Land resource planning management method and platform based on space-time big data

Abstract

The invention discloses a land resource planning management method and platform based on space-time big data, and relates to the technical field of land planning management, and the method comprises the steps: determining a to-be-planned land region; obtaining an adjacent land area connected with the to-be-planned land area; constructing a land fragmentation space-time model; traversing the land fragmentation spatial-temporal model in a land planning type library to carry out fragmentation prediction, and outputting a prediction fragmentation index set; and according to the prediction fragmentation index set, obtaining a matched land planning type corresponding to a first prediction fragmentation index, and planning the to-be-planned land region based on the matched land planning type. The technical problem that in the prior art, the land fragmentation condition and the adaptability of the land planning type and the actual area are difficult to fully consider is solved, and the technical effects of achieving optimal configuration and sustainable utilization of land resources and improving the accuracy and adaptability of land resource planning management are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of land planning management, and particularly to a land resource planning management method and platform based on spatio-temporal big data. Background Art

[0002] In the current field of land resource planning management, with the acceleration of urbanization and the rapid development of the economy, the demand for land use is increasingly diverse and complex, posing extremely high requirements for the scientificity and accuracy of land resource planning. However, traditional land resource planning management methods have many limitations. On the one hand, traditional planning often lacks in-depth analysis and effective countermeasures for land fragmentation. Land fragmentation refers to the fragmented and discontinuous state of land in spatial distribution, which may be caused by various factors such as unreasonable land division, frequent land transfers, and lack of overall planning. This fragmentation phenomenon will lead to low land use efficiency. On the other hand, when determining land planning types, traditional planning lacks a comprehensive and accurate evaluation method, usually only based on limited empirical data or simple qualitative analysis, and it is difficult to fully consider various characteristics of the land, surrounding environmental factors, and future development trends and other aspects.

[0003] There are technical problems in the prior art that it is difficult to fully consider the land fragmentation situation and the adaptability of land planning types to the actual area. Summary of the Invention

[0004] This application provides a land resource planning management method and platform based on spatio-temporal big data, which is used to solve the technical problems in the prior art that it is difficult to fully consider the land fragmentation situation and the adaptability of land planning types to the actual area.

[0005] In view of the above problems, this application provides a land resource planning management method and platform based on spatio-temporal big data.

[0006] In the first aspect of this application, a land resource planning management method based on spatio-temporal big data is provided. The method includes: Determine the area of the land to be planned; obtain the adjacent land areas connected to the area of the land to be planned, and obtain the historical spatial characteristics and historical time characteristics of the adjacent land areas; train according to the historical spatial characteristics, the historical time characteristics, and the identification information indicating the fragmentation degree, and construct a land fragmentation spatio-temporal model, where the identification information indicating the fragmentation degree includes the distribution density, connectivity, and average area of the land; the land fragmentation spatio-temporal model traverses in the land planning type library for fragmentation prediction and outputs a set of predicted fragmentation indicators; according to the set of predicted fragmentation indicators, obtain the matching land planning type corresponding to the first predicted fragmentation indicator, and the area of the land to be planned is planned based on the matching land planning type.

[0007] In the second aspect of the present application, a land resource planning and management platform based on spatio-temporal big data is provided. The platform includes: A land area determination module for determining the land area to be planned; an adjacent land area acquisition module for acquiring adjacent land areas connected to the land area to be planned and obtaining the historical spatial characteristics and historical time characteristics of the adjacent land areas; a land fragmentation spatio-temporal model construction module for training according to the historical spatial characteristics, the historical time characteristics, and identification information indicating the degree of fragmentation to construct a land fragmentation spatio-temporal model, wherein the identification information indicating the degree of fragmentation includes the distribution density, connectivity, and average area of the land; a predicted fragmentation index set output module for traversing the land planning type library by the land fragmentation spatio-temporal model for fragmentation prediction and outputting a predicted fragmentation index set; a matching land planning type acquisition module for obtaining a matching land planning type corresponding to the first predicted fragmentation index according to the predicted fragmentation index set, and planning the land area to be planned based on the matching land planning type.

[0008] One or more technical solutions provided in the present application have at least the following technical effects or advantages: Determine the land area to be planned; acquire adjacent land areas connected to the land area to be planned and obtain the historical spatial characteristics and historical time characteristics of the adjacent land areas; train according to the historical spatial characteristics, the historical time characteristics, and identification information indicating the degree of fragmentation to construct a land fragmentation spatio-temporal model; traverse the land planning type library by the land fragmentation spatio-temporal model for fragmentation prediction and output a predicted fragmentation index set; obtain a matching land planning type corresponding to the first predicted fragmentation index according to the predicted fragmentation index set, and plan the land area to be planned based on the matching land planning type. The technical effect of realizing the optimal allocation and sustainable utilization of land resources and improving the accuracy and adaptability of land resource planning and management is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0010] Figure 1Schematic flowchart of the land resource planning and management method based on spatio-temporal big data provided by the embodiments of the present application; Figure 2 Schematic structural diagram of the land resource planning and management platform based on spatio-temporal big data provided by the embodiments of the present application.

[0011] Explanation of reference numerals: land area determination module 10, adjacent land area acquisition module 20, land fragmentation spatio-temporal model construction module 30, predicted fragmentation index set output module 40, matching land planning type acquisition module 50. Detailed implementation manners

[0012] The present application provides a land resource planning and management method and platform based on spatio-temporal big data, which are used to solve the technical problems in the prior art that it is difficult to fully consider the land fragmentation situation and the adaptability between the land planning type and the actual area.

[0013] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0014] Embodiment 1, as Figure 1 shown, the present application provides a land resource planning and management method based on spatio-temporal big data, and the method includes: Step S100: Determine the land area to be planned.

[0015] Specifically, in the initial stage of the land resource planning and management process based on spatio-temporal big data, a variety of technical means and multi-source data information are comprehensively utilized. First, high-resolution satellite remote sensing images are used. With their wide coverage and accurate geolocation capabilities, the target area is comprehensively scanned to preliminarily outline the contours of the possible land areas to be planned. At the same time, basic information such as the current land use situation, topography, and vegetation coverage is obtained. Then, combined with the powerful spatial analysis function of the Geographic Information System (GIS), the satellite image data is deeply processed to further accurately define the regional boundaries and distinguish the spatial distribution relationships of different land types and related geographical elements. On this basis, referring to local urban development plans, overall land use plans, and relevant policy documents and other materials, the planning objectives and requirements are clarified, so as to determine the scope of the land areas to be planned that meet the planning direction and have development potential at the macro level. In addition, to ensure the accuracy and rationality of the regional definition, on-site surveys are also required. For areas with doubts or difficult to accurately judge in satellite images and GIS analysis, such as mountainous areas with complex terrain and the surrounding areas of waters with blurred boundaries, on-site measurements, investigations, and verifications are carried out, and on-site data is collected and compared with the previously obtained information for verification. Finally, the land areas to be planned are accurately determined, laying a solid foundation for the smooth development of subsequent land resource planning and management work.

[0016] Step S200: Obtain the adjacent land areas connected to the land areas to be planned, and obtain the historical spatial characteristics and historical time characteristics of the adjacent land areas.

[0017] Specifically, by leveraging the powerful spatial analysis function of Geographic Information System (GIS), taking the identified land area to be planned as a benchmark, accurately identify and extract the land area range that is directly adjacent to it in the geographical space, which ensures the accuracy and relevance of the selection of adjacent areas. In terms of obtaining the historical spatial characteristics of adjacent land areas, with the help of multi-period high-resolution satellite remote sensing image data, through image interpretation technology, analyze the distribution pattern of land use types in adjacent areas in different historical periods, such as the changes in the distribution of construction land, agricultural land, ecological land, etc., as well as the evolution of topographic and geomorphic features, including the changes in the shape and position of natural geographical elements such as mountains, rivers, and lakes. At the same time, use the map data, land survey data, etc. stored in the historical geographical information database to further supplement and improve the historical spatial characteristic information. For the acquisition of historical time characteristics, deeply excavate various historical archival materials, such as land registration archives, urban construction archives, etc., and sort out the change trajectory of the land use of adjacent areas over time from them, such as the transformation process from agriculture-dominated to industrial rise and then to commercial development and the corresponding time nodes. In addition, refer to local chronicles, economic development statistical materials and other documents to analyze the mutual relationship between the social and economic development trend in the region and the land use change, such as how population growth and industrial structure adjustment drive the change of land use mode, so as to comprehensively master the historical time characteristics of adjacent land areas. These rich historical spatial characteristics and historical time characteristic data provide important basic data support for the subsequent construction of the land fragmentation spatio-temporal model, help to more accurately predict the land fragmentation trend, and thus provide a strong basis for formulating a scientific and reasonable land resource planning scheme for the land area to be planned.

[0018] Step S300: Train according to the historical spatial characteristics, the historical time characteristics, and the identification information indicating the fragmentation degree to construct a land fragmentation spatio-temporal model, where the identification information indicating the fragmentation degree includes the distribution density, connectivity, and average area of the land.

[0019] Specifically, meticulous preprocessing work is carried out on the historical spatial feature and historical time feature data of the adjacent land areas obtained. For the historical spatial feature data, the coordinate system is unified and the data format is standardized to ensure the compatibility and consistency of data from different sources; for the historical time feature data, it is sorted and arranged according to the time series for subsequent analysis of the time trend of land use change. At the same time, information such as land distribution density, connectivity, and average area indicating the degree of fragmentation is accurately quantified and calculated. For example, the ratio of the number of land patches to the total area is calculated using spatial analysis tools in Geographic Information System (GIS) to obtain the distribution density, the connectivity is quantified by constructing topological relationships to analyze the connection status between land patches, and the average area is obtained by statistically calculating the average value of the land patch areas. Then, the processed historical spatial features, historical time features data, and the quantified fragmentation degree identification information are integrated into a training data set. A suitable machine learning algorithm, such as the random forest algorithm, is selected to construct the framework of the land fragmentation spatio-temporal model. During the training process, the training data set is input into the model, and the model learns the data based on the algorithm principle. By continuously adjusting internal parameters (such as the selection of splitting nodes of decision trees, the depth of trees, etc.), the model can accurately capture the complex relationship pattern between historical features and the degree of land fragmentation. After multiple rounds of iterative training, when the prediction performance of the model on the validation data set reaches a stable and satisfactory level, the construction of the land fragmentation spatio-temporal model is completed. This model can comprehensively consider spatio-temporal factors and land fragmentation characteristics, providing reliable technical support for subsequent fragmentation prediction in the land planning type library, thus laying a solid foundation for selecting the optimal land planning type for the land area to be planned.

[0020] Step S400: The land fragmentation spatio-temporal model traverses the land planning type library for fragmentation prediction and outputs a set of predicted fragmentation indicators.

[0021] Specifically, the constructed land fragmentation spatio-temporal model begins to play a key role. First, various land planning type samples are extracted one by one from the land planning type library. These samples cover planning schemes with different functional uses, development intensities, and spatial layout patterns, such as urban commercial area planning, residential area planning, industrial park planning, and ecological protection area planning, etc. For each land planning type sample, the model applies it to the area of land to be planned, simulates the changes in land use under the implementation of this planning type, and thus generates corresponding real-time spatial features and real-time time features. The real-time spatial features include the distribution pattern of land use types after simulated planning, the layout of infrastructure, and the changes in topography and landforms, etc.; the real-time time features reflect the dynamic evolution process of land use changes over time under this planning type. Then, the generated real-time spatial feature and real-time time feature data are input into the land fragmentation spatio-temporal model. Based on the historical data rules and relationship patterns learned during its training process, the model deeply predicts the land fragmentation situation in the area to be planned under each planning type. During the prediction process, the model comprehensively considers the changing trends of factors such as the distribution density, connectivity, and average area of the land, and calculates a series of predicted fragmentation indicators reflecting the degree of land fragmentation, such as the land fragmentation index, landscape connectivity index, patch shape complexity index, etc. These indicators quantify the fragmentation state of the land from different perspectives and comprehensively evaluate the impact of different planning types on the integrity and continuity of the land. Finally, the predicted fragmentation indicators calculated for each land planning type sample are summarized to form a complete set of predicted fragmentation indicators, providing a comprehensive and quantitative data basis for subsequent determination of the optimal matching land planning type.

[0022] Step S500: According to the set of predicted fragmentation indicators, obtain the matching land planning type corresponding to the first predicted fragmentation indicator, and plan the area of land to be planned based on the matching land planning type.

[0023] Specifically, a comprehensive and in-depth analysis is conducted on the set of predicted fragmentation indicators. This set contains indicators related to the possible fragmentation levels of the land area to be planned under different land planning types, and these indicators comprehensively reflect various factors such as the rationality of land use, ecological integrity, and the sustainability of future development. By comparing the numerical values of each predicted fragmentation indicator in the set, the smallest indicator is selected and determined as the first predicted fragmentation indicator. The land planning type corresponding to this indicator is the matching land planning type that, under the current data and model analysis, is most likely to keep the land area to be planned at a relatively low level of land fragmentation, achieve efficient utilization of land resources, and coordinated development of the ecological environment. Once the matching land planning type is determined, detailed planning work is carried out on the land area to be planned based on it. In terms of functional zoning, according to the characteristics of the matching planning type, different functional areas are scientifically and reasonably divided. For example, for the urban comprehensive development planning type, functional areas such as residential, commercial, industrial, public service facilities, and ecological green spaces will be divided to ensure coordinated and complementary development among the functional areas. In terms of infrastructure layout, according to the functional requirements and development orientation of the region, the layout and construction scale of infrastructure such as road networks, water and electricity supply systems, drainage and sewage systems, and communication networks are carefully planned to ensure the normal operation of production and life within the region and the needs of future development. In terms of controlling the intensity of land use, in strict accordance with the requirements of the matching planning type, indicators such as building density, floor area ratio, and green space rate in different regions are determined to avoid waste of land resources and damage to the ecological environment caused by overdevelopment. Through the above series of planning operations based on the matching land planning type, the optimal allocation and sustainable utilization of land resources in the land area to be planned are achieved, and the coordinated development of the regional economy, society, and ecological environment is promoted.

[0024] In a possible implementation manner, step S200 further includes: Step S210: Conduct sampling with replacement according to the historical spatial characteristics and historical time characteristics of the adjacent land areas, and output multiple groups of training data sets.

[0025] Step S220: Initialize a random forest, where the random forest includes multiple decision trees. Each decision tree is trained through each group of training data sets in the multiple groups of training data sets and outputs a fragmentation prediction result.

[0026] Step S230: Optimize the model parameters of the random forest according to the fragmentation prediction result to construct a spatio-temporal model of land fragmentation, where the model parameters include the number of trees, depth, and the number of features of each decision tree.

[0027] Specifically, sampling with replacement is performed based on the historical spatial characteristics and historical time characteristics of adjacent land areas to generate multiple sets of training data sets. For historical spatial characteristics, which may include land use type distribution, topographic and geomorphic information, etc., they are subdivided according to certain rules (such as spatial grid division), and samples are randomly selected in the subdivided units while recording their relevant attribute information. For historical time characteristics, such as time series data of land use changes, similar sampling operations are also carried out. By sampling with replacement, it is ensured that different combinations of characteristics are reflected in multiple sets of training data sets, and the data volume meets the model training requirements. For example, if the historical spatial characteristics include the distribution of land use types in different regions and the historical time characteristics record the time points of land use changes in the past few decades, the sampling process will fully consider various combinations of these factors to generate a rich and diverse set of training samples.

[0028] Perform the initialization operation of the random forest, allocate memory space for it and set the initial parameters, determine the initial number of decision trees in the random forest, the initial depth of the trees, and the initial number of features considered when each node splits, etc. As an ensemble learning model, the random forest contains multiple decision trees inside, and these decision trees will work together in the subsequent training process. Then, for each set of training data sets, they are sequentially input into each decision tree for independent training. During the training process, the decision tree starts from the root node and calculates indicators such as information gain or Gini impurity of each feature according to the historical spatial characteristics (such as the distribution of land use types, topographic and geomorphic features, etc.) and historical time characteristics (such as the time series of land use changes, the historical process of land development, etc.) in the training data set, selects the optimal feature as the splitting node, divides the data set into different subsets, and recursively constructs the decision tree structure. As the training progresses, the decision tree grows continuously, the nodes split continuously until the preset stopping conditions are met, such as reaching the maximum depth, the number of samples in the node is lower than the threshold, or the information gain is less than the set value, etc. After the training is completed, each decision tree makes predictions on the input training data based on the learned feature patterns and decision rules, and outputs fragmented prediction results. For example, the predicted fragmentation degree of the land under specific historical characteristics may be at different levels such as high, medium, and low, or the specific numerical ranges of land fragmentation-related indicators (such as land fragmentation index, patch connectivity index, etc.) are predicted. These fragmented prediction results will provide a basis for the optimization of the subsequent random forest model parameters, thereby gradually constructing an accurate spatio-temporal model of land fragmentation and providing reliable technical support for land resource planning.

[0029] After each decision tree outputs fragmented prediction results based on the training data set, first, these prediction results are compared and analyzed with the actual observations or known land fragmentation situations. By calculating the differences between the predicted values and the true values, such as using evaluation metrics like mean squared error and mean absolute error, the accuracy of the prediction results is quantified. Then, based on these error metrics, the gradient descent method is used to adjust the model parameters of the random forest. For the parameter of the number of trees, if it is found that the prediction error of the current model is large and shows an underfitting state, the number of decision trees can be appropriately increased to enhance the learning ability and generalization ability of the model; conversely, if the model shows signs of overfitting, the number of trees is reduced. When adjusting the depth of the tree, if the depth is too shallow and the model fails to fully learn the complex relationships in the data, resulting in inaccurate prediction results, the depth is increased; if the depth is too deep and causes overfitting, reducing the prediction performance for new data, the depth is appropriately reduced. For the number of features of each decision tree, according to the feature importance evaluation results, if some features contribute less to the prediction results, the number of features considered in the decision tree construction process can be reduced to improve the model training efficiency and avoid overfitting caused by introducing too many irrelevant features; if it is found that important features are not fully utilized, the number of features can be appropriately increased so that the decision tree can better capture the relationship between data features and land fragmentation. Through multiple iterative optimization processes, the model parameters are continuously adjusted to gradually improve the prediction performance of the random forest model on the training data set and the validation data set until the preset optimization goal is reached (such as the error metric is lower than a certain threshold or the performance improvement is not obvious after several consecutive iterations). At this time, the construction of the land fragmentation spatio-temporal model is completed, and this model can accurately predict the degree of land fragmentation based on the input land feature data, providing strong technical support for land resource planning decisions.

[0030] In a possible implementation manner, step S400 further includes: Step S410: Obtain land planning type samples from the land planning type library.

[0031] Step S420: Process the to-be-planned land area according to each land planning type sample, and output real-time spatial features and real-time temporal features.

[0032] Step S430: The land fragmentation spatio-temporal model performs fragmentation prediction according to the real-time spatial features and real-time temporal features, obtains the predicted fragmentation metrics corresponding to each land planning type sample, and outputs a set of predicted fragmentation metrics.

[0033] Specifically, land use planning type samples are carefully selected and obtained from the land use planning type library. This library covers a rich variety of planning schemes, including land use planning types with different functional orientations (such as residential, commercial, industrial, ecological protection, etc.), different development intensities (such as high-density urban construction, low-density rural planning, etc.), and different spatial layout patterns (such as centralized, grouped, ribbon-shaped, etc.). These samples are formed through long-term practical accumulation and theoretical research summary, and have broad representativeness and practicality. For example, it may include high-density, multi-functional mixed planning samples of the central business district in the city, and low-density, ecology-prioritized planning samples of the ecological protection area on the edge of the city, etc., providing rich references for the subsequent simulation analysis of the area to be planned.

[0034] Next, for each land use planning type sample obtained, it is applied to the area to be planned for simulation processing. In this process, various planning elements in the planning type sample are fully considered, such as land use zoning, infrastructure layout, building density control, etc., to simulate the land use changes after the implementation of this planning type, so as to output real-time spatial features and real-time time features. Real-time spatial features include the real-time distribution pattern of land use types (such as the specific distribution forms of residential land, commercial land, green space, etc. in different areas after simulated planning), the real-time spatial layout of infrastructure (such as the extension direction of the road network, the location of water and electricity supply facilities, etc.), and the real-time changes in topography and geomorphology under the influence of the planning (such as topographic changes after land leveling, water system transformation and other projects). Real-time time features reflect the dynamic evolution process of land use changes over time under this planning type, such as the sequence of land development, the time nodes of land use conversion in different stages, and the growth trend of land use intensity over time.

[0035] Finally, the constructed spatio-temporal model of land fragmentation plays a core role. The generated real-time spatial feature and real-time time feature data are input into the model. Based on its internal complex algorithms and the historical data patterns learned during the previous training, the model conducts in-depth fragmentation prediction on the to-be-planned land areas corresponding to each land planning type sample. The prediction process comprehensively considers the changing trends of key factors such as the distribution density, connectivity, and average area of the land after simulated planning, and calculates a series of prediction fragmentation indicators that can accurately reflect the degree of land fragmentation, such as the land fragmentation index (used to measure the fragmentation degree of land patches), the landscape connectivity index (evaluating the connectivity between different land patches), the patch shape complexity index (reflecting the irregularity of the land patch shape), etc. These indicators comprehensively quantify the fragmentation state of the land under different planning types from multiple dimensions. Finally, the prediction fragmentation indicators calculated for each land planning type sample are summarized to form a complete set of prediction fragmentation indicators, providing a comprehensive, accurate, and quantitative data basis for determining the optimal matching land planning type in the follow-up, and strongly supporting the scientificity and rationality of land resource planning decisions.

[0036] In a possible implementation manner, step S400 further includes: Step S440: If the to-be-planned land area is of a single land planning type, generate the land planning type library with the single land planning type as a variable.

[0037] Step S450: The spatio-temporal model of land fragmentation traverses the land planning type library for fragmentation prediction to determine the first prediction fragmentation indicator, where the first prediction fragmentation indicator is the smallest indicator in the set of prediction fragmentation indicators.

[0038] Step S460: Obtain the matching land planning type corresponding to the first prediction fragmentation indicator, where the matching land planning type is the matching single land planning type.

[0039] Specifically, taking this single land planning type as the core variable, deeply analyze its possible forms under different scenarios, conditions, and goals. Extensively collect relevant planning standards, specifications, and detailed information on numerous past land planning projects of the same type, including the innovation points and deficiencies of successful cases, and extract various variable elements from them, such as differences in functional layout (centralized or decentralized, etc.), levels of development intensity (high, medium, and low-intensity development), and diversity of spatial forms (regular or irregular, etc.). Combining the local natural geographical environment (topography, water system, etc.), social and economic conditions (industrial structure, population distribution, etc.), and development strategic planning (short-term and long-term goals), use professional knowledge and experience to combine and adjust these elements to generate a series of single land planning type samples with differences and pertinence. These samples together constitute the land planning type library, providing a comprehensive and diverse reference basis for subsequent evaluation of the degree of regional fragmentation and determination of land planning labels.

[0040] The land fragmentation spatio-temporal model begins a comprehensive traversal operation in the generated land planning type library. For each land planning type sample in the library, the model combines its characteristic information with the relevant data of the land area to be planned, comprehensively considers factors such as the current situation of land use, topography, and surrounding environment, and simulates the changes in the land use pattern after the implementation of this planning type. The model uses its built-in algorithm to calculate the fragmentation prediction indicators in this simulation situation based on key elements such as the distribution density, connectivity, and average area of the land, including indicator values in multiple dimensions such as land fragmentation degree and patch connectivity index. After calculating and analyzing all samples in the type library one by one, select the one with the smallest value from the numerous predicted fragmentation indicators and determine it as the first predicted fragmentation indicator. This indicator represents the situation where the land fragmentation degree is the lowest and most conducive to land resource integration and sustainable use after the implementation of the corresponding planning type under the current model evaluation, providing a key basis for subsequent acquisition of the matching land planning type.

[0041] Obtain the matching land planning type corresponding to the first predicted fragmentation index. Since the entire process focuses on a single land planning type, the matching land planning type obtained here is the single land planning mode that is most suitable for the area of the land to be planned. This matching result will provide a clear direction for subsequent actual land planning and development. During the implementation process, according to the specific requirements of this matching type, detailed functional zoning planning will be carried out, such as determining the distribution areas of different housing types in the residential area and the supporting locations of public service facilities (such as schools, hospitals, supermarkets, etc.); reasonable infrastructure construction planning will be carried out, including the design of the width and direction of roads and the layout planning of the water and electricity supply systems; and a scientific land use intensity control plan will be formulated, such as specifying key indicators such as building density and floor area ratio, to ensure the optimal allocation of land resources under a single planning type, achieve the balanced coordination of land development and environmental protection, social needs and economic development, and improve the comprehensive benefits of land use.

[0042] In a possible implementation manner, step S450 further includes: Step S451: If the area of the land to be planned is of a non-single land planning type, combine according to the number of required planning types to generate a planning combination solution space with the combined land planning type as a variable.

[0043] Step S452: Output the land planning type library according to the planning combination solution space.

[0044] Step S453: The land fragmentation spatio-temporal model traverses in the land planning type library to perform fragmentation prediction and determine the first predicted fragmentation index, where the first predicted fragmentation index is the smallest index in the set of predicted fragmentation indexes.

[0045] Step S454: Obtain the matching land planning type corresponding to the first predicted fragmentation index, where the matching land planning type is the matching combined land planning type.

[0046] Specifically, accurately determine the number of land types to be planned. This determination process is based on a comprehensive consideration of various factors such as the functional orientation, development goals, and resource endowments of the region. For example, if the goal is to create a comprehensive new urban area, it is necessary to plan various land types such as residential, commercial, public services, and ecological leisure. Then, based on these determined land types, use the combination method to carry out comprehensive combination work. Considering the mutual relationships between different land planning types in terms of spatial layout, functional coordination, traffic connection, etc., conduct orderly arrangement and matching. For example, layout residential land adjacent to supporting commercial land to facilitate residents' living and shopping; reasonably distribute public service facilities land (such as schools and hospitals) within the residential area to ensure service accessibility; at the same time, reserve a certain proportion of ecological leisure land to improve the overall environmental quality of the region. Through careful analysis and screening of various possible combination methods, integrate these solutions with combined land planning types as variables to form a complex and rich planning combination solution space. Each element in it represents a unique land use combination mode, covering key information such as the distribution ratio, location relationship, and interaction mode of different land types within the region, providing an abundant selection and analysis basis for subsequent generation of the land planning type library and determination of the optimal planning scheme.

[0047] Outputting the land planning type library based on the constructed planning combination solution space is a crucial step. Deeply analyze each combination solution in the planning combination solution space, extract detailed configuration information about various land planning types, including core elements such as the floor area, spatial distribution location, connection relationship between each other, and expected functional orientation of each land type. Then, standardize this information to meet the storage and call format requirements of the land planning type library, ensuring that each combination solution can be recorded and identified clearly, accurately, and uniformly. Next, based on the information of these processed combination solutions, construct the specific architecture and data structure of the land planning type library, and enter different combination solutions into the type library one by one as independent samples. During the entry process, assign a unique identifier to each sample to facilitate the model to quickly and accurately locate and call during traversal and analysis. At the same time, establish a perfect indexing mechanism to be able to efficiently retrieve and screen out land planning type samples that meet specific requirements according to different query conditions (such as based on land use efficiency, ecological impact degree, social service accessibility, etc.). Finally, through the above series of processing and construction steps, successfully output a complete land planning type library. Each sample in this type library represents a potential planning scheme based on a non-single land planning type combination, and they together constitute a rich and diverse and highly practical planning scheme set, providing a solid data foundation and rich decision-making reference for subsequent use of the land fragmentation spatio-temporal model for fragmentation prediction and determination of matching land planning types.

[0048] The land fragmentation spatio-temporal model begins to perform a traversal operation in the land planning type library generated from the planning combination solution space for fragmentation prediction. The model sequentially selects each combined land planning type sample from the type library. For each sample, the model deeply analyzes the characteristics of its land use layout, functional zoning, infrastructure planning, etc. Combining the geospatial data of the land area to be planned, the historical land use evolution information, and the surrounding environmental constraints, it simulates the dynamic change process of land use after the implementation of this combined planning type. During the simulation process, based on its built-in complex algorithm, the model accurately calculates the change trends of key indicators such as the distribution density, connectivity, and average area of the land at different development stages, and then obtains a series of prediction fragmentation indicators that can reflect the degree of land fragmentation. These indicators cover multiple dimensions such as the fragmentation degree of land patches, the quality of landscape connectivity, and the uniformity of land use. After comprehensively calculating and evaluating all samples in the entire type library, the model carefully compares and sorts the prediction fragmentation indicators corresponding to each sample, and selects the indicator with the smallest value from them, which is determined as the first prediction fragmentation indicator. This indicator represents the fragmentation state corresponding to the combined land planning type solution that can minimize the land fragmentation degree and optimize the land use structure in the land area to be planned under the current model evaluation system, providing a key decision-making basis for obtaining the matching land planning type subsequently.

[0049] After determining the first prediction fragmentation indicator, it is then necessary to obtain the corresponding matching land planning type. Since this process is for the case of non-single land planning types, the obtained matching land planning type is combined. Through the indexes and data associations established in the land planning type library, quickly locate the combined land planning type sample associated with the first prediction fragmentation indicator. The combined land planning type represented by this sample is the solution that the model determines can most effectively reduce the land fragmentation degree and achieve the optimal allocation of land resources after comprehensively considering various factors such as land use efficiency, ecological environment impact, and regional development coordination. It clarifies the specific layout, proportional relationship, and functional connection and spatial connection methods between different land types (such as residential, commercial, industrial, ecological land, etc.) in the land area to be planned. The obtained matching combined land planning type will serve as an important guidance for the subsequent implementation of land planning, providing detailed and scientific basis for actual land development, infrastructure construction, functional zoning determination, etc., to ensure that the land area to be planned can achieve sustainable and efficient development.

[0050] In a possible implementation manner, step S200 further includes: Step S240: Obtain the quantitative indicators of the land area to be planned.

[0051] Step S250: Obtain the first conversion coefficient, and perform conversion based on the quantization index of the to-be-planned land area according to the first conversion coefficient, and output the constrained quantization index.

[0052] Step S260: Constrain the quantization index of the adjacent land area according to the constrained quantization index.

[0053] Specifically, obtain the quantization index of the to-be-planned land area through various data collection means. These indexes cover multiple aspects such as the area, shape, terrain undulation degree, and proportion of land use types of the land. For example, use Geographic Information System (GIS) technology to accurately measure the land area, use Digital Elevation Model (DEM) data to calculate the terrain undulation degree, and determine the proportion of different land use types (such as cultivated land, construction land, forest land, etc.) in the to-be-planned area through the land use status survey data. These quantization indexes comprehensively reflect the basic characteristics and current situation of the to-be-planned land area, providing basic data support for subsequent analysis and planning.

[0054] Obtain the first conversion coefficient, which is determined comprehensively based on relevant planning standards, historical experience data, and regional development strategies, etc. To ensure a reasonable proportional relationship between the adjacent land area and the to-be-planned land area in terms of quantization indexes such as area, so as to ensure the accuracy of the fragmentation degree output. For example, if according to the requirements of the regional development plan, the adjacent area has a strong supporting or collaborative effect on the current to-be-planned area in terms of function, set the first conversion coefficient so that the area of the adjacent area must be a certain multiple (such as 20 times) of the to-be-planned area. Then, convert the quantization index of the to-be-planned land area according to the conversion coefficient, and calculate the constrained quantization index. This process not only considers the characteristics of the to-be-planned area itself, but also fully combines the overall planning requirements of the surrounding areas, making the quantization index more scientific and reasonable.

[0055] Constrain the quantization index of the adjacent land area according to the calculated constrained quantization index. In actual operation, if there is a difference between the actual quantization index of the adjacent land area and the constrained quantization index, it is necessary to adjust the boundary of the adjacent area, optimize the land use type, or make partial corrections to the planning scheme. For example, if the area of the adjacent area is smaller than the area required by the constrained quantization index, consider incorporating some suitable land around it into the adjacent area; if the proportion of land use types does not meet the requirements, re-plan and layout it to ensure that the quantization index of the adjacent land area meets the constraint conditions. In this way, the coordination and unity of the quantization indexes of the to-be-planned land area and its adjacent areas are achieved, providing a reliable data basis and a reasonable regional relationship definition for subsequent construction of the land fragmentation spatio-temporal model and accurate land planning, which helps to improve the scientificity and effectiveness of land resource planning.

[0056] In a possible implementation manner, step S500 further includes: Step S510: Obtain the geographic information data of the to-be-planned land area.

[0057] Step S520: According to the matched land planning type, conduct a land suitability analysis on the geographic information data, and output a land suitability analysis result.

[0058] Step S530: If the land suitability analysis result is less than a preset suitability, re-plan in the land planning type library, and output an updated matched land planning type.

[0059] Specifically, the geographic information data of the to-be-planned land area is obtained through a variety of advanced geographic information technology means. These data include topographic and geomorphic data (such as altitude, slope, aspect, etc.), which affect the developability of the land and the construction difficulty. For example, large-scale construction in areas with a large slope may face the risks of geological disasters and increased engineering costs; soil type data, different soil types are suitable for growing different crops or bearing different types of building foundations. For example, clay is suitable for growing rice but may not be conducive to the construction of high-rise building foundations; water system distribution data, the location and flow of rivers, lakes and other water systems will restrict the land use mode. Areas near water sources may be more suitable for developing ecological tourism or agricultural irrigation; and vegetation coverage data, areas with dense vegetation may need to consider ecological protection factors in the planning, etc. These geographic information data comprehensively and detailedly reflect the natural geographic conditions of the to-be-planned area, providing a basic basis for subsequent analysis.

[0060] Based on the obtained matching land planning type, a comprehensive and in-depth land suitability analysis is carried out on the geographical information data. First, for the matching land planning type, clarify its key requirements and characteristics. For example, if it is an industrial land planning type, focus on whether the flatness of the terrain is conducive to factory building, the convenience of transportation to ensure the transportation of raw materials and products, and whether the geological conditions can bear heavy industrial equipment, etc.; if it is an ecological protection area planning type, then focus on factors such as vegetation coverage, water conservation ability, and biodiversity. Then, compare and analyze these requirements with the geographical information data of the area to be planned one by one. Using the Geographic Information System (GIS), extract relevant elements in the geographical information data, such as slope and aspect data in topography, river direction and water volume information in water system distribution, and fertility and texture data in soil type, etc. By establishing a scientific and reasonable evaluation index system, quantitatively evaluate these elements to determine whether they meet the requirements of the matching land planning type. Finally, comprehensively consider the evaluation results of each item and output the land suitability analysis result. This result clearly shows the suitability degree of the area to be planned for the matching land planning type, such as highly suitable, moderately suitable, lowly suitable or unsuitable, etc., providing a key basis for subsequent decision-making, judging whether the currently matching land planning type is really suitable for the actual geographical conditions of this area, and thus deciding whether it is necessary to further adjust the planning scheme When the land suitability analysis result is less than the preset suitability, it means that the current matching land planning type has a poor adaptability to the actual geographical conditions of the area to be planned. At this time, it is necessary to restart the planning work in the land planning type library. First, comprehensively sort out the problems and mismatching factors found in the land suitability analysis process, such as the topography not meeting the planning requirements, the soil conditions being unfavorable for specific land use methods, etc. Then, based on this feedback information, screen out other more suitable land planning types or adjust and combine the existing planning types in the type library. Use the land fragmentation spatio-temporal model to predict the fragmentation of the new planning type or combination again, and obtain a new set of predicted fragmentation indicators. Through comparative analysis, determine the new first predicted fragmentation indicator and its corresponding matching land planning type. After multiple iterations and optimizations, until an updated matching land planning type that meets the preset suitability requirements is found, ensuring that the final planning scheme can fully fit the geographical characteristics of the area to be planned and realize the efficient and reasonable use of land resources.

[0061] Embodiment 2, based on the same inventive concept as the land resource planning and management method based on spatio-temporal big data in the foregoing embodiment, as Figure 2 shown, this application provides a land resource planning and management platform based on spatio-temporal big data. The platform in the embodiment of this application and the method embodiment are based on the same inventive concept. Among them, the platform includes: Land area determination module 10, where the land area determination module 10 is used to determine the land area to be planned.

[0062] Adjacent land area acquisition module 20, where the adjacent land area acquisition module 20 is used to acquire adjacent land areas connected to the land area to be planned, and acquire the historical spatial features and historical temporal features of the adjacent land areas.

[0063] Land fragmentation spatio-temporal model construction module 30, where the land fragmentation spatio-temporal model construction module 30 is used to train according to the historical spatial features, the historical temporal features, and identification information indicating the fragmentation degree, and construct a land fragmentation spatio-temporal model. Among them, the identification information indicating the fragmentation degree includes the distribution density, connectivity, and average area of the land.

[0064] Predicted fragmentation index set output module 40, where the predicted fragmentation index set output module 40 is used to traverse the land fragmentation spatio-temporal model in the land planning type library for fragmentation prediction, and output a set of predicted fragmentation indexes.

[0065] Matched land planning type acquisition module 50, where the matched land planning type acquisition module 50 is used to acquire the matched land planning type corresponding to the first predicted fragmentation index according to the set of predicted fragmentation indexes, and the land area to be planned is planned based on the matched land planning type.

[0066] Furthermore, the adjacent land area acquisition module 20 further includes: Training data set output unit, where the training data set output unit is used to perform sampling with replacement according to the historical spatial features and historical temporal features of the adjacent land areas, and output multiple sets of training data sets.

[0067] Fragmentation prediction result output unit, where the fragmentation prediction result output unit is used to initialize a random forest, and the random forest includes multiple decision trees. Among them, each decision tree is trained by each set of training data sets in the multiple sets of training data sets, and outputs a fragmentation prediction result.

[0068] Land fragmentation spatio-temporal model construction unit, where the land fragmentation spatio-temporal model construction unit is used to optimize the model parameters of the random forest according to the fragmentation prediction results, and construct a land fragmentation spatio-temporal model. Among them, the model parameters include the number of trees, the depth, and the number of features of each decision tree.

[0069] Furthermore, the predicted fragmentation index set output module 40 further includes: Land planning type sample acquisition unit, where the land planning type sample acquisition unit is used to acquire land planning type samples in the land planning type library.

[0070] A feature output unit, which is configured to process the area of the land to be planned according to each land planning type sample and output real-time spatial features and real-time temporal features.

[0071] A predicted fragmentation index set output unit, which is configured to perform fragmentation prediction by the land fragmentation spatio-temporal model according to the real-time spatial features and real-time temporal features, obtain the predicted fragmentation index corresponding to each land planning type sample, and output a predicted fragmentation index set.

[0072] Furthermore, the predicted fragmentation index set output module 40 further includes: A land planning type library generation unit, which is configured to generate the land planning type library with a single land planning type as a variable if the area of the land to be planned is of a single land planning type.

[0073] A first predicted fragmentation index determination unit, which is configured to perform fragmentation prediction by traversing the land planning type library by the land fragmentation spatio-temporal model to determine a first predicted fragmentation index, where the first predicted fragmentation index is the smallest index in the predicted fragmentation index set.

[0074] A matching land planning type acquisition unit, which is configured to acquire the matching land planning type corresponding to the first predicted fragmentation index, where the matching land planning type is a matching single land planning type.

[0075] Furthermore, the first predicted fragmentation index determination unit further includes: A planning combination solution space generation unit, which is configured to perform combination according to the number of types of land to be planned if the area of the land to be planned is not of a single land planning type, and generate a planning combination solution space with the combined land planning types as variables.

[0076] A land planning type library output unit, which is configured to output the land planning type library according to the planning combination solution space.

[0077] A fragmentation prediction unit, which is configured to perform fragmentation prediction by traversing the land planning type library by the land fragmentation spatio-temporal model to determine a first predicted fragmentation index, where the first predicted fragmentation index is the smallest index in the predicted fragmentation index set.

[0078] Combined land planning type acquisition unit, which is used to acquire the matching land planning type corresponding to the first predicted fragmentation index, where the matching land planning type is the matching combined land planning type.

[0079] Further, the adjacent land area acquisition module 20 further includes: Quantification index acquisition unit, which is used to acquire the quantification index of the land area to be planned.

[0080] Constrained quantification index output unit, which is used to acquire the first conversion coefficient, convert the quantification index of the land area to be planned according to the first conversion coefficient, and output the constrained quantification index.

[0081] Quantification index constraint unit, which is used to constrain the quantification index of the adjacent land area according to the constrained quantification index.

[0082] Further, the matching land planning type acquisition module 50 further includes: Geographic information data acquisition unit, which is used to acquire the geographic information data of the land area to be planned.

[0083] Land suitability analysis result output unit, which is used to perform land suitability analysis on the geographic information data according to the matching land planning type, and output the land suitability analysis result.

[0084] Land planning type library planning unit, which is used to re-plan in the land planning type library if the land suitability analysis result is less than the preset suitability, and output the updated matching land planning type.

[0085] It should be noted that the above-mentioned sequence of embodiments of the present application is only for description and does not represent the advantages or disadvantages of the embodiments. And the above describes specific embodiments of this specification. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0086] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0087] This specification and the drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications therein.

Claims

1. A land resource planning and management method based on spatiotemporal big data, characterized in that: The method comprises: Identify the land area to be planned; Acquire an adjacent land area connected to the land area to be planned, and acquire historical spatial characteristics and historical temporal characteristics of the adjacent land area; Training is performed according to the historical spatial characteristics, the historical time characteristics, and identification information indicating the degree of fragmentation to construct a spatiotemporal model of land fragmentation, wherein the identification information indicating the degree of fragmentation includes distribution density, connectivity, and average area of ​​the land; The land fragmentation spatiotemporal model traverses the land planning type library to perform fragmentation prediction, and outputs a set of predicted fragmentation indicators; According to the predicted fragmentation indicator set, a matching land planning type corresponding to a first predicted fragmentation indicator is obtained, and the land area to be planned is planned based on the matching land planning type.

2. The method according to claim 1, characterized in that The method further comprises: Performing sampling with replacement according to the historical spatial characteristics and historical temporal characteristics of the adjacent land areas, and outputting multiple sets of training data sets; Initializing a random forest, wherein the random forest includes a plurality of decision trees, wherein each decision tree is trained by each of the plurality of training data sets, and outputs a fragmented prediction result; The model parameters of the random forest are optimized according to the fragmentation prediction results to construct a spatiotemporal model of land fragmentation, wherein the model parameters include the number and depth of trees and the number of features of each decision tree.

3. The method according to claim 1, characterized in that The land fragmentation spatiotemporal model is traversed in the land planning type library to perform fragmentation prediction, and the method includes: Obtaining a land planning type sample from the land planning type library; Processing the land area to be planned according to each land planning type sample, and outputting real-time spatial features and real-time temporal features; The land fragmentation spatiotemporal model performs fragmentation prediction according to the real-time spatial characteristics and real-time temporal characteristics, obtains the predicted fragmentation index corresponding to each land planning type sample, and outputs a set of predicted fragmentation indexes.

4. The method according to claim 1, characterized in that The land fragmentation spatiotemporal model is traversed in the land planning type library to perform fragmentation prediction, and the method includes: If the land area to be planned is of a single land planning type, the land planning type library is generated with the single land planning type as a variable; The land fragmentation spatiotemporal model traverses the land planning type library to perform fragmentation prediction, and determines a first predicted fragmentation index, wherein the first predicted fragmentation index is the smallest index in the predicted fragmentation index set; Obtain a matching land planning type corresponding to the first predicted fragmentation index, wherein the matching land planning type is a matching single land planning type.

5. The method according to claim 4, characterized in that The land fragmentation spatiotemporal model is traversed in the land planning type library to perform fragmentation prediction, and the method further includes: If the land area to be planned is not a single land planning type, the number of planning types required is combined, and the planning combination solution space is generated with the combined land planning type as a variable; Outputting the land planning type library according to the planning combination solution space; The land fragmentation spatiotemporal model traverses the land planning type library to perform fragmentation prediction, and determines a first predicted fragmentation index, wherein the first predicted fragmentation index is the smallest index in the predicted fragmentation index set; Obtain a matching land planning type corresponding to the first predicted fragmentation index, wherein the matching land planning type is a matching combined land planning type.

6. The method according to claim 1, characterized in that Acquiring an adjacent land area connected to the land area to be planned, the method further includes: Obtaining quantitative indicators of the land area to be planned; Acquire a first conversion coefficient, convert the quantitative index of the land area to be planned according to the first conversion coefficient, and output the constraint quantitative index; The quantitative index of the adjacent land area is constrained according to the constrained quantitative index.

7. The method according to claim 1, characterized in that After obtaining the matching land planning type corresponding to the first predicted fragmentation index, the method further includes: Acquiring geographic information data of the land area to be planned; According to the matching land planning type, performing land suitability analysis on the geographic information data, and outputting the land suitability analysis result; If the land suitability analysis result is less than the preset suitability, re-planning is performed in the land planning type library, and an updated matching land planning type is output.

8. The land resource planning and management platform based on spatiotemporal big data is characterized by: The platform is used to execute the land resource planning and management method based on spatiotemporal big data according to any one of claims 1 to 7, and the platform includes: A land area determination module, the land area determination module is used to determine the land area to be planned; An adjacent land area acquisition module, the adjacent land area acquisition module is used to acquire an adjacent land area connected to the land area to be planned, and acquire historical spatial characteristics and historical temporal characteristics of the adjacent land area; a land fragmentation spatiotemporal model construction module, the land fragmentation spatiotemporal model construction module is used to train according to the historical spatial characteristics and the historical time characteristics and identification information identifying the degree of fragmentation to construct a land fragmentation spatiotemporal model, wherein the identification information identifying the degree of fragmentation includes the distribution density, connectivity and average area of ​​the land; A predicted fragmentation indicator set output module, which is used to traverse the land fragmentation spatiotemporal model in the land planning type library to perform fragmentation prediction and output a predicted fragmentation indicator set; A matching land planning type acquisition module is used to obtain a matching land planning type corresponding to a first predicted fragmentation indicator according to the predicted fragmentation indicator set, and the land area to be planned is planned based on the matching land planning type.

Images