A Geographic Information System (GIS) is used to integrate data from different sources for spatial analysis and map generation. The GIS system first converts the raw imagery data from remote sensing and UAVs into GeoTIFF format to ensure that the data can be compared and analyzed under a unified coordinate system. Comprehensive monitoring of the land at different spatial scales and accuracies yields heterogeneous monitoring data, providing foundational data for subsequent data analysis and processing. Time-series analysis is then performed on the heterogeneous land monitoring data. Taking satellite remote sensing data as an example, time-series data analysis methods (such as time-series smoothing and trend analysis) are used to analyze changes over a period of time and extract the time series of land cover changes. Spatial sequence analysis is based on UAV monitoring data and GIS data, using spatial autocorrelation analysis methods (such as Moran's I index) to perform spatial correlation analysis on data from different areas of the ground, obtaining the spatial distribution characteristics of land change in each area. Combining the time-series analysis results, spatiotemporal alignment mapping of heterogeneous data is performed to generate heterogeneous land spatiotemporal sequence data. Heterogeneous feature dimension interaction analysis is then performed based on heterogeneous land monitoring data. For example, satellite remote sensing data contains multiple features such as land cover, vegetation index, and urban heat island effect, while UAV monitoring data contains more detailed information such as building types and land use functions. When performing interactive analysis on these features, multidimensional data analysis techniques (such as principal component analysis (PCA) and factor analysis) are used to reduce the dimensionality of different features, extract the main features, and generate a heterogeneous feature multidimensional interactive dataset. Based on the heterogeneous feature multidimensional interactive data, 3D modeling technology is used to perform three-dimensional land modeling. This process utilizes GIS and remote sensing technologies, combining topographic data, land use information, and spatial analysis results to construct a realistic 3D national land model. For example, spatial interpolation methods (such as Kriging or IDW interpolation) are used to map elevation data, land use types, and other features of different regions into 3D space, constructing a 3D model for each region. Furthermore, information such as building height and shape provided by UAV data can be combined to further refine the urban structure in the model. Ultimately, the generated integrated 3D national land model will include multiple data dimensions such as ground morphology, building structure, and road network, providing a visualized and comprehensive national land data support platform for land planning, ecological environment monitoring, and urban construction.

[0021] Step S2: Extract land attribute features from the integrated 3D land model to generate land attribute feature data; perform adaptive recursive zoning of the integrated 3D land model based on the land attribute feature data to generate adaptive zoning land spatial data. In this embodiment of the invention, the integrated 3D model of the national territory is analyzed to extract attribute data related to national territory characteristics. These include topography, land use type, building density, vegetation cover, and soil type. The integrated 3D model of the national territory is then spatially partitioned according to a preset national territory division ratio. During implementation, a reasonable partitioning ratio is first determined by analyzing existing land use data, topographic features, and regional spatial requirements. Assuming a preset partitioning ratio of 1:4, the national territory is divided into four equal parts. Then, 3D modeling tools, such as ArcGIS's spatial analysis module, are used to divide different geographic units in the integrated 3D model of the national territory. The partitioning data generated for each unit includes the geographic coordinates, area, and other information of the region, forming partitioned national territory spatial data. The information entropy of the national territory attribute characteristics of each partition is calculated to quantify the heterogeneity within each region. Specifically, this involves collecting attribute characteristic data for each partition, such as land use type, building density, and vegetation cover, and calculating the attribute characteristic information entropy for each partition based on this data. Information entropy is an indicator that measures the complexity of the distribution of various attributes within a partition. A higher entropy value indicates a more uniform distribution of attributes and higher heterogeneity within the region. Generating attribute characteristic information entropy data for each partition reflects its spatial heterogeneity. This data will provide decision support in subsequent spatial partitioning and adjustment processes, helping to optimize the partitioning scheme. Based on the attribute characteristic information entropy data of each partition, it is determined whether further recursive partitioning of certain partitions is necessary. First, a preset information entropy threshold is set, for example, 0.8. If the attribute characteristic information entropy of a partition is greater than this threshold, it indicates that the region has high spatial heterogeneity and requires more refined partitioning. For example, if the entropy value of a partition is 0.9, then recursive partitioning is required. During the recursive partitioning process, the partition is spatially divided again using a preset land division ratio, that is, the currently analyzed block of land space is divided into four equal parts again. Each newly generated sub-region is re-evaluated based on its new information entropy value. If the information entropy value of a recursive partition is still greater than the threshold, and its spatial range is greater than the preset minimum spatial threshold (e.g., 10 square kilometers), the partitioning continues until the stopping condition is met. If the information entropy of a recursive partition is less than or equal to the threshold, or its spatial range is less than the minimum spatial threshold, the recursive partitioning stops, and the partition is marked as adaptive partitioned territorial spatial data. For example, if the spatial range of a partition is 8 square kilometers and its information entropy is lower than the preset threshold, the partitioning of that partition stops, and it is marked as final adaptive partitioned territorial spatial data. Ultimately, the output recursive partitioned territorial spatial data consists of a series of spatial units that meet the adaptive requirements, have low heterogeneity, and conform to the preset spatial and information entropy thresholds.

[0022] Step S3: Calculate the minimum bounding space of the adaptive partitioned territorial spatial data to generate minimum bounding space data; initialize the minimum leaf node landmark index based on the minimum bounding space data to generate minimum leaf node landmark index data; analyze the decision mapping relationship between territorial space and spatial index based on the minimum leaf node landmark index data to generate a territorial spatial index decision tree model; transfer the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space to generate indexed partitioned territorial spatial data. In this embodiment of the invention, based on adaptive partitioned land space data, the minimum bounding space of the land is calculated to find the minimum bounding space that can completely enclose each adaptive partition region. Specifically, the geometric boundary of each partition region is first determined, and the minimum bounding space of each partition region is calculated using the convex hull algorithm. For partitions with polygonal boundaries, the bounding space is determined by calculating the vertices of the polygons, and the generated minimum bounding space data includes the bounding rectangle boundary information of each partition. The minimum bounding space data is then used to initialize the minimum leaf node landmark index of the adaptive partitioned land space. Specifically, based on the minimum bounding space of the partition, a quadtree or R-tree algorithm is used for spatial indexing. Within the partitioned space, each leaf node represents a specific area, and its coordinates are determined by the bounding box of the minimum bounding space. Taking a quadtree as an example, the entire partitioned space is first divided into four sub-regions, and then each sub-region is recursively divided into smaller regions until a preset resolution or spatial size requirement is reached. Through this process, the minimum leaf node landmark index data is initialized, including the spatial range of each leaf node and its corresponding index number. Spatial relationship analysis is performed on adaptive zoning land spatial data to determine the relationship between each zoning and its superior zoning. Spatial intersection analysis (e.g., based on buffer analysis) is used to determine the hierarchical relationship between adjacent zoning. For example, within a zoning space, there are multiple sub-zoning spaces. Intersection algorithms are used to determine the boundary relationships of each zoning, ultimately generating superior zoning relationship data for land spatial data, including the parent-child relationships of each zoning and their spatial overlap. Further analysis is conducted based on the superior zoning relationship data to determine the relationship between the minimum leaf node landmark index data and the superior zoning. Spatial overlap analysis is used to determine which minimum leaf nodes belong to specific superior zoning. For example, for a small area containing multiple leaf nodes, its spatial relationship with the entire urban area is analyzed. By calculating the intersection area or overlap, the superior landmark index leaf node data is updated to accurately assign each leaf node to its corresponding superior zoning. Combining the superior landmark index leaf node data and the minimum leaf node landmark index data, a decision tree algorithm is used to analyze the spatial characteristics of each leaf node (such as location, size, adjacency, etc.) to generate a spatial feature-based decision mapping model. For example, the landmark index number of a certain region is related to the spatial shape and area of ​​that region. Therefore, a decision tree model is used for data mapping to generate a spatial index number corresponding to each spatial region. Adaptive zoning land spatial data is transmitted to the land spatial index decision tree model, and this model is used to process the landmark index identification of the zoning land spaces. Using the mapping rules generated in the decision tree model, the data of each zoning region is matched with the corresponding landmark index, ultimately generating zoning land spatial data containing spatial index numbers.For example, for a certain urban area, the model identifies its corresponding landmark index as "X1" and marks it as the spatial index of that area, along with the latitude and longitude of the corresponding landmark index. The output of this process is indexed zonal spatial data, which provides efficient spatial indexing information for subsequent spatial analysis and monitoring tasks.

[0023] Step S4: Perform time-series data analysis on land change in the indexed partition land space data to generate time-series data on land change in the indexed partition land space; perform adjacency characteristic land change transfer trend analysis on the time-series data on land change in the indexed partition land space to generate adjacency characteristic land change transfer trend data. In this embodiment of the invention, a time-series hierarchical window data for land change monitoring is designed, including fine-grained window data and coarse-grained window data. Fine-grained window data is used to capture land changes within shorter time periods, while coarse-grained window data is suitable for capturing change trends over longer periods. The fine-grained and coarse-grained window data are respectively input into the land change detection model. For example, assuming remote sensing image difference analysis, such as change vector analysis, is used to identify changes in land type, information about the changed areas is extracted by comparing image data from different time windows. For the fine-grained window, the analysis results show land use changes occurring in a specific area within a month, such as farmland being converted to construction land; while the coarse-grained window reveals broader change trends, such as a decrease in green space area in a region within a year. The analysis results output land change data for each time period, specifically including the type of change, the area of ​​change, and its corresponding spatial location. Through spatial adjacency analysis, the boundary relationships between adjacent zones are identified, determining which areas are spatially adjacent. By calculating the spatial geometric relationships between adjacent areas, for example using buffer analysis, a buffer is created on the boundary of each partition, and adjacency relationships are defined by the intersection of these buffers. The generated adjacency relationship data includes information such as adjacent area pairs, intersection area, and adjacent boundary length, providing detailed data on spatial relationships and supporting subsequent land change analysis. Spatial analysis of land change trends, combined with the change characteristics of adjacent areas, predicts the transfer trends of land change. In practice, fine-grained and coarse-grained land change time-series data are first integrated to analyze land change patterns in adjacent areas. By performing regression analysis or Markov chain models on these change data, the transfer trends of land change can be extracted. Specifically, the indexed partition adjacency characteristic land change transfer trend data will include the following: land environment change transfer trend data, showing the trend of changes in regional environmental conditions (such as water bodies and green spaces); land use change transfer trend data, showing the transformation trend between different land use types by considering the impact of adjacent land use changes, such as the conversion of agricultural land to construction land. The generated indexed partition adjacency characteristic land change transfer trend data is used to analyze and predict the future land change trends of each partition.

[0024] Step S5: Based on the land change transfer trend data of the adjacency characteristics of the index partition, perform ecological suitability analysis of land change in the index partition to generate ecological suitability data of land change in the index partition; perform land change monitoring weight analysis of the index partition based on the ecological suitability data of land change in the index partition to generate land change monitoring weight data of the index partition; execute intelligent feedback operation of land change monitoring for each index partition through the land change monitoring weight data of the index partition.

[0025] In this embodiment of the invention, based on land environment change and transfer trend data, statistical methods, such as extreme value theory or regression analysis, are used to conduct return period analysis. The goal of return period analysis is to predict the same type of land environment change within a specific time period based on historical change data. Return period methods in statistics (such as Poisson regression models or extreme value theory) are used to calculate the average time interval between change events. The generated return period data can provide a time expectation of land change trends and the environmental conditions during the return period, helping to consider the potential risks of land environment change in planning. Land use types are analyzed based on indexed partitioned land spatial data. This analysis uses spatial data (such as remote sensing imagery or land use layers) to classify the use type of each plot, typically into categories such as cultivated land, forest land, construction land, grassland, and water area. By classifying and processing this spatial data, land use type data within each indexed partition is generated. Based on the land use type data, a series of indicators are designed to assess the impact and sustainability of land use types. These indicators include land use intensity, ecological carrying capacity, and land use efficiency. Through these indicators, the health of land use types and potential environmental risks can be assessed. Based on land use type data and related environmental and economic indicators, a multiple regression analysis was conducted to design an assessment model suitable for the region's land use sustainability. Through the analysis of land use type indicators, a sustainability assessment of land use in the region was provided, assisting decision-makers in formulating reasonable land use plans. Based on the land use type indicator assessment, further analysis was conducted on the adjacent characteristic land use change transfer trend data of the indexed zones. By assessing land use change trends, future land use changes were predicted, and corresponding assessment data were generated. The assessment process relies on the aforementioned land use type indicators, combined with adjacent characteristic data, and uses regression models or time series analysis methods to assess the impact and potential risks of land change in adjacent areas. Based on the indexed zone land use change indicator assessment data, the ecological sensitivity of land change was analyzed. This analysis, based on the land use change situation of each zone, identified the impact of land use change on the ecological environment, focusing on areas vulnerable to environmental changes. In practice, an ecological sensitivity model was established, combined with regional ecological data, to assess the impact of land change on the ecosystem. Ecological suitability analysis combined the return period data of land environmental change trends with ecological sensitivity data to assess the suitability of different land types and change trends. For example, if a region has high ecological sensitivity and a short return period for land use change, it indicates that the region may face high ecological pressure. Furthermore, analyzing the extreme impacts of the environment on ecological sensitivity within the return period helps assess whether the region is suitable for continued development or other types of land use. By comprehensively considering factors such as ecological environmental protection needs, the temporality of land use change, and the sustainability of change, ecological suitability data for land use change can be obtained.Based on the aforementioned ecological suitability analysis data, a monitoring weight analysis of land change in each indexed zone is conducted. This analysis aims to determine the importance of different regions in land use change monitoring; the weight reflects the region's criticality in ecological protection, resource utilization, and other aspects. Through comprehensive analysis, land use change monitoring weight data for each indexed zone is generated. Using this weight data, an intelligent feedback operation for land use change monitoring in each indexed zone is executed. This operation, through an automated system, combines the weight data of each region to monitor land use changes in real time and prioritizes feedback on changes in key areas based on the weight analysis results. During the operation, the monitoring system continuously assesses the change trends in each zone, and immediately provides feedback and activates an early warning mechanism upon detecting potential environmental risks or changes exceeding thresholds. Based on the land use change monitoring weight data for each indexed zone, the system frequently monitors urban expansion areas, especially surrounding ecologically sensitive areas, and updates changes in real time through an intelligent feedback mechanism to ensure the accuracy and timeliness of land use change monitoring and avoid resource waste or environmental damage.

[0026] Furthermore, step S1 includes the following steps: Step S11: Use satellite remote sensing system, UAV monitoring system and geographic information system respectively to collect and process heterogeneous land monitoring data of the target area of ​​land monitoring to obtain heterogeneous land monitoring data; In this embodiment of the invention, a satellite remote sensing system is used to monitor a large area of ​​the target region. The satellite remote sensing system uses high-resolution remote sensing imagery for data acquisition, with an image resolution of at least 1 meter to ensure clear identification of ground objects such as buildings, vegetation, roads, and lakes. An unmanned aerial vehicle (UAV) monitoring system is used to supplement data that is difficult to obtain via satellite, especially in densely populated urban areas or mountainous regions. UAVs are used for low-altitude flight to collect detailed ground data, with a flight altitude not exceeding 500 meters and a data acquisition frequency of 10 frames per second to ensure high spatial resolution. A Geographic Information System (GIS) is used to integrate data from different sources for spatial analysis and map generation. The GIS system first converts the raw imagery data from remote sensing and UAVs into GeoTIFF format to ensure that the data can be compared and analyzed within a unified coordinate system. Comprehensive monitoring of the land at different spatial scales and accuracies yields heterogeneous monitoring data, providing foundational data for subsequent data analysis and processing.

[0027] Step S12: Perform temporal and spatial sequence analysis on heterogeneous land monitoring data to obtain heterogeneous land spatiotemporal sequence data; perform heterogeneous land monitoring data spatiotemporal alignment mapping on heterogeneous land monitoring data through heterogeneous land spatiotemporal sequence data to generate mapped land monitoring data; In this embodiment of the invention, time-series analysis is performed on heterogeneous land monitoring data. Taking satellite remote sensing data as an example, time-series data analysis methods (such as time-series smoothing and trend analysis) are used to analyze changes over a period of time and extract the time series of land cover changes. Spatial sequence analysis is based on UAV monitoring data and GIS data. Spatial autocorrelation analysis methods (such as Moran's I index) are used to perform spatial correlation analysis on data from different areas of the ground to obtain the spatial distribution characteristics of land changes in each area. Combining the time-series analysis results, spatiotemporal alignment mapping of heterogeneous data is performed. Specifically, a spatiotemporal grid model is used to unify the processing of data from different sources, generating a dataset with consistent coordinates and time standards. The spatiotemporal alignment process ensures that different data sources have consistent geographic identifiers at the same time points, generating heterogeneous land spatiotemporal sequence data, which includes information on land use, vegetation changes, building additions, etc., at different time and spatial points, providing a spatiotemporally integrated data foundation for subsequent data analysis.

[0028] Step S13: Perform heterogeneous feature dimension interaction data analysis based on heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; In this embodiment of the invention, heterogeneous feature dimension interaction analysis is performed based on heterogeneous land monitoring data. Taking satellite remote sensing data as an example, it contains multiple features such as land cover, vegetation index, and urban heat island effect, while UAV monitoring data contains more detailed information such as building types and land use functions. When performing interaction analysis on these features, multidimensional data analysis techniques (such as principal component analysis (PCA) and factor analysis) are used to reduce the dimensionality of different features and extract the main features. For example, when performing land use change analysis, PCA is used to combine land use types (such as farmland, forest, urban construction land, etc.) with relevant features such as vegetation index and heat island effect from satellite images to obtain the comprehensive features of each region. After heterogeneous feature dimension interaction analysis, a heterogeneous feature multidimensional interaction dataset is finally generated, specifically containing interaction information on multiple dimensions such as land use type, vegetation cover, climate change, and urbanization process, providing multi-angle data support for subsequent modeling and prediction.

[0029] Step S14: Based on the heterogeneous feature multidimensional interactive data, perform heterogeneous dimensional feature integration on the mapped land monitoring data to generate a land integrated three-dimensional model.

[0030] In this embodiment of the invention, based on heterogeneous multidimensional interactive data, 3D modeling technology is used to perform three-dimensional land modeling. This process utilizes GIS and remote sensing technologies, combining topographic data, land use information, and spatial analysis results to construct a realistic three-dimensional land model. For example, spatial interpolation methods (such as Kriging or IDW interpolation) are used to map elevation data, land use types, and other features of different regions into three-dimensional space, constructing a three-dimensional model for each region. Furthermore, information such as building height and shape provided by UAV data can be combined to further refine the urban structure in the model. Ultimately, the generated integrated three-dimensional land model will include multiple data dimensions such as ground morphology, building structure, and road network, providing a visualized and three-dimensional land data support platform for land planning, ecological environment monitoring, and urban construction. Through the implementation of these specific steps, the integration and application of multi-source heterogeneous data are ultimately realized, thereby providing comprehensive, accurate, and three-dimensional technical support for land monitoring and management.

[0031] Furthermore, step S13 includes the following steps: Step S131: Perform heterogeneous type characteristic analysis on heterogeneous land monitoring data to generate heterogeneous type characteristic data; In this embodiment of the invention, heterogeneous type feature analysis is performed on heterogeneous land monitoring data. The main goal of heterogeneous type feature analysis is to extract representative features from different data sources (such as satellite remote sensing, UAV monitoring, and geographic information systems) for subsequent multidimensional data fusion and modeling. Satellite remote sensing data is first used to extract land cover types (such as forests, cultivated land, water bodies, and urban construction land). Supervised classification algorithms (such as Support Vector Machines (SVM) or Random Forest classifiers) are used to classify the images and generate land cover type data for each pixel. UAV monitoring data can be used to refine the extraction of building types, vegetation types, and land use functions within the area. Deep learning models (such as Convolutional Neural Networks (CNN)) are used to automatically classify high-resolution images taken by UAVs. Based on these analysis results, combined with land use data in GIS, the features from different monitoring sources are integrated to obtain heterogeneous type feature data that includes land cover type, building type, and land use function. For example, satellite remote sensing data of a certain urban area identifies approximately 60% of the area as urban construction land, while UAV monitoring data provides detailed information such as building height and density. Ultimately, heterogeneous feature data contains multidimensional feature information across different spatial and temporal scales.

[0032] Step S132: Based on the heterogeneous spatiotemporal sequence data of the mapped land monitoring data, use the heterogeneous type dimension modeling benchmark sequence, and perform heterogeneous feature dimension tensor modeling processing through the heterogeneous type dimension modeling benchmark sequence and heterogeneous type feature data to generate heterogeneous feature dimension tensor modeling data. In this embodiment of the invention, heterogeneous land spatiotemporal sequence data includes temporal trends and spatial geographic distribution. Taking satellite remote sensing data as an example, satellite imagery reflects changes in land use at different time points. Through time series analysis, data on changes in land cover types in different areas at each time point are obtained. UAV monitoring data provides spatial information for each monitoring point, such as building type and land use function. This data is integrated through a spatiotemporal coordinate system to generate a heterogeneous type dimensional modeling benchmark sequence based on time series and spatial coordinates. This benchmark sequence effectively aligns the heterogeneous monitoring data through time index and spatial coordinates, ensuring that the monitoring data corresponding to each time point and spatial location has a consistent structure and dimension. Based on this, heterogeneous feature dimension tensor modeling processing is performed. In specific operation, tensor modeling methods (such as higher-order tensor decomposition) are used to process the data, mapping different heterogeneous data (such as land use, building type, vegetation cover, etc.) into a high-dimensional tensor space. Each tensor dimension corresponds to different time, space, and feature types. In this way, through multi-dimensional interactive analysis, the relationships between different features and their patterns of change over time are revealed. For example, satellite imagery of a certain urban area shows large-scale urban expansion over a certain period of time, while drone monitoring data shows an increase in building density. By using tensor modeling, these heterogeneous data can be fused together and their spatiotemporal trends can be analyzed.

[0033] Step S133: Perform heterogeneous feature dimension interaction data analysis on the heterogeneous feature dimension tensor modeling data to generate heterogeneous feature multidimensional interaction data.

[0034] In this embodiment of the invention, heterogeneous feature-dimensional tensor modeling data is used for further interactive analysis. The purpose of this analysis is to uncover the interrelationships between different feature dimensions and how they interact at different temporal and spatial scales. During implementation, an interactive tensor decomposition method (such as CP decomposition or Tucker decomposition) is first used to decompose the high-dimensional tensor into multiple low-dimensional tensors, from which interactive information between different dimensions (such as time, space, and features) is extracted. Correlation between different heterogeneous monitoring data is extracted to reveal the impact of land use type changes in a specific area on the surrounding environment within a specific time period. Taking urban expansion as an example, if there is a strong correlation between the increase in buildings and the decrease in agricultural use of surrounding land in a certain area, interactive analysis can quantify this trend and provide corresponding predictions. In this analysis, the change patterns of each feature in the temporal and spatial dimensions are calculated using interrelated feature dimensions, and heterogeneous feature multidimensional interactive data are generated based on the analysis results. For example, the land use conversion in a certain area changes from agricultural land to urban land, accompanied by a decrease in vegetation cover. Through interactive analysis, multidimensional interactive data on the impact of land change on the ecological environment and the contribution of building density to the urban heat island effect are generated. This data provides a foundation for subsequent prediction and decision analysis. Ultimately, heterogeneous multidimensional interactive data can effectively describe the complex relationships between various dimensions, providing important evidence for the management, planning, and forecasting of land resources.

[0035] Furthermore, step S2 includes the following steps: Step S21: Extract land attribute features based on the integrated 3D land model to generate land attribute feature data; In this embodiment of the invention, an integrated 3D model of the national territory is analyzed to extract attribute data related to national territory characteristics. These attribute data include topography, land use type, building density, vegetation cover, and soil type. Using geographic information and spatial data from the 3D model, the elevation and soil type at each location are estimated through spatial interpolation. Simultaneously, by combining satellite remote sensing imagery and UAV data, different land use types (such as farmland, urban areas, and forests) are identified. For example, in a certain region, data extracted from remote sensing imagery shows that the main land cover type is arable land, accounting for approximately 70% of the area, while urban construction land accounts for 30%. Through these analyses, attributes such as land use type, soil information, and building characteristics are integrated to form national territory attribute feature data, providing basic data support for subsequent spatial analysis.

[0036] Step S22: Use the preset land division ratio data to perform preliminary zoning of the land integrated 3D model to generate zoning land spatial data; In this embodiment of the invention, the integrated 3D model of the national territory is spatially partitioned according to a preset national territory division ratio. During implementation, firstly, a reasonable partitioning ratio is determined by analyzing existing land use data, terrain features, and regional spatial requirements. Assuming the preset partitioning ratio is 1:4, the national territory is divided into four equal parts. Then, using 3D modeling tools, such as ArcGIS's spatial analysis module, different geographic units in the integrated 3D model of the national territory are divided. The partitioning data generated for each unit includes the geographic coordinates, area, and other information of the region, forming partitioned national territory spatial data. This partitioning data provides the spatial basis for subsequent spatial analysis.

[0037] Step S23: Calculate the regional land attribute feature information entropy based on the land attribute feature data to generate regional land attribute feature information entropy data; In this embodiment of the invention, the information entropy of the land attribute characteristics of different zones is calculated to quantify the heterogeneity within each region. Specifically, this involves collecting attribute characteristic data for each zone, such as land use type, building density, and vegetation cover, and calculating the attribute characteristic information entropy for each zone based on this data. Information entropy is an indicator that measures the complexity of the distribution of various attributes within a zone; a higher entropy value indicates a more uniform distribution of attributes within the region and higher heterogeneity. First, several important land attributes are selected for analysis, such as land use type (e.g., arable land, urban, forest), building density, and vegetation cover. This data is obtained through satellite remote sensing images and UAV monitoring data. Next, based on this data, the probability distribution of various attributes within each zone is calculated. For example, in a typical urban area, land use mainly includes urban construction land, road networks, and public facilities, with agricultural land accounting for a smaller proportion. Based on this data, the probability of each attribute within the zone is calculated. For example, in a certain zone, assuming arable land accounts for 70% and urban land accounts for 30%, the attribute entropy value of that zone is obtained through probability distribution calculation. If the calculated information entropy value is 0.65, it indicates that the land use in the area is relatively simple and the heterogeneity is low; while a value of 0.9 indicates that the land use types in the area are more complex and the heterogeneity is higher. This process generates attribute characteristic information entropy data for each partition, reflecting the spatial heterogeneity of each partition. This data will provide decision support in subsequent spatial partitioning and adjustment processes, helping to optimize the partitioning scheme.

[0038] Step S24: When the entropy data of the zonal land attribute features of the zonal land spatial data is greater than the preset threshold of the land interval entropy, the zonal land spatial data is recursively divided again using the preset land division ratio data, and the recursive zonal land spatial data is output; when the entropy data of the zonal land attribute features corresponding to the recursive zonal land spatial data of each region is not greater than the preset threshold of the land interval entropy or the spatial range of the recursive zonal land spatial data of each region is less than the preset minimum spatial threshold of land division, the recursive zonal land spatial division operation is ended, and the final output recursive zonal land spatial data is marked as adaptive zonal land spatial data.

[0039] In this embodiment of the invention, the information entropy data of the attribute features of each partition is used to determine whether recursive partitioning of certain partitions is necessary. Specifically, a preset information entropy threshold is first set, for example, 0.8. If the information entropy of the attribute features of a partition is greater than this threshold, it indicates that the region has high spatial heterogeneity and requires more refined partitioning. For example, if the entropy value of a partition is 0.9, recursive partitioning is required. During the recursive partitioning process, the partition is spatially partitioned again using a preset land division ratio, that is, the currently analyzed block of land space is divided into four equal parts again. Each newly generated sub-region is re-evaluated based on its new information entropy value. If the information entropy value of a partition after recursive partitioning is still greater than the threshold, and its spatial range is greater than a preset minimum spatial threshold (e.g., 10 square kilometers), the partitioning continues until the stopping condition is met. If the information entropy of a partition after recursive partitioning is less than or equal to the threshold, or its spatial range is less than the minimum spatial threshold, the recursive partitioning stops, and the partition is marked as adaptive partitioned land space data. For example, if a partition has a spatial extent of 8 square kilometers and its information entropy is below a preset threshold, the partitioning of that partition is stopped, and it is marked as the final adaptive partitioned land spatial data. Ultimately, the output recursive partitioned land spatial data consists of a series of spatial units that meet the adaptation requirements, exhibit low heterogeneity, and conform to preset spatial and information entropy thresholds. This data will be used for further analysis and modeling.

[0040] Furthermore, step S3 includes the following steps: Step S31: Perform minimum bounding space calculation on the adaptive partitioned territorial spatial data to generate minimum bounding space data for the territorial spatial data; In this embodiment of the invention, the minimum bounding space of the land is calculated based on adaptive partitioned land spatial data. The purpose is to find the minimum circumscribed space that can completely enclose each adaptive partition area. Specifically, the geometric boundary of each partition area is first determined, and the minimum bounding space of each partition area is calculated using the convex hull algorithm. For partitions with polygonal boundaries, the bounding space is determined by calculating the vertices of the polygon. Taking a long strip-shaped land partition as an example, let the vertex coordinates of the partition be (x1,y1), (x2,y2), (x3,y3). The minimum bounding space is calculated using this method, and the result is the smallest rectangle or polygon containing the partition. The final generated minimum bounding space data includes the bounding rectangle boundary information of each partition (such as length, width, area, etc.), providing a spatial basis for subsequent data processing.

[0041] Step S32: Initialize the minimum leaf node landmark index based on the minimum enclosing spatial data of the territory to generate minimum leaf node landmark index data; In this embodiment of the invention, minimum leaf node landmark indexes are initialized for adaptive partitioned land space using minimum bounding space data. Specifically, the method is based on the minimum bounding space of the partition, employing a quadtree or R-tree algorithm for spatial indexing. Within the partitioned space, each leaf node represents a specific region, and its coordinates are determined by the bounding box of the minimum bounding space. Taking a quadtree as an example, the entire partitioned space is first divided into four sub-regions, and then each sub-region is recursively divided into smaller regions until a preset resolution or spatial size requirement is reached. Through this process, the minimum leaf node landmark index data is initialized, including the spatial extent of each leaf node and its corresponding index number.

[0042] Step S33: Perform a higher-level spatial zoning relationship analysis based on the adaptive zoning territorial spatial data to generate higher-level spatial zoning relationship data; In this embodiment of the invention, spatial relationship analysis is performed on adaptive partitioned territorial spatial data to determine the relationship between each partition and its superior partition. Spatial intersection analysis (e.g., based on buffer analysis) is used to determine the hierarchical relationship between adjacent partitions. For example, within a partitioned space, there are multiple sub-partitioned spaces. Through an intersection algorithm, the boundary relationships of each partition are determined, ultimately generating territorial spatial superior partition relationship data, including the parent-child partition relationships of each partition and the spatial overlap between them.

[0043] Step S34: Based on the higher-level spatial zoning relationship data, perform higher-level landmark index leaf node analysis on the minimum leaf node landmark index data to generate higher-level landmark index leaf node data; In this embodiment of the invention, the relationship between the minimum leaf node landmark index data and the higher-level zoning data is further analyzed based on the higher-level spatial zoning relationship data. Spatial overlap analysis is used to determine which minimum leaf nodes belong to specific higher-level zonings. For example, for a small area containing multiple leaf nodes, its spatial relationship with the entire urban area is analyzed. By calculating the intersection area or overlap degree, the higher-level landmark index leaf node data is updated to accurately assign each leaf node to its corresponding higher-level zoning.

[0044] Step S35: Analyze the decision mapping relationship between land space and spatial index based on the upper-level landmark index leaf node data and the corresponding minimum leaf node landmark index data, and generate a land space index decision tree model; In this embodiment of the invention, the decision mapping relationship between land space and spatial indexes is analyzed by combining the leaf node data of the higher-level landmark index and the landmark index data of the lowest-level leaf node. The goal of this step is to create a decision tree model to establish the mapping relationship between spatial data and its index. Using a decision tree algorithm, the spatial characteristics (such as location, size, adjacency, etc.) of each leaf node are analyzed to generate a decision mapping model based on spatial characteristics. For example, the landmark index number of a certain region is related to the spatial shape and area of ​​that region; therefore, the decision tree model is used to map the data and generate the spatial index number corresponding to each spatial region.

[0045] Step S36: Transmit the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space, and generate indexed partitioned territorial spatial data.

[0046] In this embodiment of the invention, adaptive partitioned territorial spatial data is transmitted to a territorial spatial index decision tree model, and the model is used to perform landmark index identification processing on the partitioned territorial spaces. Using the mapping rules generated in the decision tree model, the data of each partition area is matched with the corresponding landmark index, ultimately generating partitioned territorial spatial data containing spatial index numbers. For example, for a certain urban area, the model identifies its corresponding landmark index as "X1," identifies it as the spatial index of that partition, and identifies the latitude and longitude of the corresponding landmark index. The output of this process is indexed partitioned territorial spatial data, which provides efficient spatial index information for subsequent spatial analysis and monitoring tasks.

[0047] Furthermore, step S4 includes the following steps: Step S41: Design time series hierarchical window data for land change monitoring, wherein the land change monitoring time series hierarchical window data includes fine-grained window data and coarse-grained window data; In this embodiment of the invention, a time-series hierarchical window data for land change monitoring is designed, including fine-grained window data and coarse-grained window data. Fine-grained window data is used to capture land changes within shorter time periods, while coarse-grained window data is suitable for capturing trends over longer periods. For example, a fine-grained window is set to one month or one quarter, collecting land change data monthly using satellite remote sensing data; while a coarse-grained window is set to six months or one year to analyze land changes on a larger scale. Assuming the fine-grained window collects data monthly, the data includes information such as land use status, vegetation cover, and building distribution for each region. The coarse-grained window statistically analyzes land use change trends every six months or year, encompassing change data for a large area. By designing monitoring windows of different granularities, patterns of land change can be captured at multiple time scales.

[0048] Step S42: Utilize the time-series hierarchical window data of land change monitoring to perform time-series data analysis on the indexed partition land space data, and generate time-series data on land change in the indexed partition land space; In this embodiment of the invention, fine-grained window data and coarse-grained window data are respectively input into the land change detection model. For example, assuming remote sensing image difference analysis, such as change vector analysis, is used to identify changes in land type, information about the changed areas is extracted by comparing image data from different time windows. For the fine-grained window, the analysis results show land use changes that occur in a specific area within a month, such as farmland being converted to building land; while the coarse-grained window reveals broader trends, such as a decrease in green space area in a region within a year. The analysis results are output as land change data for each time period, specifically including the type of change, the area of ​​change, and its corresponding spatial location.

[0049] Step S43: Perform adjacency analysis of the indexed partition land space based on the indexed partition land space data, and generate indexed partition land space adjacency data; In this embodiment of the invention, spatial adjacency analysis identifies the boundary relationships between adjacent partitions, determining which areas are spatially adjacent. By calculating the spatial geometric relationships between adjacent areas, for example using buffer analysis, a buffer is created on the boundary of each partition, and the adjacency relationship is defined by the intersection of these buffers. For example, if the partition spaces of region A and region B are within the intersection of their buffers, they are considered adjacent. The generated adjacency relationship data includes information such as adjacent region pairs, intersection area, and adjacent boundary lengths, providing detailed data on spatial relationships and supporting subsequent land change analysis.

[0050] Step S44: Based on the indexed partition land space adjacency relationship data, perform indexed partition adjacency characteristic land change transfer trend analysis on the indexed partition land space land change time series data to generate indexed partition adjacency characteristic land change transfer trend data, wherein the indexed partition adjacency characteristic land change transfer trend data includes indexed partition adjacency characteristic land environment change transfer trend data and indexed partition adjacency characteristic land use change transfer trend data.

[0051] In this embodiment of the invention, spatial analysis of land change trends, combined with the change characteristics of adjacent areas, predicts the transfer trends of land change. In practice, fine-grained and coarse-grained land change time-series data are first integrated to analyze land change patterns in adjacent areas. For example, if a large-scale conversion of green space to building land occurs in area A, and this area is adjacent to area B, it is predicted that similar land use changes will occur in area B as well. By performing regression analysis or Markov chain models on these change data, the transfer trends of land change can be extracted. Specifically, in terms of data generation, the indexed partition adjacency characteristic land change transfer trend data will include the following: land environment change transfer trend data, showing the trend of changes in regional environmental conditions (such as water bodies and green spaces); land use change transfer trend data, showing the transformation trend between different land use types by considering the impact of adjacent land use changes, such as the conversion of agricultural land to construction land. The generated indexed partition adjacency characteristic land change transfer trend data is used to analyze and predict the future land change trends of each partition. This data provides a scientific basis for further decisions on land planning, environmental protection, and urban development. For example, by analyzing land use change trends, it is possible to predict development plans for certain urban expansion areas in advance, providing a basis for rational planning and resource allocation.

[0052] Furthermore, step S44 includes the following steps: Step S441: Based on the indexed partition land space adjacency relationship data and the indexed partition land space land change time series data, perform land space adjacency land change transfer impact coefficient analysis to generate land space adjacency land change transfer impact coefficient; In this embodiment of the invention, the transfer impact coefficient of adjacent land changes in national spatial context is analyzed based on indexed partition land spatial adjacency relationship data and land change time series data. The mutual influence of land changes in adjacent areas is modeled to quantify the potential impact of land changes in one area on land changes in adjacent areas. First, the adjacency relationships of each indexed partition are determined, and change patterns are analyzed based on the changes in adjacent areas. For example, when area A undergoes a land use type change (e.g., from agricultural land to building land), this change will affect adjacent area B, such as increasing the probability of land change in area B. The impact coefficient is determined by analyzing factors such as the magnitude of change, the type of change, and the time window. For example, if area A has experienced a large magnitude of land change in the past six months, and area B is adjacent to the boundary of area A, then a higher value is assigned to the change transfer coefficient of area B. The impact coefficient of each area on the land changes of adjacent areas is calculated through regression analysis or similarity measures (e.g., Euclidean distance, Manhattan distance, etc.), thus obtaining the specific transfer impact coefficient. For example, in the land use change analysis of a certain city, the land in area A changes from farmland to residential area, and this area borders adjacent area B. Based on historical data, if land changes in region A are significant and region B has the potential for similar changes, then the influence coefficient of A on B is calculated to be 0.8, indicating that changes in A have a strong impact on region B.

[0053] Step S442: Based on the land change and transfer impact coefficient of adjacent land in national land space and the time series data of land change in indexed partitions, establish the land change and transfer network of adjacent characteristics of each indexed partition, and generate the land change and transfer network of adjacent characteristics of indexed partitions; In this embodiment of the invention, a land change transfer network based on the adjacency characteristics of indexed partitions is constructed using the influence coefficient of adjacent land change transfer and land change time series data. The nodes and edges of the network are defined according to the adjacency relationship and influence coefficient of each partition. Each node represents an indexed partition, and each edge represents the influence relationship between two adjacent partitions; the weight of the edge is determined by the influence coefficient. When constructing the network, land change time series data is used to determine changes, and each edge is weighted according to its influence coefficient. For example, if the influence coefficient of region A on region B is 0.8 and region A has experienced significant changes in the past year, then the transfer edge between A and B will be assigned a higher weight. By using the influence coefficient between different partitions as the edge weights, a complete land change transfer network is formed. This network not only reflects the relationship between adjacent regions but also reveals the spatial transmission pattern of land change. Further graph theory analysis can then be performed to identify the regions experiencing the most change. For example, in a case analyzing urban expansion, assume that regions A, B, and C are adjacent, and the influence coefficient of A on B is 0.9, while the influence coefficient of B on C is 0.7. This data is used to build a weighted adjacency network, where the weight of each edge represents the intensity of land change transmission between regions, further helping to identify hotspots of change.

[0054] Step S443: Perform an analysis of the land change and transfer trends of the index partitions based on the land change and transfer network of the adjacency characteristics of the index partitions, and generate land change and transfer trend data of the adjacency characteristics of the index partitions.

[0055] In this embodiment of the invention, by combining the nodes (partitions) and edges (adjacency relationships) in the network with land change time-series data, the analysis identifies which areas have a spreading effect on changes, or which areas will trigger land changes in adjacent areas. The analysis method uses graph algorithms such as PageRank and Betweenness Centrality to identify the most influential areas in the network, and then predicts the land change trends of their adjacent areas. For example, if large-scale urbanization occurs in area A of a city, and A is adjacent to B and C, according to the analysis of the transfer network and influence coefficients, areas B and C will be affected by the changes in A, resulting in land use changes. Based on previous data analysis, if the influence coefficient of A on B is large and the historical changes in area B conform to a specific pattern (such as the conversion of agricultural land to residential areas), then it is predicted that area B will also experience similar changes in the future. Furthermore, through time-series analysis, combined with historical land change data and adjacency characteristics, a mathematical model of transfer trends is established. These models predict which partitions will become the mainstream of land change in the future, and which areas' change trends will spread to adjacent areas. For example, in predicting land use change trends in a city, if the land use change impact coefficient of region A is 0.85 and its change history is relatively significant, then the predicted change trends of regions B and C will be similar to those of A, providing numerical prediction data for these changes (e.g., the probability of change occurring in region B is 70%). This data provides important basis for land use planning and risk assessment. A detailed analysis of the intrinsic connections and transfer trends of land space changes will construct a more scientific and accurate land change prediction system, thereby assisting decision-makers in making more precise adjustments and response strategies in land management and planning.

[0056] Furthermore, step S441 includes the following steps: Based on the adjacency relationship data of the indexed partition land space, a collinearity analysis of the land space adjacency change time series data of the indexed partition land space is performed to generate collinearity data of the land space adjacency change. In this embodiment of the invention, collinearity analysis is performed on land change time series data using indexed partition land spatial adjacency relationship data. The purpose of collinearity analysis is to assess whether the land change of a certain partition exhibits a high correlation with the land change of its adjacent areas, i.e., whether similar change patterns exist. This includes selecting each adjacent partition within a time window, extracting their land change characteristics, and performing statistical analysis. For example, by calculating the Pearson correlation coefficient or Spearman rank correlation coefficient, the correlation between the land change data of each indexed partition and its adjacent areas within the same time period is analyzed. If, within a certain time period, land in area A changes from farmland to construction land, and the adjacent area B also exhibits a similar change pattern, then the collinearity between A and B is high, indicating a highly correlated land change trend. Through multidimensional collinearity analysis, a land change collinearity matrix of adjacent areas is further constructed, thereby generating collinearity data. This data is used to reveal the synchronicity of land change in adjacent areas, reflecting their common behavior within the same time period. For example, in a study on urban expansion, it is assumed that area A and area B are located in the city center and the city periphery, respectively, and both areas have experienced a large-scale conversion of farmland to construction land. If the changes in these two regions are highly correlated within the same year, the collinearity analysis results show that the Pearson correlation coefficient between A and B is 0.92, indicating that their change patterns are extremely similar and that they have strong collinearity.

[0057] Based on the collinearity data of adjacent land changes in national land space, we conduct a dependency characteristic analysis of adjacent land changes in national land space to generate dependency characteristic data of adjacent land changes in national land space. In this embodiment of the invention, the dependency characteristics of land change in adjacent areas are further explored by analyzing collinearity data. The purpose of dependency characteristic analysis is to identify the causal relationship or interdependence of changes in adjacent areas during land change. For example, in the process of land use change in area A, changes in adjacent area B affect the type of change in A, and vice versa. Statistical methods such as regression analysis, covariance analysis, or principal component analysis (PCA) are used to conduct in-depth analysis of collinearity data to explore the dependencies between land changes. Based on the statistical results, it is identified which land changes depend on changes in adjacent areas. For example, if, within a certain time window, land change in area A is significantly dependent on land change in area B, the dependence strength of area A on changes in area B (e.g., 0.8) can be obtained through covariance analysis or regression models, generating dependency characteristic data. Dependency characteristic data reflects the mutual influence and intensity of land change between areas, providing a basis for subsequent influence coefficient analysis. For example, in the land use change of a certain city, the expansion of industrial land in area A is strongly affected by the infrastructure construction (such as the construction of transportation routes) around area B. Regression analysis results show that the dependence between the change in area A and the change in infrastructure construction in area B reaches 0.75. This indicates a high degree of dependence between land changes in area A and area B.

[0058] Based on the data on the characteristics of adjacent land changes in national land space, an analysis of the transfer impact coefficient of adjacent land changes in national land space is conducted to generate the transfer impact coefficient of adjacent land changes in national land space.

[0059] In this embodiment of the invention, based on land change dependency feature data, the transfer impact coefficient of land change between adjacent areas is further calculated. This coefficient is used to quantify the potential impact of land change in one area on land change in adjacent areas. This includes establishing an impact transmission model, considering the causal relationships and degree of dependency revealed in the dependency feature data, and calculating the impact coefficient of land change between each pair of adjacent areas. For example, by using a weighted regression model or decision tree algorithm, with dependency feature data as input, the intensity of the impact of change in one area on changes in adjacent areas can be predicted. The higher the impact coefficient, the more significant the impact of land change in one area on changes in adjacent areas. The value of each impact coefficient represents the intensity of the transfer of change from one area to adjacent areas. For example, suppose in a land change study in a city, land in area A changes from agricultural land to residential land, and land in area B changes from agricultural land to commercial land. Through dependency feature analysis, the impact coefficient of change in area A on change in area B is found to be 0.85, indicating that land use change in area A has a significant impact on area B and is likely to drive land change in area B. Ultimately, the generated transfer impact coefficient data helps to understand the propagation patterns of land change and provides a quantitative basis for subsequent land change prediction and planning. The data on the impact coefficients of adjacent land change generated through the above steps will provide data support for subsequent land change trend analysis and decision-making, helping to identify hotspots of land change and potential change chains, and promoting more precise land space management.

[0060] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S5 is shown below. In this embodiment, step S5 includes the following steps: Step S51: Based on the land environment change and transfer trend data of the adjacency characteristics of the index partition, perform a return period analysis of the land environment change trend of the index partition to generate return period data of the land environment change trend of the index partition. In this embodiment of the invention, based on land environment change trend data, statistical methods, such as extreme value theory or regression analysis, are used to conduct return period analysis. The goal of return period analysis is to predict the same type of land environment change within a specific time frame based on historical change data. For example, by performing time-series statistics and trend prediction on land change data (such as the trend of land conversion from agricultural land to urban construction land), the frequency of the same type of land change can be assessed. By calculating the interval between each land change event, return period methods in statistics (such as Poisson regression models or extreme value theory) are used to estimate the average time interval between change events. For example, if the interval between land change events in a certain area is 5 years, then the return period is 5 years, indicating that the probability of this type of change repeating is once every 5 years. The return period of land environment change in this area is then estimated by analyzing past change patterns. For example, a regression model can be established to predict the probability of this type of change occurring within the next ten years. Ultimately, the generated return period data can provide a time expectation of land change trends and the environmental conditions in which the return period exists, helping to consider the potential risks of land environment change during planning.

[0061] Step S52: Perform land use type analysis of the indexed partition based on the indexed partition land space data, and generate land use type data of the indexed partition; In this embodiment of the invention, land use types are analyzed based on indexed partitioned territorial spatial data. This analysis uses spatial data (such as remote sensing imagery or land use layers) to classify the use type of each plot, typically into categories such as cultivated land, forest land, construction land, grassland, and water area. By classifying and processing this spatial data (e.g., using machine learning methods such as K-means clustering or support vector machines), land use type data within each indexed partition is generated. The analysis process also considers the spatiotemporal changes of land, for example, by comparing data from different time periods to determine the trend of land use changes.

[0062] Step S53: Design land use type index evaluation decision based on indexed partitioned land use type data; In this embodiment of the invention, a series of indicators are designed based on land use type data to assess the impact and sustainability of land use types. These indicators include land use intensity, ecological carrying capacity, and land use efficiency. These indicators enable the assessment of the health of land use types and potential environmental risks. For example, if urbanization in a region is too rapid, the land use intensity indicator will be very high, leading to ecosystem imbalance. Specific operations include conducting multiple regression analysis based on land use type data and relevant environmental and economic indicators to design an assessment model suitable for the sustainability of land use in that region. For example, for a densely agricultural area, an indicator such as "farmland cultivation intensity" is designed to assess the extent of land use. If the farmland cultivation intensity in this area is too high, it leads to soil degradation. Therefore, by analyzing land use type indicators, a sustainability assessment of land use in the region is provided, helping decision-makers formulate reasonable land use plans.

[0063] Step S54: Use land use type indicators to evaluate the land use change trend data of the index partition adjacency characteristics and generate index partition land use change indicator evaluation data. In this embodiment of the invention, based on land use type indicators for assessment and decision-making, further analysis is performed on the land use change transfer trend data of adjacent characteristics in the indexed partition. By assessing the land use change trend, future land use changes are predicted, and corresponding assessment data is generated. The assessment process relies on the aforementioned land use type indicators, combined with adjacent characteristic data, and uses regression models or time series analysis methods to assess the impact and potential risks of land changes in adjacent areas. For example, suppose that the land use type in a certain area has changed from agricultural land to residential land, and similar changes have occurred in adjacent areas. By combining the impact indicators of land use change with the change trend data of adjacent areas, the type of land use change that will occur in the area in the future can be determined. For example, if the residentialization process in adjacent areas is relatively rapid, it is expected that agricultural land in that area will be converted to residential land. Finally, this process generates land use change indicator assessment data, providing a quantitative basis for further land use planning.

[0064] Step S55: Perform ecological sensitivity analysis of land use change in the indexed partitions based on the land use change index assessment data, and generate ecological sensitivity data of land use change in the indexed partitions. In this embodiment of the invention, the ecological sensitivity of land use change is analyzed based on indexed partitioned land use change index assessment data. This analysis identifies the impact of land use change on the ecological environment based on the land use change situation in each partition, focusing on areas vulnerable to environmental changes (such as ecologically sensitive areas like wetlands and forests). Specifically, an ecological sensitivity model is established, combined with regional ecological data (such as biodiversity, soil quality, and water conservation capacity), to assess the impact of land change on the ecosystem. For example, if agricultural land in a region is converted to construction land, and that region is a water conservation area or wetland, then the ecological sensitivity of land use change in that region is high. By analyzing the land use change in each region and its impact on the ecological environment, ecological sensitivity data of land change is generated. In a land use change analysis, if wetlands near a city are converted to commercial land, it will affect the local hydrological cycle and biodiversity, leading to increased ecological sensitivity. Through ecological sensitivity analysis, an ecological sensitivity index for the region is derived, thereby guiding the optimization of land use planning.

[0065] Step S56: Based on the return period data of land environment change trends in the indexed partitions and the ecological sensitivity data of land change in the indexed partitions, conduct an ecological suitability analysis of land change in the indexed partitions to generate ecological suitability data of land change in the indexed partitions. In this embodiment of the invention, ecological suitability analysis combines land environment change trend return period data with ecological sensitivity data to assess the suitability of different land types and change trends. For example, if a region has high ecological sensitivity and a short return period for land change, it indicates that the region may face high ecological pressure. Furthermore, the analysis of the extreme impacts of the environment on ecological sensitivity under the return period further assesses whether the region is suitable for continued development or other types of land use. By comprehensively considering factors such as ecological environmental protection needs, the temporality of land use change, and the sustainability of change, ecological suitability data for land change is obtained. This data can help decision-makers identify ecologically sensitive areas that are unsuitable for development or require key protection. For example, a region may have a short return period for agricultural land (once every 3 years), and the environment under the return period has a significant impact on agriculture. Simultaneously, the region exhibits high ecological sensitivity (e.g., adjacent to a wetland reserve). Therefore, based on this data, the ecological suitability analysis results show that the region has low ecological suitability and requires ecological protection measures.

[0066] Step S57: Perform land change monitoring weight analysis on the indexed partition based on the ecological suitability data of land change in the indexed partition, and generate land change monitoring weight data for the indexed partition; In this embodiment of the invention, based on the aforementioned ecological suitability analysis data, a monitoring weight analysis is performed on land change in each index partition. This analysis aims to determine the importance of different regions in land change monitoring; the weight reflects the region's criticality in ecological protection, resource utilization, and other aspects. For example, regions with lower ecological suitability (such as sensitive ecological areas or important water source protection areas) are assigned higher monitoring weights to strengthen the monitoring and management of land changes in these areas. The determination of monitoring weights considers not only ecological suitability data but also factors such as the intensity of human activities and the rate of land use change within the region. Through comprehensive analysis, land change monitoring weight data for the index partitions is generated. For example, if the ecological suitability data of a certain region indicates that it is a highly sensitive ecological area, while the adjacent area is ordinary farmland, the monitoring weight for the sensitive ecological area is set to 0.8, while the weight for the farmland area is 0.2. Such weight allocation ensures that more resources are invested in monitoring changes in ecologically sensitive areas.

[0067] Step S58: Execute intelligent feedback operations for land change monitoring in each indexed partition using the indexed partition land change monitoring weight data.

[0068] In this embodiment of the invention, land use change monitoring weight data is used to perform intelligent feedback operations for land use change monitoring in each indexed partition. This operation, through an automated system, combines weight data from each region to monitor land use changes in real time and prioritizes feedback from key areas based on weight analysis results. For example, areas with higher weights (such as ecologically sensitive areas) are prioritized for real-time monitoring and dynamically adjusted. This intelligent feedback system can promptly capture risk points in the land use change process and provide data support for land management decisions. During the operation, the monitoring system continuously assesses the change trends in each partition. Once potential environmental risks or changes exceeding thresholds are detected, it immediately provides feedback and activates an early warning mechanism. For example, in an area with multiple land use types, urban expansion is underway. Based on the indexed partition land use change monitoring weight data, the system frequently monitors the urban expansion area, especially the surrounding ecologically sensitive areas, and updates changes in real time through an intelligent feedback mechanism, ensuring the accuracy and timeliness of land use change monitoring and avoiding resource waste or environmental damage.

[0069] This invention provides a system based on integrated land monitoring data for performing the method described above, the system comprising: The heterogeneous land integrated monitoring module is used to acquire heterogeneous land monitoring data; perform heterogeneous feature dimension interaction data analysis based on the heterogeneous land monitoring data to generate heterogeneous feature multidimensional interaction data; and perform land integrated three-dimensional model analysis based on heterogeneous feature multidimensional interaction data and heterogeneous land monitoring data to generate a land integrated three-dimensional model. The adaptive zoning module for national land space is used to extract national land attribute features from the integrated 3D model of national land and generate national land attribute feature data; and to perform adaptive recursive zoning of the integrated 3D model of national land space based on the national land attribute feature data to generate adaptive zoning national land space data. The territorial spatial index partitioning module is used to calculate the minimum bounding space of the adaptive partitioned territorial spatial data, generating minimum bounding space data; to initialize the minimum leaf node landmark index based on the minimum bounding space data, generating minimum leaf node landmark index data; to analyze the decision mapping relationship between territorial space and spatial index based on the minimum leaf node landmark index data, generating a territorial spatial index decision tree model; and to transfer the adaptive partitioned territorial spatial data to the territorial spatial index decision tree model for landmark index identification processing of the partitioned territorial space, generating indexed partitioned territorial spatial data. The Index Partition Adjacency Characteristics Land Change and Transfer Trend Analysis Module is used to perform time-series data analysis on land changes in index partition land space data, generating time-series data on land changes in index partition land space; and to perform adjacency characteristics land change and transfer trend analysis on the time-series data on land changes in index partition land space, generating trend data on land changes in index partition adjacency characteristics. The index partition land change monitoring weight analysis module is used to perform ecological suitability analysis of index partition land change based on the adjacency characteristics of land change transfer trend data, and generate ecological suitability data of index partition land change; perform land change monitoring weight analysis of index partition land change based on the ecological suitability data of index partition land change, and generate land change monitoring weight data of index partition land change; and execute intelligent feedback operation of land change monitoring for each index partition through the land change monitoring weight data of index partition land change.

[0070] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0071] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for integrated homeland monitoring data based on, characterized by, The method comprises the following steps: Step S1: acquiring heterogeneous land monitoring data; Based on the heterogeneous feature dimension interaction data analysis of the heterogeneous land monitoring data, the heterogeneous feature multi-dimensional interaction data is generated; according to the heterogeneous feature multi-dimensional interaction data and the heterogeneous land monitoring data, the land integrated three-dimensional model analysis is carried out, and the land integrated three-dimensional model is generated; Step S2: according to the land integrated three-dimensional model, the land attribute feature extraction is carried out, and the land attribute feature data is generated; based on the land attribute feature data, the adaptive land space recursive partition processing is carried out on the land integrated three-dimensional model, and the adaptive partition land space data is generated; Step S3: the land minimum enclosing space data is generated by carrying out land minimum enclosing space calculation on the adaptive partition land space data; the minimum leaf node landmark index data is generated by carrying out minimum leaf node landmark index initialization processing according to the land minimum enclosing space data; the decision mapping relationship analysis of land space and space index is carried out based on the minimum leaf node landmark index data, and the land space index decision tree model is generated; the adaptive partition land space data is transmitted to the land space index decision tree model to carry out landmark index identification processing of the partition land space, and the index partition land space data is generated; Step S4: the index partition land space land change time series data is generated by carrying out index partition land space land change time series data analysis on the index partition land space data; The index partition adjacent characteristic land change transfer trend data is generated by carrying out index partition adjacent characteristic land change transfer trend analysis on the index partition land space land change time series data; Step S5: the index partition land change ecological suitability data is generated by carrying out index partition land change ecological suitability analysis based on the index partition adjacent characteristic land change transfer trend data; the index partition land change monitoring weight data is generated according to the index partition land change ecological suitability data; the intelligent feedback work of land change monitoring of each index partition is carried out through the index partition land change monitoring weight data.

2. The method for monitoring integrated homeland based on data according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: the heterogeneous land monitoring data is obtained by using satellite remote sensing system, unmanned aerial vehicle monitoring system and geographic information system to collect and process the land monitoring target area respectively; Step S12: the time sequence and space sequence analysis is carried out according to the heterogeneous land monitoring data, and the heterogeneous land space-time sequence data is obtained; the heterogeneous data space-time alignment mapping processing of land monitoring is carried out on the heterogeneous land monitoring data through the heterogeneous land space-time sequence data, and the mapping land monitoring data is generated; Step S13: the heterogeneous feature dimension interaction data analysis is carried out based on the heterogeneous land monitoring data, and the heterogeneous feature multi-dimensional interaction data is generated; Step S14: the heterogeneous dimension feature integrated land three-dimensional modeling processing is carried out on the mapping land monitoring data according to the heterogeneous feature multi-dimensional interaction data, and the land integrated three-dimensional model is generated.

3. The method for monitoring integrated homeland based on data according to claim 2, characterized in that, Step S13 comprises the following steps: Step S131: the heterogeneous type feature analysis is carried out on the heterogeneous land monitoring data, and the heterogeneous type feature data is generated; Step S132: modeling the heterogeneous type dimension reference sequence according to the heterogeneous land space-time sequence data mapping the land monitoring data, performing heterogeneous feature dimension tensor modeling processing on the heterogeneous type dimension reference sequence and the heterogeneous type feature data, and generating heterogeneous feature dimension tensor modeling data; Step S133: performing heterogeneous feature dimension interaction data analysis on the heterogeneous feature dimension tensor modeling data, and generating heterogeneous feature multi-dimensional interaction data.

4. The method for monitoring integrated homeland data based on claim 1, characterized in that, Step S2 includes the following steps: Step S21: extracting land attribute feature data from the land integrated three-dimensional model; Step S22: performing preliminary partitioned land space processing on the land integrated three-dimensional model using the preset land division proportion data, and generating partitioned land space data; Step S23: calculating the partitioned land attribute feature information entropy data according to the land attribute feature data; Step S24: when the partitioned land attribute feature information entropy data of the partitioned land space data is greater than the preset land interval information entropy threshold, performing land space recursive partitioning operation on the partitioned land space data again using the preset land division proportion data, and outputting the recursive partitioned land space data; when the partitioned land attribute feature information entropy data corresponding to the recursive partitioned land space data of each region is not greater than the preset land interval information entropy threshold or the spatial range of the recursive partitioned land space data of each region is less than the preset land division minimum spatial threshold, ending the land space recursive partitioning operation, and marking the finally output recursive partitioned land space data as adaptive partitioned land space data.

5. The method for monitoring integrated homeland data based on claim 1, characterized by, Step S3 includes the following steps: Step S31: performing land minimum enclosing space calculation on the adaptive partitioned land space data, and generating land minimum enclosing space data; Step S32: performing minimum leaf node landmark index initialization processing on the land minimum enclosing space data, and generating minimum leaf node landmark index data; Step S33: performing land space upper partition relationship analysis on the adaptive partitioned land space data, and generating land space upper partition relationship data; Step S34: performing upper landmark index leaf node analysis processing on the minimum leaf node landmark index data according to the land space upper partition relationship data, and generating upper landmark index leaf node data; Step S35: performing land space and spatial index decision mapping relationship analysis according to the upper landmark index leaf node data and the corresponding minimum leaf node landmark index data, and generating a land space index decision tree model; Step S36: transmitting the adaptive partitioned land space data to the land space index decision tree model to perform landmark index identification processing on the partitioned land space, and generating index partitioned land space data.

6. The method for monitoring integrated homeland data based on claim 1, characterized by, Step S4 includes the following steps: Step S41: designing land change monitoring time sequence level window data, wherein the land change monitoring time sequence level window data includes fine-grained window data and coarse-grained window data; Step S42: land change time series data analysis of the index partitioned land space is performed on the index partitioned land space data using the land change monitoring time series level window data, to generate land change time series data of the index partitioned land space; Step S43: index partitioned land space adjacency relationship analysis is performed on the index partitioned land space data, to generate index partitioned land space adjacency relationship data; Step S44: index partitioned adjacency characteristic land change transfer trend analysis is performed on the index partitioned land space land change time series data based on the index partitioned land space adjacency relationship data, to generate index partitioned adjacency characteristic land change transfer trend data, wherein the index partitioned adjacency characteristic land change transfer trend data includes index partitioned adjacency characteristic land environment change transfer trend data and index partitioned adjacency characteristic land use change transfer trend data.

7. The method for monitoring integrated homeland data based on claim 6, characterized in that, Step S44 includes the following steps: Step S441: land space adjacency land change transfer impact coefficient analysis is performed based on the index partitioned land space adjacency relationship data and the index partitioned land space land change time series data, to generate land space adjacency land change transfer impact coefficient; Step S442: land change transfer network of each index partitioned adjacency characteristic is established based on the land space adjacency land change transfer impact coefficient and the index partitioned land space land change time series data, to generate index partitioned adjacency characteristic land change transfer network; Step S443: index partitioned adjacency characteristic land change transfer trend analysis is performed according to the index partitioned adjacency characteristic land change transfer network, to generate index partitioned adjacency characteristic land change transfer trend data.

8. The method for monitoring integrated homeland data based on claim 7, characterized by, Step S441 includes the following steps: Land space adjacency land change collinearity analysis is performed on the index partitioned land space land change time series data based on the index partitioned land space adjacency relationship data, to generate land space adjacency land change collinearity data; Land space adjacency land change dependency feature analysis is performed based on the land space adjacency land change collinearity data, to generate land space adjacency land change dependency feature data; Land space adjacency land change transfer impact coefficient analysis is performed based on the land space adjacency land change dependency feature data, to generate land space adjacency land change transfer impact coefficient.

9. The method for monitoring integrated homeland data based on claim 6, characterized by, Step S5 includes the following steps: Step S51: index partitioned land environment change trend return period analysis is performed according to the index partitioned land environment change transfer trend data, to generate index partitioned land environment change trend return period data; Step S52: index partitioned land use type analysis is performed according to the index partitioned land space data, to generate index partitioned land use type data; Step S53: land use type index evaluation decision is designed based on the index partitioned land use type data; Step S54: index partitioned land use change index evaluation data is generated by performing index partitioned land use change index evaluation data on the index partitioned adjacency characteristic land use change transfer trend data using the land use type index evaluation decision, to generate index partitioned land use change index evaluation data; Step S55: Index partition land change ecological sensitivity analysis is performed according to the index partition land use change index evaluation data, and index partition land change ecological sensitivity data is generated; Step S56: Index partition land change ecological suitability analysis is performed based on the index partition land environment change trend return period data and the index partition land change ecological sensitivity data, and index partition land change ecological suitability data is generated; Step S57: Index partition land change monitoring weight analysis is performed according to the index partition land change ecological suitability data, and index partition land change monitoring weight data is generated; Step S58: The index partition land change monitoring intelligent feedback work of each index partition is performed through the index partition land change monitoring weight data.

10. A system for monitoring integrated homeland data, characterized by, The system for performing the method based on integrated land monitoring data according to claim 1 comprises: A heterogeneous land integration monitoring module is configured to acquire heterogeneous land monitoring data, perform heterogeneous feature dimension interaction data analysis based on the heterogeneous land monitoring data, generate heterogeneous feature multi-dimensional interaction data, perform land integration three-dimensional model analysis according to the heterogeneous feature multi-dimensional interaction data and the heterogeneous land monitoring data, and generate land integration three-dimensional model data. A land space adaptive partitioning module is configured to extract land attribute feature data from the land integration three-dimensional model data, perform adaptive land space recursive partitioning processing on the land integration three-dimensional model based on the land attribute feature data, and generate adaptive partitioning land space data. A land space index partitioning module is configured to perform land minimum enclosing space calculation on the adaptive partitioning land space data, generate land minimum enclosing space data, perform minimum leaf node landmark index initialization processing according to the land minimum enclosing space data, generate minimum leaf node landmark index data, perform land space and spatial index decision mapping relationship analysis based on the minimum leaf node landmark index data, generate land space index decision tree model, and perform landmark index identification processing on the partitioning land space data by transmitting the adaptive partitioning land space data to the land space index decision tree model, and generate index partitioning land space data. An index partitioning adjacent characteristic land change transfer trend analysis module is configured to perform index partitioning land space land change time series data analysis on the index partitioning land space data, generate index partitioning land space land change time series data, and perform index partitioning adjacent characteristic land change transfer trend analysis on the index partitioning land space land change time series data, and generate index partitioning adjacent characteristic land change transfer trend data. An index partitioning land change monitoring weight analysis module is configured to perform index partitioning land change ecological suitability analysis based on the index partitioning adjacent characteristic land change transfer trend data, generate index partitioning land change ecological suitability data, perform index partitioning land change monitoring weight analysis according to the index partitioning land change ecological suitability data, generate index partitioning land change monitoring weight data, and perform intelligent feedback work of each index partition land change monitoring through the index partitioning land change monitoring weight data.

Images