A method for predicting rural ecological space based on bayesian network

Abstract

This invention discloses a method for predicting rural ecological space based on Bayesian networks. The method includes: establishing a rural ecological space evaluation index system based on selected suitable factors; determining the time points involved in the Bayesian network model for rural ecological space prediction; acquiring basic data and creating a thematic map of rural ecological space evaluation indicators using preprocessed basic data; constructing a Bayesian network model for rural ecological space prediction by combining the causal relationships of the rural ecological space evaluation indicators; using the Bayesian network model for parameter learning to obtain a conditional probability table; and using the inference engine of the Bayesian network model to predict rural ecological space. This invention, based on the mechanism of action of factors influencing the evolution of rural ecological space and the game relationship between different functional spaces, employs Bayesian networks to solve the problems of the randomness of human activities and the uncertainty of dynamic changes over time in the evolution of rural ecological space, thus achieving the prediction of rural ecological space and improving the accuracy of the prediction results.

Description

Technical Field

[0001] This invention belongs to the field of rural ecological landscape planning technology, specifically involving a rural ecological space prediction method based on Bayesian networks. Background Technology

[0002] Predicting the expansion and contraction of rural ecological space can effectively reflect the evolutionary process of ecological space, identify potential areas in rural areas with ecological loss risk and restoration value, and contribute to the protection of rural ecological space. Currently, research on ecological space largely focuses on the identification of important ecological functional zones and ecologically sensitive and vulnerable areas. The delineation of ecological protection red lines targets core ecological protection zones within ecological space, rather than the overall ecological space. The erosion of rural ecological space outside ecological protection red lines has also not received sufficient attention. Currently, the "source-sink" theory is often used to construct minimum cumulative resistance models to predict rural ecological space patterns. However, this method is an idealized simulation of ecological processes and has limitations. On the one hand, this method fails to consider the uncertainty of the dynamic changes of natural and anthropogenic influencing factors over time; the resistance surface is constructed based on the current ecological condition, and the resulting predictions cannot reflect the evolutionary characteristics of ecological space over time. On the other hand, rural ecological space is formed under the interplay of different functional spaces; considering only the causes of ecological space change leads to one-sided research results. While traditional land use pattern evolution simulation methods such as CA-Markov can reflect the changing processes of various functional spaces, they essentially rely on analyzing large amounts of sample data to obtain statistical relationships and then making predictions. Therefore, they fail to reflect the underlying ecological processes and the mechanisms of action of various influencing factors, resulting in insufficient reliability and accuracy. Thus, a scientific prediction method is needed to accurately predict future rural ecological spaces, taking into account the temporal process and the interplay between various functional spaces. Summary of the Invention

[0003] To address the shortcomings mentioned in the background section, the present invention aims to provide a method for predicting rural ecological space based on Bayesian networks.

[0004] The objective of this invention can be achieved through the following technical solution: a method for predicting rural ecological space based on Bayesian networks, the method comprising the following steps:

[0005] By combining knowledge of rural ecological characteristics and land use evolution patterns, suitable factors are selected, and a rural ecological space evaluation index system is established using these suitable factors for Bayesian network model construction.

[0006] Determine the time points involved in the Bayesian network model for predicting rural ecological space, including time point 1, time point 2, and time point 3;

[0007] Acquire basic data and preprocess it. Then, use the preprocessed basic data to create a thematic map of rural ecological space evaluation indicators.

[0008] By combining the causal relationships among the indicators in the thematic map of rural ecological space evaluation indicators, a Bayesian network model for predicting rural ecological space is constructed.

[0009] By using a Bayesian network model for predicting rural ecological space, parameter learning is performed to obtain a conditional probability table of rural ecological space evaluation indicators, which serves as the basis for inference in the Bayesian network model for predicting rural ecological space.

[0010] Preferably, the rural ecological space evaluation index system includes spatial factors, ecological suitability factors, land use change factors, policy factors, and target factors.

[0011] Preferably, the spatial factors include elevation, slope, distance from water area, distance from road, distance from building, and distance from forest; the ecological suitability factors include ecological sensitivity and importance of ecosystem service functions; the land use change factors include changes in ecological land use, changes in construction land use, and changes in agricultural land use; the policy factors include ecological protection red lines; and the target factor is the predicted rural ecological space.

[0012] Preferably, the ecological sensitivity includes topographic sensitivity, hydrological sensitivity, vegetation sensitivity and land use sensitivity, and the importance of ecosystem service functions includes water conservation capacity and soil and water conservation capacity.

[0013] Preferably, the time points involved in the Bayesian network model for predicting rural ecological space include time point 1, time point 2, and time point 3. The time span between time point 1 and time point 2, and between time point 2 and time point 3, must be consistent. Time point 3 is the year corresponding to the predicted rural ecological space.

[0014] Preferably, the basic data includes satellite remote sensing data at time point 1 and time point 2, digital line map data, and ecological protection red line data;

[0015] The preprocessing of basic data includes projection conversion, image fusion, geometric correction, image enhancement and stitching of satellite remote sensing data, and data format conversion, resampling, reclassification and cropping of digital line map data and ecological protection red line data.

[0016] The creation of thematic maps for rural ecological space evaluation indicators includes:

[0017] Rural land use type raster map: Based on the LUCC land use classification system, the pre-processed satellite remote sensing data at time point 1 and time point 2 are combined with human-computer interactive interpretation to obtain the rural land use map. The land use types are divided into 6 categories, including: construction land, forest land, water area, shrubland and grassland, agricultural land and bare land. After rasterization, the rural land use type raster map at time point 1 and time point 2 is obtained.

[0018] Rural ecological space raster map: By combining rural land use type raster maps at time point 1 and time point 2 with satellite remote sensing data, forest land, water area, shrubland and grassland are merged to obtain rural ecological space raster maps at time point 1 and time point 2.

[0019] Spatial factor raster maps: Elevation and slope raster maps for time point 1 and time point 2 were obtained using GIS (Geographic Information System) spatial analysis methods. Multi-ring buffer analysis was used to obtain raster maps of distances to water bodies, roads, buildings, and forests.

[0020] Ecological suitability factor raster: The ecological sensitivity and importance of ecosystem service functions at time point 1 and time point 2 were obtained by using the analytic hierarchy process and weighted overlay method.

[0021] Policy Factor Raster Map: Identify the areas inside and outside the ecological protection red line at time points 1 and 2, and obtain the ecological protection red line area raster map;

[0022] Land use change factor raster map: By overlaying and analyzing the land use maps at time point 1 and time point 2, raster maps of ecological land use change, construction land use change, and agricultural land use change are obtained.

[0023] Preferably, the formula for the Bayesian network model for predicting rural ecological space is as follows:

[0024] S = (V, L) (1)

[0026] In the formula, S represents the Bayesian structure, and S is composed of the set of node variables V (V = {V1, V2, V3, ..., V...}). n}) and directed edge L (L=V i V j |V i V j Composed of ( ∈ V), where the node variable V i It is an abstract representation of the predicted ecological space, where directed edges L are node variables V. i V j The dependency or causal relationship between them;

[0027] The parameters between node variables are a probability distribution set reflecting the local correlation between the nodes, and their expression is as follows:

[0028] P = {P(V)} i |V1, V2, V3, ..., V i -1), V i ∈V (2)

[0030] Use V pi Represents variable V i If the set of parent nodes of V is given, then the joint probability distribution of V is:

[0031]

[0032] Preferably, the process of using a Bayesian network model for rural ecological space prediction to learn parameters and obtain a conditional probability table includes the following steps:

[0033] A Bayesian network model for predicting rural ecological space was constructed using rural ecological space evaluation indicators as node variables.

[0034] The raster data in the spatial factor raster map, ecological suitability factor raster map, land use change factor raster map, policy factor raster map at time point 1, and the ecological spatial raster map at time point 2 are discretized. Among them, the spatial factor and ecological suitability factor raster data are discretized into 3 levels, and the land use change factor, policy factor raster data and the ecological spatial raster data at time point 2 are discretized into 2 levels.

[0035] All discretized raster data are converted into point data, exported as a set of sample point data containing the values ​​and coordinates of each feature, and converted into ASCII format as training sample data.

[0036] The Bayesian network model for predicting rural ecological space uses training sample data to learn parameters and obtains a complete conditional probability table after training.

[0037] Preferably, the process of predicting rural ecological space using the inference engine of the Bayesian network model for rural ecological space prediction includes the following steps:

[0038] In GIS, the raster data of spatial factor raster, ecological suitability factor raster, land use change factor raster, and policy factor raster at time point 2 are discretized. The spatial factor and ecological suitability factor raster data are discretized into 3 levels, and the land use change factor and policy factor raster data are discretized into 2 levels.

[0039] All discretized raster data are converted into point data, exported as a sample point data set containing the values ​​and coordinates of each feature, and converted into ASCII format as evidence sample data.

[0040] The evidence sample data was imported into the Bayesian network model for predicting rural ecological space, and the data was updated in the Bayesian network model for predicting rural ecological space.

[0041] The probability distribution of each sample point in the rural ecological space at time point 3 was calculated using the joint tree inference engine.

[0042] The most probable interpretation is used to determine whether a sample point is located within the rural ecological space, and the result is output as a set of point data.

[0043] The point data set was transformed into a data raster map for analysis and visualization, resulting in the rural ecological space at time point 3.

[0044] Preferably, an apparatus includes:

[0045] One or more processors;

[0046] Memory, used to store one or more programs;

[0047] When the one or more programs are executed by the one or more processors, the one or more processors implement a Bayesian network-based method for predicting rural ecological space as described above.

[0048] Preferably, a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a Bayesian network-based rural ecological spatial prediction method as described above.

[0049] The beneficial effects of this invention are:

[0050] The proposed method for predicting rural ecological space addresses the randomness of human activities and the uncertainty of dynamic changes over time in rural spatial evolution, thus improving the reliability of prediction results. Furthermore, it effectively reflects the interaction mechanisms among factors influencing rural ecological space evolution and the game-theoretic relationships between different types of functional spaces through a Bayesian network structure, quantifying these relationships using conditional probability tables to enhance accuracy. This invention identifies key factors influencing rural ecological space evolution and their probabilities, providing a reference for the optimal allocation of these key factors and supporting the protection and development of rural ecological space. Based on Bayesian networks, this invention constructs an open prediction tool for rural ecological space, applicable to predictions of rural ecological space at different times and in different regions, demonstrating strong universality. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0053] Figure 2 This is a schematic diagram of a rural ecological space evaluation index system according to an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of a Bayesian network model for predicting rural ecological space according to an embodiment of the present invention.

[0055] Figure 4 This is a schematic diagram of training sample data for a Bayesian network model according to an embodiment of the present invention;

[0056] Figure 5 This is a schematic representation of the conditional probability of a Bayesian network model according to an embodiment of the present invention;

[0057] Figure 6 This is a schematic diagram of evidence sample data for a Bayesian network model according to an embodiment of the present invention;

[0058] Figure 7 This is a schematic diagram of rural ecological space at time point 3 according to an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] See Figures 1-7 This embodiment takes a village in Jiangning District, Nanjing City as an example to provide a method for predicting rural ecological space based on Bayesian networks:

[0061] like Figure 1 As shown, a method for predicting rural ecological space based on Bayesian networks includes the following steps:

[0062] S1: Combining knowledge of rural ecological characteristics and land use evolution patterns, suitable factors are selected, and a rural ecological space evaluation index system is established using these suitable factors for Bayesian network model construction, such as... Figure 2 As shown;

[0063] The spatial factors in the rural ecological space evaluation index system include elevation, slope, distance from water bodies, distance from roads, distance from buildings, and distance from forest land; ecological suitability factors include ecological sensitivity and the importance of ecosystem service functions; among them, ecological sensitivity includes topographic sensitivity, hydrological sensitivity, vegetation sensitivity, and land use sensitivity, and the importance of ecosystem service functions includes water conservation capacity and soil and water conservation capacity; land use change factors include changes in ecological land use, changes in construction land use, and changes in agricultural land use; policy factors include ecological protection red lines; and the target factor is the predicted ecological space.

[0064] S2: Determine the time points involved in the Bayesian network model for predicting rural ecological space, including time point 1, time point 2, and time point 3;

[0065] The time points involved in the Bayesian network model for predicting rural ecological space include time point 1, time point 2, and time point 3. The time span between time point 1 and time point 2, and between time point 2 and time point 3, must be consistent. Time point 3 is the year corresponding to the predicted rural ecological space. In this case, time point 1 is 2010, time point 2 is 2020, and time point 3 is 2030.

[0066] S3: Obtain basic data, preprocess the basic data, and create thematic maps based on the specific circumstances of each evaluation indicator, including:

[0067] S3.1 Basic data includes satellite remote sensing data of a village in 2010 and 2020, digital line map data, and ecological protection red line data;

[0068] S3.2 preprocessing includes projection conversion, image fusion, geometric correction, image enhancement and stitching processing of satellite remote sensing data, and data format conversion, resampling, reclassification and cropping of digital line map data and ecological protection red line data;

[0069] S3.3 The creation of thematic maps includes:

[0070] Land use type raster map: Based on the LUCC land use classification system of the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, land use maps of a village were obtained by combining preprocessed remote sensing image data from 2010 and 2020 with human-computer interactive interpretation. Land use types were divided into six categories: construction land, forest land, water area, shrubland and grassland, agricultural land, and bare land. Rasterized land use type raster maps of the village for 2010 and 2020 were then obtained.

[0071] Ecological space raster map: Combining land use raster maps of a village in 2010 and 2020 with remote sensing satellite imagery, forest land, water areas, shrubland and grassland were merged to obtain ecological space raster maps of a village in 2010 and 2020.

[0072] Spatial factor raster maps: Elevation and slope raster maps for 2010 and 2020 were obtained using GIS (Geographic Information System) spatial analysis methods. Multi-ring buffer analysis was used to obtain raster maps of distances to water bodies, roads, buildings, and forests.

[0073] Raster maps of ecological suitability factors: Raster maps of ecological sensitivity and importance of ecosystem service functions in 2010 and 2020 were obtained using the analytic hierarchy process (AHP) and weighted overlay method.

[0074] Policy Factor Raster Map: Based on the ecological protection red line delineated by Nanjing City, identify the areas inside and outside the ecological protection red line in 2010 and 2020, and obtain the ecological protection red line area raster map;

[0075] Land use change factor raster map: By overlaying and analyzing the land use maps of a village in 2010 and 2020, raster maps of changes in ecological land use, construction land use, and agricultural land use are obtained;

[0076] S4: Construct a Bayesian network model for predicting rural ecological space by combining the causal relationships between various indicators, such as... Figure 3 As shown, the formula for the Bayesian network model is:

[0077] S = (V, L) (1)

[0079] In equation (1), S represents the Bayesian structure, which is composed of the set of node variables V (V = {V1, V2, V3, ..., V...}). n}) and directed edge L (L=V i V j |V i V j The system consists of nodes (Vi, ∈V). Here, node variable Vi is an abstract representation of the predicted ecological space, and directed edges L are nodes (V, ∈V). i V j The dependency or causal relationship between them.

[0080] The parameters between node variables are a probability distribution set reflecting the local correlation between the nodes, and their expression is as follows:

[0081] P = {P(V)} i |V1, V2, V3, ..., V i-1 V i ∈V (2)

[0083] If V pi Represents variable V i If the set of parent nodes of V is given, then the joint probability distribution of V is:

[0084]

[0085] S5: Utilize a Bayesian network model for parameter learning to obtain a conditional probability table of rural ecological space evaluation indicators. The specific steps are as follows:

[0086] S5.1 uses rural ecological space evaluation indicators as node variables to construct a Bayesian network model structure;

[0087] S5.2 discretizes the spatial factor raster map, ecological suitability factor raster map, land use change factor raster map, policy factor raster map from 2010, and the ecological spatial raster map from 2020. Specifically, the spatial factor and ecological suitability factor raster data are discretized into three levels, while the land use change factor, policy factor raster data, and the 2020 ecological spatial raster data are discretized into two levels.

[0088] S5.3 converts all discretized raster data into point data, exports it as a set of sample point data containing the values ​​and coordinates of each feature, and converts it into ASCII format as training sample data (e.g., ...). Figure 4 (as shown);

[0089] S4.4 performs Bayesian network parameter learning on the training sample data, and obtains a complete conditional probability table after training (e.g., ...). Figure 5 (as shown);

[0090] S5: Import the conditional probability table into the Bayesian network model for rural ecological space prediction and perform rural ecological space prediction through the network inference engine. The specific steps are as follows:

[0091] S5.1 discretizes the spatial factor raster map, ecological suitability factor raster map, land use change factor raster map, and policy factor raster map of 2020 in GIS. The spatial factor and ecological suitability factor raster data are discretized into 3 levels, and the land use change factor and policy factor raster data are discretized into 2 levels.

[0092] S5.2 converts all discretized raster data into point data, exports it as a sample point data set containing the values ​​and coordinates of each feature, and converts it into ASCII format as evidence sample data;

[0093] S5.3 Import the evidence sample data into the Bayesian network model and update the model with data;

[0094] S5.4 uses a joint tree inference engine to calculate the probability distribution of each sample point in the rural ecological space in 2030;

[0095] S5.5 calculates the most probable interpretation to determine whether a sample point is located within the rural ecological space, and outputs the result as a set of point data.

[0096] S5.6 transforms point data into a raster map for analysis and visualization, yielding the ecological space of a village in 2030, such as... Figure 6 As shown in the figure, the area filled with diagonal lines represents the rural ecological space area obtained after Bayesian network inference.

[0097] Based on the same inventive concept, this invention also provides a computer device, comprising: one or more processors, and a memory for storing one or more computer programs; the programs include program instructions, and the processor executes the program instructions stored in the memory. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, used to implement one or more instructions, specifically for loading and executing one or more instructions stored in a computer storage medium to implement the above-described method.

[0098] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, performs the above-described method. This storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0099] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0100] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for predicting rural ecological space based on Bayesian networks, characterized in that, The method includes the following steps: By combining rural ecological characteristics and land use evolution patterns, suitable factors are selected, and a rural ecological space evaluation index system is established using these suitable factors to construct a Bayesian network model for rural ecological space prediction. Determine the time points involved in the Bayesian network model for predicting rural ecological space, including time point 1, time point 2, and time point 3; The time points involved in the Bayesian network model for predicting rural ecological space include time point 1, time point 2, and time point 3. The time span between time point 1 and time point 2, and between time point 2 and time point 3, must be consistent. Time point 3 is the year corresponding to the predicted rural ecological space. The formula for the Bayesian network model for predicting rural ecological space is as follows: （1） In the formula S Representing a Bayesian structure, S From the set of node variables V(V ={V 1 , V 2 , V 3 , …, V n }) and directed edges L(L = V i V j |V i ,V j ,∈V) Composition, including node variables V i It is an abstract representation of the predicted ecological space, with directed edges. L Node variables V i 、V j The dependency or causal relationship between them; The parameters between node variables are a probability distribution set reflecting the local correlations between nodes, and their expression is as follows: （2） use V pi Representing variables V i The set of parent nodes, then V The joint probability distribution is: （3） Acquire basic data and preprocess it. Then, use the preprocessed basic data to create a thematic map of rural ecological space evaluation indicators. By combining the causal relationships among the indicators in the thematic map of rural ecological space evaluation indicators, a Bayesian network model for predicting rural ecological space is constructed. By using a Bayesian network model for predicting rural ecological space, parameter learning is performed to obtain a conditional probability table of rural ecological space evaluation indicators, which serves as the basis for inference in the Bayesian network model for predicting rural ecological space. Rural ecological space prediction is performed using the inference engine of a Bayesian network model for rural ecological space prediction.

2. The method for predicting rural ecological space based on Bayesian networks according to claim 1, characterized in that, The rural ecological space evaluation index system includes spatial factors, ecological suitability factors, land use change factors, policy factors, and target factors.

3. The method for predicting rural ecological space based on Bayesian networks according to claim 2, characterized in that, The spatial factors include elevation, slope, distance from water bodies, distance from roads, distance from buildings, and distance from forest land; the ecological suitability factors include ecological sensitivity and the importance of ecosystem service functions; the land use change factors include changes in ecological land use, changes in construction land use, and changes in agricultural land use; the policy factors include ecological protection red lines; the target factors are the predicted rural ecological space; the ecological sensitivity includes topographic sensitivity, hydrological sensitivity, vegetation sensitivity, and land use sensitivity; and the importance of ecosystem service functions includes water conservation capacity and soil and water conservation capacity.

4. The method for predicting rural ecological space based on Bayesian networks according to claim 1, characterized in that, The basic data includes satellite remote sensing data at time points 1 and 2, digital line map data, and ecological protection red line data; The preprocessing of basic data includes projection conversion, image fusion, geometric correction, image enhancement and stitching of satellite remote sensing data, and data format conversion, resampling, reclassification and cropping of digital line map data and ecological protection red line data. The creation of thematic maps for rural ecological space evaluation indicators includes: Rural land use type raster map: Based on the LUCC land use classification system, the pre-processed satellite remote sensing data at time point 1 and time point 2 are combined with human-computer interactive interpretation to obtain the rural land use map. The land use types are divided into 6 categories, including: construction land, forest land, water area, shrubland and grassland, agricultural land and bare land. After rasterization, the rural land use type raster map at time point 1 and time point 2 is obtained. Rural ecological space raster map: By combining rural land use type raster maps at time point 1 and time point 2 with satellite remote sensing data, forest land, water area, shrubland and grassland are merged to obtain rural ecological space raster maps at time point 1 and time point 2. Spatial factor raster maps: Elevation and slope raster maps for time points 1 and 2 were obtained using GIS spatial analysis methods, and raster maps of distances to water bodies, roads, buildings, and forests were obtained using multi-ring buffer analysis. Ecological suitability factor raster: The ecological sensitivity and importance of ecosystem service functions at time point 1 and time point 2 were obtained by using the analytic hierarchy process and weighted overlay method. Policy Factor Raster Map: Identify the areas inside and outside the ecological protection red line at time points 1 and 2, and obtain the ecological protection red line area raster map; Land use change factor raster map: By overlaying and analyzing the land use maps at time point 1 and time point 2, raster maps of ecological land use change, construction land use change, and agricultural land use change are obtained.

5. The method for predicting rural ecological space based on Bayesian networks according to claim 1, characterized in that, The process of using a Bayesian network model for rural ecological space prediction to learn parameters and obtain a conditional probability table of rural ecological space evaluation indicators includes the following steps: A Bayesian network model for predicting rural ecological space was constructed using rural ecological space evaluation indicators as node variables. The raster data in the spatial factor raster map, ecological suitability factor raster map, land use change factor raster map, policy factor raster map at time point 1, and the ecological spatial raster map at time point 2 are discretized. Among them, the spatial factor and ecological suitability factor raster data are discretized into 3 levels, and the land use change factor, policy factor raster data and the ecological spatial raster data at time point 2 are discretized into 2 levels. All discretized raster data are converted into point data, exported as a set of sample point data containing the values ​​and coordinates of each feature, and converted into ASCII format as training sample data. The Bayesian network model for predicting rural ecological space uses training sample data to learn parameters and obtains a complete conditional probability table after training.

6. The method for predicting rural ecological space based on Bayesian networks according to claim 1, characterized in that, The process of predicting rural ecological space using the inference engine of the Bayesian network model for rural ecological space prediction includes the following steps: In GIS, the raster data of spatial factor raster, ecological suitability factor raster, land use change factor raster, and policy factor raster at time point 2 are discretized. The spatial factor and ecological suitability factor raster data are discretized into 3 levels, and the land use change factor and policy factor raster data are discretized into 2 levels. All discretized raster data are converted into point data, exported as a sample point data set containing the values ​​and coordinates of each feature, and converted into ASCII format as evidence sample data. The evidence sample data was imported into the Bayesian network model for predicting rural ecological space, and the data was updated in the Bayesian network model for predicting rural ecological space. The probability distribution of each sample point in the rural ecological space at time point 3 was calculated using the joint tree inference engine. The most probable interpretation is used to determine whether a sample point is located within the rural ecological space, and the result is output as a set of point data. The point data set was transformed into a data raster map for analysis and visualization, resulting in the rural ecological space at time point 3.

7. A computer device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When one or more of the programs are executed by one or more of the processors, the one or more processors implement a Bayesian network-based method for predicting rural ecological space as described in any one of claims 1-6.

8. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform a rural ecological space prediction method based on Bayesian networks as described in any one of claims 1-6.

Images