Method and device for identifying non-agrochemical high-risk area of cultivated land based on three-dimensional feature space

Abstract

The invention relates to the technical field of natural resource supervision, and particularly discloses a cultivated land non-agrochemical high-risk area identification method and device based on a three-dimensional feature space, and the method comprises the steps: obtaining basic data, and carrying out the preprocessing and time-space alignment processing; based on the processed data and an integrated learning model, determining target features corresponding to the cultivated land non-agrochemical degree and contribution values of the target features to the cultivated land non-agrochemical degree so as to determine weight values of the target features; calculating a natural score and a social economic score of the target area; the natural score of the target area is used as an X coordinate value of the target area, the social economic score is used as a Y coordinate value, and the cultivated land non-agrochemical degree value is used as a Z coordinate value; constructing a three-dimensional feature space model based on the coordinate information of all the target areas; and based on a tip mutation theory, determining a mutation threshold value of the three-dimensional feature space model, and based on the mutation threshold value, determining a cultivated land non-agrochemical high-risk area. Therefore, the identification accuracy of the non-agrochemical high-risk area of the cultivated land can be greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of natural resource supervision, and specifically discloses a method and device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space. Background Art

[0002] In the existing field of natural resource supervision, the identification and early warning of cultivated land non-agriculturalization are generally achieved through single remote sensing image analysis or manual interpretation, such as through the comparison of historical cultivated land vector maps and real-time remote sensing data to achieve change detection and early warning. However, in practical applications, both nature and social economy have certain impacts on cultivated land non-agriculturalization. Therefore, the accuracy of the existing technology for identifying and early warning cultivated land non-agriculturalization is not high, and the efficiency is low. Therefore, how to efficiently and accurately identify cultivated land non-agriculturalization has become an urgent problem to be solved at present. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method and device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, so as to overcome the problems of low efficiency and poor accuracy in the current identification of cultivated land non-agriculturalization.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present application provides a method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, including: Obtain basic data and perform preprocessing and spatio-temporal alignment processing on the basic data; wherein, the basic data includes land use data, remote sensing geographical data, meteorological data, and social economic data; Based on the processed basic data and the integrated learning model, determine the target features corresponding to the degree of cultivated land non-agriculturalization, and the contribution values of each target feature to the degree of cultivated land non-agriculturalization, and based on the contribution values of each target feature, determine the weight values of each target feature; wherein, the target features include natural features determined by remote sensing geographical data and meteorological data, and social economic features determined by social economic data, and the degree of cultivated land non-agriculturalization is determined based on land use data; Calculate the natural score of the target area based on the natural features and the corresponding weight values of the natural features; and calculate the social economic score of the target area based on the social economic features and the corresponding weight values of the social economic features; Take the natural score of the target area as the X coordinate value of the target area, the social economic score as the Y coordinate value of the target area, and the value of the degree of cultivated land non-agriculturalization as the Z coordinate value of the target area; Construct a three-dimensional feature space model based on the coordinate information of all target areas; Based on the cusp catastrophe theory, determine the mutation threshold of the three-dimensional feature space model, and determine the high-risk areas of cultivated land non-agriculturalization based on the mutation threshold.

[0005] Further, in some embodiments of the present application, the obtaining of the basic data and the preprocessing and spatio-temporal alignment processing of the basic data successively include: Performing data cleaning and missing value supplementation on the basic data; Performing data spatialization processing on the preprocessed meteorological data and socioeconomic data; Unifying the land use data, remote sensing geographic data, preprocessed meteorological data, and socioeconomic data under the target coordinate system; Making the spatial resolution of the data under the target coordinate system the same through upscaling and downscaling; Aligning the time nodes of the data; Performing normalization processing on all the data.

[0006] Further, in some embodiments of the present application, the remote sensing geographic data includes remote sensing images and DEM data; The natural features include NDVI and vegetation coverage determined from remote sensing images, terrain slope determined from DEM data, and air temperature and rainfall determined from meteorological data; The socioeconomic features include population density, urbanization rate, and traffic network density.

[0007] Further, in some embodiments of the present application, the target area includes the area corresponding to each spatial resolution grid under the same spatial resolution.

[0008] Further, in some embodiments of the present application, the calculating of the natural score of the target area based on the natural features and the corresponding weight values of the natural features includes: For each target area, taking the sum of the products of the values of the natural features of the target area and the corresponding weight values of the natural features as the natural score of the target area.

[0009] Further, in some embodiments of the present application, the calculating of the socioeconomic score of the target area based on the socioeconomic features and the corresponding weight values of the socioeconomic features includes: For each target area, taking the sum of the products of the values of the socioeconomic features of the target area and the corresponding weight values of the socioeconomic features as the socioeconomic feature score of the target area.

[0010] Further, in some embodiments of the present application, the degree of cultivated land non-agriculturalization is used to characterize the area and rate of the conversion of cultivated land to non-cultivated land.

[0011] Further, in some embodiments of the present application, the determining of the mutation threshold of the three-dimensional feature space model based on the cusp catastrophe theory includes: Determining the system potential energy function of the three-dimensional feature space model; Based on the equation that the second derivative of the system potential energy function is zero, determine the mutation point, and use the coordinates of the mutation point as the mutation threshold.

[0012] Further, in some embodiments of the present application, determining the high-risk area of cultivated land non-agriculturalization based on the mutation threshold includes: Determine the target area where the X coordinate value is greater than the X coordinate value of the mutation point, the Y coordinate value is greater than the Y coordinate value of the mutation point, and the Z coordinate value is greater than the Z coordinate value of the mutation point as the high-risk area of cultivated land non-agriculturalization.

[0013] In a second aspect, the present application provides a device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, which is characterized by including a processor and a memory, and the processor is connected to the memory: Wherein, the processor is used to call and execute the program stored in the memory; The memory is used to store the program, and the program is at least used to execute the above-mentioned method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space.

[0014] The present invention relates to the technical field of natural resources supervision, and specifically relates to a method and device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space. The method includes: obtaining basic data and performing preprocessing and spatio-temporal alignment processing on the basic data; wherein, the basic data includes land use data, remote sensing geographical data, meteorological data, and social and economic data; based on the processed basic data and an integrated learning model, determine the target features corresponding to the degree of cultivated land non-agriculturalization, and the contribution values of each target feature to the degree of cultivated land non-agriculturalization, and based on the contribution values of each target feature, determine the weight values of each target feature; wherein, the target features include natural features determined by remote sensing geographical data and meteorological data, and social and economic features determined by social and economic data, and the degree of cultivated land non-agriculturalization is determined based on land use data; calculate the natural score of the target area based on the natural features and the corresponding weight values of the natural features; and calculate the social and economic score of the target area based on the social and economic features and the corresponding weight values of the social and economic features; use the natural score of the target area as the X coordinate value of the target area, the social and economic score as the Y coordinate value of the target area, and the value of the degree of cultivated land non-agriculturalization as the Z coordinate value of the target area; construct a three-dimensional feature space model based on the coordinate information of all target areas; based on the cusp catastrophe theory, determine the mutation threshold of the three-dimensional feature space model, and determine the high-risk area of cultivated land non-agriculturalization based on the mutation threshold. In this way, the accuracy of identifying high-risk areas of non-agriculturalization can be greatly improved. Description of the Drawings

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 is a schematic flowchart of a method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by an embodiment of the present invention.

[0017] Figure 2 is a schematic diagram of the principle of a method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by an embodiment of the present invention.

[0018] Figure 3 is a schematic structural diagram of a device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by an embodiment of the present invention. Detailed implementation manners

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will describe the technical solutions of the present invention in detail. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other implementation manners obtained by those of ordinary skill in the art without creative efforts fall within the scope protected by the present invention.

[0020] Figure 1 is a schematic flowchart of a method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by an embodiment of the present invention. Please refer to Figure 1 , this embodiment may include the following steps: S101. Obtain basic data and perform preprocessing and spatio-temporal alignment processing on the basic data.

[0021] Specifically, the basic data includes land use data, remote sensing geographical data, meteorological data, and social and economic data. The preprocessing includes data cleaning and missing value supplementation for the data. The spatio-temporal alignment processing includes spatializing the non-spatial structured data such as meteorological data and social and economic data after preprocessing and aligning all the obtained data in terms of time and space for subsequent use.

[0022] S102. Based on the processed basic data and the integrated learning model, determine the target features corresponding to the degree of cultivated land non-agriculturalization and the contribution values of each target feature to the degree of cultivated land non-agriculturalization, and determine the weight values of each target feature based on the contribution values of each target feature.

[0023] Specifically, in this application, based on the integrated learning model, the Shapley Additive explanations (SHAP) is used to determine the target features corresponding to the degree of cultivated land conversion to non-agricultural use, and the contribution values of each target feature to the degree of cultivated land conversion to non-agricultural use. Then, based on the contribution values of each target feature, the weight values of each target feature are determined.

[0024] Among them, the target features include natural features determined by remote sensing geographic data and meteorological data, and socioeconomic features determined by socioeconomic data.

[0025] S103. Calculate the natural score of the target area based on the natural features and the corresponding weight values of the natural features; and calculate the socioeconomic score of the target area based on the socioeconomic features and the corresponding weight values of the socioeconomic features.

[0026] S104. Use the natural score of the target area as the X coordinate value of the target area, the socioeconomic score as the Y coordinate value of the target area, and the value of the degree of cultivated land conversion to non-agricultural use as the Z coordinate value of the target area.

[0027] S105. Construct a three-dimensional feature space model based on the coordinate information of all target areas.

[0028] Specifically, in this application, after the spatio-temporal alignment processing of the basic data, the spatial resolutions of all data will be unified, that is, the spatial resolution grids corresponding to the data are of the same specification (such as 100 meters). On this basis, the area corresponding to each spatial resolution grid is regarded as a target area. In this way, using the natural score as the X axis, the socioeconomic score as the Y axis, and the value of the degree of cultivated land conversion to non-agricultural use as the Z axis, after calculating the X coordinate value, Y coordinate value, and Z coordinate value of each target area respectively, a three-dimensional feature space model is constructed. It can be understood that in this three-dimensional feature space model, each point corresponds to the natural score, socioeconomic score, and the value of the degree of cultivated land conversion to non-agricultural use of a target area.

[0029] S106. Based on the cusp catastrophe theory, determine the catastrophe threshold of the three-dimensional feature space model, and determine the high-risk areas of cultivated land conversion to non-agricultural use based on the catastrophe threshold.

[0030] Specifically, in practical applications, the system potential energy function of the three-dimensional feature space model can be determined first, and the catastrophe threshold of the system of the three-dimensional feature space model, that is, the coordinates of the catastrophe point, can be determined by using the equation that the second derivative of the potential energy function is zero. Then, based on the catastrophe threshold, the high-risk areas of cultivated land conversion to non-agricultural use are determined from all target areas.

[0031] The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by this application takes into account both natural and socio-economic factors affecting the degree of cultivated land non-agriculturalization, avoiding the insufficient explanatory power of a single natural or socio-economic factor for the non-agriculturalization status. At the same time, by constructing a three-dimensional feature space and identifying the tip mutation points to quantify the non-agriculturalization risk threshold, the process is simple and highly operable, which can improve the accuracy, dynamics, and robustness of the identification of high-risk areas of cultivated land non-agriculturalization, provide an efficient quantification tool for the risk identification of cultivated land non-agriculturalization in natural resource monitoring and supervision, and support the transformation of the governance model from passive response to active early warning.

[0032] In some embodiments of this application, the remote sensing geographical data may specifically include remote sensing images (such as Landsat and Sentinel images) and Digital Elevation Model (DEM) data. Correspondingly, the natural features include the Normalized Difference Vegetation Index (NDVI) and vegetation coverage determined by the remote sensing images, the terrain slope determined by the DEM data, and the temperature and rainfall determined by the meteorological data (the observation data of meteorological stations). The socio-economic data covers population density (such as from WorldPop), urbanization rate (such as from public data of the National Bureau of Statistics and the World Bank, etc.), and traffic network density (such as from Open StreetMap). Correspondingly, the socio-economic features may include the population density, urbanization rate, and traffic network density mentioned above. The land use data, that is, the cultivated land data, is used to quantify the non-agriculturalization area and rate of the conversion of cultivated land to non-cultivated land.

[0033] Figure 2 It is a schematic diagram of the principle of the method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space provided by an embodiment of the present invention. As Figure 2 shown, in some embodiments of this application, obtaining the basic data and performing preprocessing and spatio-temporal alignment processing on the basic data may specifically include, in sequence: Data preprocessing specifically includes: data cleaning and missing value supplementation for all basic data. Among them, different operations can be performed on different data. For example, cloud masking and outlier removal processing are performed on remote sensing images, and spatial interpolation processing is performed on meteorological data.

[0034] Then spatio-temporal alignment processing is performed. It should be noted that since some of the data obtained in this application above is in the form of a spatial structure, and some are not in the form of a spatial structure (that is, whether each data is for a spatial region), when performing spatio-temporal alignment processing on non-spatial structure data, it is first necessary to convert it into spatial structure data, that is, first perform data spatialization, and then perform spatio-temporal alignment. Specifically: Data spatialization, specifically including: performing data spatialization processing on the preprocessed meteorological data and socioeconomic data, so that the meteorological data and socioeconomic data are converted into a spatial structure form, laying a foundation for subsequent spatio-temporal alignment of data. Since the land use data and remote sensing geographic data are already in the form of a spatial structure, data spatialization processing can be skipped, and the spatio-temporal alignment step can be directly carried out.

[0035] Spatio-temporal alignment, specifically including: First, unify the land use data, remote sensing geographic data, meteorological data and socioeconomic data after data spatialization processing to the target coordinate system, such as unifying to the WGS84 UTM coordinate system; then make the spatial resolutions of the data the same by upscaling and downscaling, such as making the spatial resolution grids all 100 meters; and align the time nodes of the data to unify the data to the same time year. For example, unify the data around 2000 in the above four types of data to 2000, and the data around 2010 to 2010, so as to perform data processing and analysis based on the unified time year subsequently.

[0036] Normalization processing, specifically including: unifying the data to the dimensionless scale of (0, 1) (in some embodiments, only the remote sensing geographic data, meteorological data and socioeconomic data can be normalized with multiple factors).

[0037] Furthermore, when determining the target features and corresponding weights in this application, screening and weight quantification can be performed based on the feature importance of ensemble learning.

[0038] Specifically, the contribution of each feature to the prediction of cultivated land non-agriculturalization can be quantified based on the SHAP theory, and it can be judged whether the relationship between the feature and the final prediction result is positive correlation, negative correlation or other more complex correlations, so as to screen the relevant features, determine the target features, and determine the weights of each target feature based on the SHAP of the screened target features.

[0039] Among them, the formula for determining the weights of each target feature is specifically:

[0040] Among them, represents the weight of the target feature i; is the importance, that is, the contribution value of the target feature i determined based on the SHAP theory; N is the total number of the screened target features.

[0041] It should be noted that under the influence of climate change, human activities, etc., the structure and function of a system may undergo large-scale mutations, leading to the system transitioning from a relatively stable state to another stable state. This phenomenon is called steady-state conversion. In this application, by establishing a three-dimensional relationship among natural factors, socioeconomic factors, and the system state (i.e., the degree of farmland conversion to non-farmland), the mutation threshold of farmland conversion to non-farmland is quantified in three-dimensional space, thereby identifying high-risk areas of farmland conversion to non-farmland.

[0042] Specifically, first for each target area mentioned above, the sum of the products of the values of each natural feature of the target area and the corresponding weight values of the natural features is used as the natural score of the target area. And the sum of the products of the values of each socioeconomic feature of the target area and the corresponding weight values of the socioeconomic features is used as the socioeconomic feature score of the target area. The specific formulas are as follows:

[0043] Among them, represents the natural score, represents the i-th natural feature, and M is the total number of natural features.

[0044]

[0045] Among them, represents the socioeconomic score, represents the i-th socioeconomic feature, and N is the total number of socioeconomic features.

[0046] On this basis, in this application, the degree of farmland conversion to non-farmland is used as the Z-axis representing the sustainable or unsustainable state of the system, the natural score is used as the X-axis, and the socioeconomic score is used as the Y-axis. X-Y together constitute the disturbing factors of the non-ideal state affecting farmland conversion to non-farmland, and a three-dimensional state diagnosis space of the risk state of farmland conversion to non-farmland and the natural and social system is constructed.

[0047] In some embodiments of this application, the coordinate system and coordinate values can be standardized. For example, first, the degree of farmland conversion to non-farmland is defined as the proportion of the standardized non-farmland area, which is expressed as follows:

[0048] Among them, A 耕地初始 represents the initial farmland area of the target area within the calculation time range; A 非农化 represents the area of the farmland in the target area converted to non-farmland within the calculation time range; is the calculation time range.

[0049] On this basis, using statistical principles, the coordinate values calculated above are standardized and converted, which is specifically expressed as follows: X' =

[0050] Y' =

[0051] Z’ =

[0052] Wherein, 、 and respectively correspond to the coordinate values after the standardization of X, Y, and Z. X represents the mean value of all X coordinates. X represents the standard deviation of all X coordinates (subsequently, Y, Z, Y and Z have the same principle and can be understood by referring to X and X, and will not be elaborated here).

[0053] Thus, based on the data principle, the system potential energy function of the three-dimensional feature space model constructed above can be expressed as:

[0054] The system stable state satisfies that the first derivative of the potential energy function is zero (equilibrium surface equation), and can be expressed as:

[0055] The mutation threshold is determined by the second derivative of the potential energy function being zero, and can be expressed as:

[0056]

[0057] Thus, by using the equation where the second derivative of the above potential energy function is zero, the mutation points in the three-dimensional feature space model can be determined, and the coordinates of the mutation points are used as the mutation threshold. Furthermore, the high-risk area of cultivated land non-agriculturalization is determined by using the coordinates of the mutation points, that is, the mutation threshold. For example, the target area where the X coordinate value is greater than the X coordinate value of the mutation point, the Y coordinate value is greater than the Y coordinate value of the mutation point, and the Z coordinate value is greater than the Z coordinate value of the mutation point can be determined as the high-risk area of cultivated land non-agriculturalization (or in some other embodiments of the present application, the high-risk area of cultivated land non-agriculturalization can also be determined by comparing only one or two coordinates, such as the target area where the X coordinate value is greater than the X coordinate value of the mutation point and the Z coordinate value is greater than the Z coordinate value of the mutation point is determined as the high-risk area of cultivated land non-agriculturalization).

[0058] Based on the same inventive concept, the present invention also provides a device for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, which is used to implement the above method embodiments.Figure 3 FIG. Figure 3 is a schematic structural diagram of an identification device for high - risk areas of cultivated land non - agriculturalization based on a three - dimensional feature space provided by an embodiment of the present invention. As Figure 3 shown, the identification device for high - risk areas of cultivated land non - agriculturalization based on a three - dimensional feature space in this embodiment includes a processor 11 and a memory 12, and the processor 11 is connected to the memory 12. Among them, the processor 11 is used to call and execute the program stored in the memory 12; the memory 12 is used to store the program, and this program is at least used to execute the method for identifying high - risk areas of cultivated land non - agriculturalization based on a three - dimensional feature space in the above embodiments.

[0059] The specific implementation scheme of the identification device for high - risk areas of cultivated land non - agriculturalization based on a three - dimensional feature space provided by the embodiments of the present application can refer to the implementation manner of the method for identifying high - risk areas of cultivated land non - agriculturalization based on a three - dimensional feature space in any of the above embodiments, and will not be elaborated here.

[0060] It can be understood that the same or similar parts in the above embodiments can be referred to each other, and the content not detailed in some embodiments can be seen in the same or similar content in other embodiments.

[0061] It should be noted that in the description of the present invention, terms such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "a plurality of" refers to at least two.

[0062] Any process or method description in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0063] It should be understood that each part of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in the memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following well - known technologies in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application - specific integrated circuits with suitable combinational logic gate circuits, programmable gate arrays (PGA), field - programmable gate arrays (FPGA), etc.

[0064] Those of ordinary skill in the art can understand that all or part of the steps carried out in the method of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0065] In addition, each functional unit in the various embodiments of the present invention can be integrated in a processing module, or each unit can exist physically alone, or two or more units can be integrated in one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0066] The above-mentioned storage medium can be a read-only memory, a magnetic disk or an optical disc, etc.

[0067] In the description of this specification, the descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0068] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, characterized in that Including: Obtain basic data and perform preprocessing and spatio-temporal alignment processing on the basic data; wherein, the basic data includes land use data, remote sensing geographical data, meteorological data and socioeconomic data; Based on the processed basic data and the integrated learning model, determine the target features corresponding to the degree of cultivated land non-agriculturalization, and the contribution value of each target feature to the degree of cultivated land non-agriculturalization, and based on the contribution value of each target feature, determine the weight value of each target feature; wherein, the target features include natural features determined by remote sensing geographical data and meteorological data, and socioeconomic features determined by socioeconomic data, and the degree of cultivated land non-agriculturalization is determined based on land use data; Calculate the natural score of the target area based on the natural features and the weight values corresponding to the natural features; and calculate the socioeconomic score of the target area based on the socioeconomic features and the weight values corresponding to the socioeconomic features; Take the natural score of the target area as the X coordinate value of the target area, the socioeconomic score as the Y coordinate value of the target area, and the value of the degree of cultivated land non-agriculturalization as the Z coordinate value of the target area; Construct a three-dimensional feature space model based on the coordinate information of all target areas; Based on the cusp catastrophe theory, determine the mutation threshold of the three-dimensional feature space model, and determine the high-risk area of cultivated land non-agriculturalization based on the mutation threshold.

2. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 1, wherein The obtaining of basic data and performing preprocessing and spatio-temporal alignment processing on the basic data sequentially includes: Perform data cleaning and missing value supplementation on the basic data; Perform data spatialization processing on the preprocessed meteorological data and socioeconomic data; Unify the land use data, remote sensing geographical data, meteorological data and socioeconomic data after data spatialization processing under the target coordinate system; Make the data spatial resolution the same under the target coordinate system by upscaling and downscaling; Align the time nodes of the data; Normalize all the data.

3. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 1, wherein The remote sensing geographical data includes remote sensing images and DEM data; The natural features include NDVI and vegetation coverage determined by remote sensing images, terrain slope determined by DEM data, and air temperature and rainfall determined by meteorological data; The socioeconomic features include population density, urbanization rate and traffic network density.

4. The method for identifying high-risk areas of cultivated land conversion to non-agricultural use based on a three-dimensional feature space according to claim 2, wherein, The target area includes the area corresponding to each spatial resolution grid under the same spatial resolution.

5. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 4, wherein The calculating of the natural score of the target area based on the natural features and the weight values corresponding to the natural features includes: For each target area, take the sum of the products of the values of each natural feature of the target area and the weight values corresponding to the natural features as the natural score of the target area.

6. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 4, characterized in that The calculating of the socioeconomic score of the target area based on the socioeconomic features and the weight values corresponding to the socioeconomic features includes: For each target area, take the sum of the products of the values of each socioeconomic feature of the target area and the weight values corresponding to the socioeconomic features as the socioeconomic feature score of the target area.

7. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 1, wherein The degree of cultivated land non-agriculturalization is used to characterize the area and rate of the conversion of cultivated land to non-cultivated land.

8. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 1, characterized in that The determining of the mutation threshold of the three-dimensional feature space model based on the cusp catastrophe theory includes: Determine the system potential energy function of the three-dimensional feature space model; Based on the equation that the second derivative of the system potential energy function is zero, determine the mutation point, and use the coordinates of the mutation point as the mutation threshold.

9. The method for identifying high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space according to claim 6, wherein, Determining the high-risk area of cultivated land non-agriculturalization based on the mutation threshold includes: Determine the target area where the X coordinate value is greater than the X coordinate value of the mutation point, the Y coordinate value is greater than the Y coordinate value of the mutation point, and the Z coordinate value is greater than the Z coordinate value of the mutation point as the high-risk area of cultivated land non-agriculturalization.

10. An identifying device for high-risk areas of cultivated land non-agriculturalization based on a three-dimensional feature space, characterized in that, It includes a processor and a memory, and the processor is connected to the memory: Among them, the processor is used to call and execute the program stored in the memory; The memory is used to store the program, and the program is at least used to execute the method for identifying the high-risk area of cultivated land non-agriculturalization based on the three-dimensional feature space according to any one of claims 1-9.

Images