Land tilling suitability prediction model construction method and device and land tilling suitability evaluation method and device

Abstract

The invention relates to the technical field of ecological suitability evaluation, and discloses a land tillage suitability prediction model construction method and device and a land tillage suitability evaluation method and device. The land cultivability prediction model construction method comprises the steps that target land cultivability evaluation result grid data, target evaluation index grid data and a land cultivability index system of a target area are acquired, the land cultivability index system comprises a plurality of evaluation dimensions, and each evaluation dimension comprises a plurality of evaluation indexes; a Bayesian network structure is constructed based on the land cultivability index system, root nodes in the Bayesian network structure are a plurality of evaluation indexes in the land cultivability index system, middle nodes are evaluation dimensions, and leaf nodes are land cultivability evaluation results; and converting the target land cultivability evaluation result raster data and the target evaluation index raster data into a sample data table, and inputting the sample data table into a Bayesian network structure for parameter learning to obtain a land cultivability prediction model.

Description

Technical Field

[0001] This invention relates to the field of ecological suitability assessment technology, specifically to the construction of land arability prediction models and methods and apparatus for land arability assessment. Background Technology

[0002] Land suitability assessment is a core component of agricultural resource management, helping farmers accurately determine the arable potential and limiting factors of their own land plots. However, related technologies often focus on a macro-level perspective, outputting land suitability assessment results primarily to provide decision support for management departments or planning agencies in developing land use plans. Assessment models are designed for regional planning, with evaluation indicators at a macro-level, failing to consider the heterogeneity at the micro-level. This leads to a disconnect between the assessment results and farmers' micro-level needs, making it impossible to directly guide farmers in developing reserve arable land resources at the plot-level. Summary of the Invention

[0003] This invention provides a land arability prediction model construction method and apparatus for land arability assessment, in order to solve the problem that in related technologies, the assessment model design is oriented towards regional planning, the spatial scale of the evaluation indicators is relatively macroscopic, and the heterogeneity differences at the micro scale are not considered, resulting in the evaluation results being out of touch with the micro-needs of farmers and unable to directly guide farmers in developing reserve arable land resources at the field-level micro scale.

[0004] In a first aspect, the present invention provides a method for constructing a land arability prediction model. The method includes: acquiring raster data of target land arability evaluation results, target evaluation index raster data, and a land arability index system for a target area. The land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators. A Bayesian network structure is constructed based on the land arability index system. In the Bayesian network structure, each root node represents multiple evaluation indicators in the land arability index system, the intermediate nodes represent evaluation dimensions, and the leaf nodes represent land arability evaluation results. The target land arability evaluation result raster data and the target evaluation index raster data are converted into a sample data table. The sample data table is then input into the Bayesian network structure for parameter learning to obtain the land arability prediction model.

[0005] The land arableness prediction model construction method provided by this invention constructs a Bayesian network structure with evaluation indicators as root nodes, evaluation dimensions as intermediate nodes, and arableness evaluation results as leaf nodes. It utilizes macro-level raster data for parameter learning, effectively solving the problem of macro-level evaluation scales being disconnected from farmers' micro-level needs in related technologies. Compared to traditional deterministic models oriented towards regional planning, the Bayesian network of this invention can probabilistically fuse prior knowledge from macro-level raster data with posterior data from farmers' field experience. This preserves statistical regularities at the regional level while fully considering heterogeneous differences at the field level, transforming the evaluation results from macro-level planning basis for management departments into micro-level decision support that can directly guide farmers' development of reserve arable land resources. Simultaneously, the Bayesian network naturally supports incomplete data input and quantifies uncertainty through probability distribution and confidence intervals, significantly improving the model's applicability and robustness in scenarios where farmers can only provide partial experience data.

[0006] In one optional implementation, the target land arability evaluation result raster data and target evaluation index raster data are obtained through the following steps: obtaining the land arability evaluation result raster data and evaluation index raster data of the target area; performing coordinate unification processing on the land arability evaluation result raster data and evaluation index raster data to obtain the processed land arability evaluation result raster data and evaluation index raster data; performing spatial scale unification processing on the processed land arability evaluation result raster data and evaluation index raster data to obtain the target land arability evaluation result raster data and target evaluation index raster data.

[0007] The method provided by this optional implementation method ensures the consistency of the target raster data in terms of spatial reference and resolution by sequentially performing coordinate and spatial scale unification processing on multi-source raster data. It eliminates errors caused by differences in format and scale of raster data from different sources, providing standardized and high-quality basic data for subsequent conversion of raster data into sample data tables and Bayesian network parameter learning. This ensures the accuracy and scientific nature of the construction of the land arability prediction model, and also lays a reliable data foundation for subsequent integration of micro-data from farmers' fields to achieve macro-to-micro arability evaluation.

[0008] Secondly, the present invention provides a method for assessing land arability. The method includes: acquiring a post-verification dataset of the field to be assessed, wherein the post-verification dataset is determined based on a field experience dataset of the field to be assessed, the field experience dataset being used to characterize the empirical values ​​of multiple assessment indicators for the field to be assessed; inputting the post-verification dataset into a pre-constructed land arability level prediction model, so that the land arability level prediction model calculates the land arability probability distribution and confidence interval of the field to be assessed, the land arability level prediction model being constructed according to the land arability prediction model construction method of the first aspect or its corresponding implementation; and determining the arability assessment result of the field to be assessed based on the land arability probability distribution and confidence interval of the field to be assessed.

[0009] The land arableness assessment method provided by this invention uses a Bayesian network-based land arableness level prediction model pre-constructed from standardized macro-grid data. It transforms farmers' empirical assessment index values ​​based on actual field conditions into a post-validation data dataset input to the model. Through model inference, it obtains the field-specific arableness probability distribution and confidence interval, thereby determining the assessment result. This method overcomes the limitation of existing macro-evaluation techniques that cannot directly guide farmers' micro-level arable land development. It effectively integrates macro-regional prior knowledge with farmers' micro-level field experience, fully considering the heterogeneity characteristics at the field level, ensuring the assessment results accurately match farmers' actual production needs. Simultaneously, leveraging the characteristics of the Bayesian network model, it adapts to the reality of inaccurate and incomplete farmer experience data, and can quantify the reliability of the assessment results through confidence intervals. Compared to the singular macro-evaluation conclusions output by related technologies, the assessment results of this invention are more targeted and referential, directly guiding farmers in making decisions regarding the development of reserve arable land resources at the field level. This allows land arableness assessment to truly move from the macro-level of serving regional planning to the micro-level of serving farmers' production.

[0010] In an optional implementation, the method further includes: if the arability level of the field to be evaluated is determined to be lower than a preset level based on the arability assessment results, obtaining the land arability probability distribution for each assessment dimension, wherein the land arability probability distribution for each assessment dimension is calculated using a land arability level prediction model; determining the suitability level for the corresponding assessment dimension based on the land arability probability distribution for each assessment dimension; and generating structured management text based on the suitability level for each assessment dimension.

[0011] The method provided by this optional implementation method, for fields that do not meet the suitability level, determines the suitability level of each dimension based on the probability distribution of arability of each assessment dimension directly output by the model, and generates structured management text. This solves the problem that existing macro-evaluations only output overall results and lack targeted improvement guidance. It can accurately locate the weak dimensions of field suitability, provide farmers with concrete and implementable improvement and management suggestions, and transform the field-level arability assessment results from a simple decision reference into a practical basis for guiding farmers' actual farmland development and improvement. This greatly improves the practicality and implementability of the assessment results and is more in line with the actual needs of farmers' micro-production.

[0012] In one alternative implementation, after the step of generating structured management text based on the suitability levels of each assessment dimension, the method further includes sending the structured management text of the field to be assessed and the arability assessment results of the field to be assessed to a display terminal for display.

[0013] The method provided by this optional implementation sends the arability assessment results and structured management text to a display terminal for display. This meets the actual needs of farmers who lack professional operation skills, allowing them to intuitively and conveniently obtain the arability conclusions of the fields and targeted improvement suggestions, directly guiding the development of arable land and enhancing its practical value.

[0014] Thirdly, the present invention provides a land arability prediction model construction device, which includes: a first acquisition module, used to acquire raster data of target land arability evaluation results, target evaluation index raster data, and a land arability index system for a target area, wherein the land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators; a construction module, used to construct a Bayesian network structure based on the land arability index system, wherein each root node in the Bayesian network structure is a multiple evaluation indicator in the land arability index system, the intermediate nodes are evaluation dimensions, and the leaf nodes are land arability evaluation results; and a first determination module, used to convert the target land arability evaluation result raster data and the target evaluation index raster data into a sample data table, input the sample data table into the Bayesian network structure for parameter learning, and obtain a land arability prediction model.

[0015] Fourthly, the present invention provides a land arability assessment device, comprising: a second acquisition module for acquiring a post-verification data set of the field to be assessed, the post-verification data set being determined based on a field experience data set of the field to be assessed, the field experience data set being used to characterize the empirical values ​​of multiple assessment indicators of the field to be assessed; a calculation module for inputting the post-verification data set into a pre-constructed land arability level prediction model, so that the land arability level prediction model calculates the land arability probability distribution and confidence interval of the field to be assessed, the land arability level prediction model being constructed according to the land arability prediction model construction method of the first aspect or its corresponding implementation; and a second determination module for determining the arability assessment result of the field to be assessed based on the land arability probability distribution and confidence interval of the field to be assessed.

[0016] Fifthly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the land arability prediction model construction method of the first aspect or its corresponding embodiment, or to perform the land arability assessment method of the second aspect or any corresponding embodiment.

[0017] In a sixth aspect, the present invention provides a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the land arability prediction model construction method of the first aspect or its corresponding embodiment, or to execute the land arability assessment method of the second aspect or any corresponding embodiment.

[0018] In a seventh aspect, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the land arability prediction model construction method of the first aspect or its corresponding embodiment, or to execute the land arability assessment method of the second aspect or any corresponding embodiment. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention;

[0021] Figure 2This is a flowchart illustrating the method for constructing a land arability prediction model according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the first process of the land arability assessment method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the second process of the land suitability assessment method according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a land arability grade prediction model based on Bayesian networks in an embodiment of this application; Figure 6 This is a schematic diagram of the evaluation results of the land arability grade prediction model based on Bayesian network in the embodiments of this application; Figure 7 This is a structural block diagram of a land arability prediction model construction device according to an embodiment of the present invention; Figure 8 This is a structural block diagram of a land suitability assessment device according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0023] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0024] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0025] As an optional application scenario of this invention, the specific application environment architecture or specific hardware architecture on which the land arability prediction model construction method or land arability assessment method depends is described herein. For example... Figure 1 As shown, the architecture system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0026] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0027] In related technologies, most focus on a macro-scale approach, outputting land suitability assessment results, primarily providing decision support for management departments or planning agencies in preparing land use plans. The assessment model design is geared towards regional planning, with evaluation indicators at a relatively macro-scale, failing to consider the heterogeneity at the micro-scale. This leads to a disconnect between the assessment results and farmers' micro-level needs, making it impossible to directly guide farmers in developing reserve arable land resources at the plot-level micro-scale.

[0028] In view of this, this application provides a method for constructing a land arability prediction model, which can be applied to a server to construct the model. The method provided in this application constructs a Bayesian network structure with evaluation indicators as root nodes, evaluation dimensions as intermediate nodes, and arability evaluation results as leaf nodes, and utilizes macro-level raster data to complete parameter learning. This effectively solves the problem in related technologies where the evaluation scale is macroscopic and disconnected from farmers' micro-level needs. Compared to traditional deterministic models oriented towards regional planning, the Bayesian network of this invention can probabilistically fuse prior knowledge from macro-level raster data with posterior data from farmers' field experience. This preserves statistical regularities at the regional level while fully considering heterogeneous differences at the field level, transforming the evaluation results from a macro-level planning basis for management departments into micro-level decision support that can directly guide farmers' development of reserve arable land resources. Simultaneously, the Bayesian network naturally supports incomplete data input and quantifies uncertainty through probability distribution and confidence intervals, significantly improving the model's applicability and robustness in scenarios where farmers can only provide partial experience data.

[0029] According to an embodiment of the present invention, a method for constructing a land arability prediction model is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] This embodiment provides a method for constructing a land suitability prediction model, which can be used in the aforementioned server. Figure 2 This is a flowchart of a method for constructing a land arability prediction model according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain raster data of target land arability evaluation results, target evaluation index raster data, and land arability index system for the target area. The land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators.

[0031] For example, the target area refers to the specific geographical area targeted by this invention for constructing the land arability prediction model and subsequently conducting field-level arability assessments. This area can be defined according to actual needs, such as a county, mountainous region, irrigation district, or specific agricultural production area. The target land arability evaluation result raster data can be preprocessed raster data within the target area, using regular raster cells as the basic unit. Each raster cell is labeled with the corresponding comprehensive land arability evaluation result. In this embodiment, the comprehensive land arability evaluation result may include, but is not limited to, land arability level information. Multiple evaluation dimensions may include, but are not limited to, four dimensions: topography, climate, soil, and location. Each dimension may include 3 to 5 evaluation indicators.

[0032] In some optional implementations, the raster data of the target land arability evaluation results and the target evaluation index raster data are obtained through the following steps: Step a1: Obtain raster data of land suitability evaluation results and evaluation index raster data for the target area.

[0033] For example, in this embodiment of the application, the raster data of the regional scale land arability evaluation results and the raster data of the evaluation indicators are obtained from the suitability level data generated by a professional data platform or other authoritative sources.

[0034] Step a2 involves performing coordinate unification processing on the raster data of land arability evaluation results and the raster data of evaluation indicators to obtain the processed raster data of land arability evaluation results and the raster data of evaluation indicators.

[0035] For example, in this embodiment of the application, geographic information system software is used to transform all input data into a geographic coordinate system using a projection conversion tool, thereby obtaining processed land arability evaluation result raster data and evaluation index raster data.

[0036] Step a3 involves performing spatial scale unification processing on the processed land arability evaluation result raster data and evaluation index raster data to obtain target land arability evaluation result raster data and target evaluation index raster data.

[0037] For example, in this embodiment of the application, one of the bilinear interpolation method, nearest neighbor interpolation method, or ordinary kriging interpolation method is used to perform batch resampling processing on the processed land arability evaluation result raster data and evaluation index raster data through ArcGIS's Resample tool or Python's Rasterio library. The resampling goal is to unify to the highest resolution in the data source to ensure maximum preservation of detailed information. Finally, the target land arability evaluation result raster data and target evaluation index raster data with uniform spatial resolution are output.

[0038] Step S202: Construct a Bayesian network structure based on the land arability index system. In the Bayesian network structure, each root node is a multiple evaluation index in the land arability index system, the intermediate nodes are evaluation dimensions, and the leaf nodes are the land arability evaluation results.

[0039] For example, in this embodiment of the application, a directed acyclic Bayesian network structure is constructed based on the land arability index system. The Bayesian network structure includes node definitions and dependencies. The root node is the evaluation index, the intermediate nodes are the arability levels of terrain, climate, soil, and location, and the leaf nodes are the land arability levels. The network topology can be built using Netica software or Python's pgmpy library.

[0040] Step S203: Convert the raster data of the target land arability evaluation results and the target evaluation index raster data into a sample data table, input the sample data table into a Bayesian network structure for parameter learning, and obtain the land arability prediction model.

[0041] For example, in this embodiment of the application, the raster data of the target land arability evaluation results and the target evaluation index raster data are converted into a sample data table, input into a Bayesian network structure, and the Expectation-Maximization Algorithm is used for parameter learning. The conditional probability distribution of each arability level is calculated through the maximum posterior probability to obtain a land arability level prediction model based on the Bayesian network, which includes a prior probability vector (e.g., highly suitable = 0.6, fairly suitable = 0.3, unsuitable = 0.1).

[0042] The land arableness prediction model construction method provided in this embodiment constructs a Bayesian network structure with evaluation indicators as root nodes, evaluation dimensions as intermediate nodes, and arableness evaluation results as leaf nodes. It utilizes macro-level raster data for parameter learning, effectively solving the problem of macro-level evaluation scales being disconnected from farmers' micro-level needs in related technologies. Compared to traditional deterministic models oriented towards regional planning, the Bayesian network of this invention can probabilistically fuse prior knowledge from macro-level raster data with posterior data from farmers' field experience. This preserves statistical regularities at the regional level while fully considering heterogeneous differences at the field level, transforming the evaluation results from a macro-level planning basis for management departments into micro-level decision support that can directly guide farmers' development of reserve arable land resources. Simultaneously, the Bayesian network naturally supports incomplete data input and quantifies uncertainty through probability distribution and confidence intervals, significantly improving the model's applicability and robustness in scenarios where farmers can only provide partial experience data.

[0043] This embodiment also provides a land suitability assessment method, which can be used in the aforementioned server. Figure 3 This is a flowchart of a land suitability assessment method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: Obtain the post-validation data set of the field to be evaluated. The post-validation data set is determined based on the field experience data set of the field to be evaluated. The field experience data set is used to characterize the empirical values ​​of multiple evaluation indicators of the field to be evaluated.

[0044] For example, in this embodiment of the application, farmers input all or part of the information they have into the terminal to obtain a field experience dataset. The field experience dataset may include: (1) location information: geographical coordinates of the field (latitude and longitude); (2) terrain information: altitude, slope, and aspect; (3) soil observation data: such as descriptive information such as soil texture, soil moisture, and soil depth; (4) location data: such as walking time to the nearest water source, walking time to the settlement, and walking time to the nearest road. The terminal device performs unit unification and verification on the data, and uses data verification algorithms (such as regular expression matching and range checking) to standardize the format and perform logical verification on the input data. For example, the Pandas library of Python is used for data cleaning to ensure that the coordinates are within a reasonable range (latitude -90° to 90°, longitude -180° to 180°) and the soil texture conforms to the preset category (such as sandy soil, loam, and clay). The data is encrypted and uploaded to the cloud server through the HTTPS protocol or MQTT communication protocol, and the data is encapsulated in JSON format. The server uses a database management system (such as MySQL or MongoDB) for storage and indexing to obtain a post-evaluation data dataset of the fields to be evaluated, which can be used for subsequent calls.

[0045] Step S302: Input the post-validation data dataset into the pre-built land arability level prediction model so that the land arability level prediction model can calculate the probability distribution and confidence interval of the land arability of the field to be evaluated. The land arability level prediction model is constructed according to the land arability prediction model construction method in the above embodiment.

[0046] For example, in this embodiment of the application, the post-validation data dataset is input into a pre-constructed land arability grade prediction model to update the information of the root node, calculate the posterior probability distribution of the target node "field arability", and obtain the probability distribution of the arability grade of the field to be evaluated (e.g., highly arable probability = 0.7, moderately arable probability = 0.2, unarable probability = 0.1) and confidence interval (e.g., 95% confidence interval). This achieves the fusion of multi-source heterogeneous data and uncertainty, and provides robust and interpretable evaluation results through probabilistic inference, solving the problem that traditional deterministic models cannot handle incomplete data input.

[0047] Step S303: Determine the arability assessment result of the field to be assessed based on the probability distribution of land arability and the confidence interval of the field to be assessed.

[0048] For example, in this embodiment of the application, the arable level of the field with the maximum probability is determined according to the probability distribution of land arableness, and the arable level and probability of the field are used as the arableness assessment result of the field to be evaluated.

[0049] The land arableness assessment method provided in this embodiment uses a Bayesian network-based land arableness level prediction model pre-constructed from standardized macro-grid data. It transforms farmers' empirical assessment index values ​​based on actual field conditions into a post-validation data dataset input to the model. Through model inference, it obtains the field-specific arableness probability distribution and confidence interval, thereby determining the assessment result. This method overcomes the limitation of existing macro-evaluation techniques that cannot directly guide farmers' micro-level arable land development. It effectively integrates macro-regional prior knowledge with farmers' micro-level field experience, fully considering the heterogeneity characteristics at the field level, ensuring the assessment results accurately match farmers' actual production needs. Simultaneously, leveraging the characteristics of the Bayesian network model, it adapts to the reality of inaccurate and incomplete farmer experience data, and can quantify the reliability of the assessment results through confidence intervals. Compared to the singular macro-evaluation conclusions output by related technologies, the assessment results of this invention are more targeted and referential, directly guiding farmers in making decisions regarding the development of reserve arable land resources at the field level. This allows land arableness assessment to truly move from the macro-level of serving regional planning to the micro-level of serving farmers' production.

[0050] This embodiment provides a method for assessing land suitability, which can be used in the aforementioned server. Figure 4This is a flowchart of a land suitability assessment method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: Step S401: Obtain the post-validation data set for the field to be evaluated. This post-validation data set is determined based on the field experience dataset for the field to be evaluated. The field experience dataset is used to characterize the empirical values ​​of multiple evaluation indicators for the field to be evaluated. For details, please refer to [link to relevant documentation]. Figure 3 Step S301 of the illustrated embodiment will not be described again here.

[0051] Step S402: Input the post-validation data dataset into the pre-built land arability grade prediction model so that the land arability grade prediction model can calculate the probability distribution and confidence interval of the land arability of the plot to be evaluated. The land arability grade prediction model is constructed according to the land arability prediction model construction method of the above embodiment. For details, please refer to... Figure 3 Step S302 of the illustrated embodiment will not be described again here.

[0052] Step S403: Determine the arability assessment result of the field to be assessed based on the probability distribution of land arability and the confidence interval. For details, please refer to [link to relevant documentation]. Figure 3 Step S303 of the illustrated embodiment will not be described again here.

[0053] Step S404: If the arability assessment results of the field to be assessed determine that the arability level of the field to be assessed is lower than the preset level, obtain the land arability probability distribution of each assessment dimension. The land arability probability distribution of each assessment dimension is calculated using the land arability level prediction model.

[0054] For example, in this embodiment of the application, if the farmland's arability level is highly suitable with the highest probability, then the farmland's arability level label is directly output; if the farmland's arability level is "relatively suitable" or "unsuitable", the probability distribution of land arability across four evaluation dimensions—topography, soil, climate, and location—is extracted (e.g., low soil arability level).

[0055] Step S405: Determine the suitability level of the corresponding assessment dimension based on the probability distribution of land arability for each assessment dimension.

[0056] For example, in the embodiments of this application, the land arability probability distribution of each assessment dimension determines the suitability level of the corresponding dimension.

[0057] Step S406: Generate structured management text based on the suitability level of each assessment dimension.

[0058] For example, in this application embodiment, text suggestions are generated based on the suitability levels of four assessment dimensions: topography, soil, climate, and location (e.g., low soil arability level). These suggestions include "This field is restricted by soil conditions and has a 60% probability of being unsuitable for cultivation."

[0059] Step S407: The structured management text of the field to be evaluated and the arability evaluation results of the field to be evaluated are sent to the display terminal for display.

[0060] For example, the display terminal can be a terminal device capable of displaying functions. This application embodiment does not limit the specific content of the display terminal, as long as it is reasonable. In this application embodiment, a visualization library (such as ECharts or D3.js) is used to present text reports on the terminal device interface (such as a mobile app or web page). The interface design follows the principle of farmer friendliness and uses large fonts.

[0061] The following specific embodiment illustrates the construction of the land arability prediction model and the land arability assessment method provided in this application.

[0062] Example: To more intuitively illustrate the practical application effects of the method of this invention, the following example, using a specific case from City A, demonstrates the complete implementation process of constructing a land arability prediction model and using a land arability assessment method in the micro-evaluation at the field level. This example uses a reserve arable land field in City A as the object, demonstrating the entire process from data input to result output.

[0063] (1) Application environment description: Geographical location: City A, 92°47.624′E, 28°52.232′N, altitude 3734 meters; Field characteristics: This field is a scattered reserve farmland in the mountainous area, which the farmers plan to develop for spring barley cultivation. The region has a plateau temperate semi-arid climate, with sandy loam as the main soil type and significant topographic relief. Background requirement: Farmers need to assess the arability level of a field to decide whether to cultivate it and optimize management practices. However, farmers can only provide some empirical indicators and lack complete data support.

[0064] (2) Application process: ① Phase 1: Prior Knowledge Integration. Raster data of regional land arability assessment results are obtained from the "Scientific Data Center," along with raster data of regional land arability assessment indicators, including topography, climate, soil, and location indicators. A standardized raster dataset is generated through coordinate system unification (conversion to WGS84) and spatial scale unification (bilinear interpolation downscaling to 10m resolution). Based on a land suitability index system (4 categories and 13 indicators, including topography, climate, soil, and location), a Bayesian network structure is constructed. Prior probability vectors are learned using the expectation-maximization algorithm (e.g., 0.2% prior probability for highly arable land, 2.4% for moderately arable land, 67.4% for barely arable land, and 30.0% for unsuitable land). The resulting structure is as follows: Figure 5 The model shown is a land arability rating prediction model based on Bayesian networks.

[0065] Phase 2: Posterior knowledge input. Farmer data input includes: Topographical indicators: slope range 2°~5°, slope direction northeast, elevation 3734 meters; Soil parameters: gravel content approximately 10%, soil texture predominantly sandy (qualitative description); Location indicators: 10-20 minutes walking time to the nearest water source, 5-10 minutes walking time to the settlement, and 5-10 minutes walking time to the nearest road; Other indicators: Climate indicators (such as precipitation and temperature) and soil chemical indicators (such as pH value) are unknown and are supplemented by macroscopic prior knowledge; Through terminal verification and standardization, the data is uploaded to the cloud server to generate a structured post-verification dataset.

[0066] ③ Stage 3: Bayesian Network Inference. The posterior validation evidence input by farmers is fused with the prior probability matrix, and the posterior probability distribution of the field's arability level is calculated using Bayesian network inference. Specifically, a Bayesian network-based land arability level prediction model is used to update the root node information and calculate the probability of the leaf node "field arability".

[0067] Phase 4: Visualizing the results.

[0068] The probability distribution is mapped to suitability levels, and management recommendations are generated and presented in text and chart form through the terminal interface, specifically including: Probabilistic Output: Bayesian network inference results show that the probability distribution of the land's arability level is as follows: barely arable: 66%, moderately arable: 30%, highly arable: 4%. The model also outputs a 95% confidence interval, indicating robustness. The evaluation results of the land arability level prediction model based on Bayesian networks are as follows: Figure 6 As shown.

[0069] Grade determination: Based on the principle of maximum probability, the arability grade of the field is determined to be "barely arable", and the output label is "66% probability of barely arable".

[0070] Management recommendations generated: Due to the non-high suitability level, the model automatically analyzes limiting factors: low climate suitability level (probability 57.3% barely arable, 29.5% moderately arable), low terrain suitability level (probability 60.4% moderately arable), and low soil suitability level (probability 42.3% moderately arable, 29.5% barely arable).

[0071] The generated text suggests: "Due to climate and location limitations, the farmland's suitability is rated as 'barely suitable for cultivation'."

[0072] This embodiment also provides a land arability prediction model construction device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as already described. As used below, the term "module" can be a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0073] This embodiment provides a device for constructing a land suitability prediction model, such as... Figure 7 As shown, it includes: The first acquisition module 701 is used to acquire raster data of the target land arability evaluation results, target evaluation index raster data, and land arability index system of the target area. The land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators. Module 702 is used to construct a Bayesian network structure based on the land arability index system. In the Bayesian network structure, each root node is a multiple evaluation index in the land arability index system, the middle nodes are evaluation dimensions, and the leaf nodes are the land arability evaluation results. The first determining module 703 is used to convert the raster data of the target land arability evaluation results and the target evaluation index raster data into a sample data table, input the sample data table into a Bayesian network structure for parameter learning, and obtain a land arability prediction model.

[0074] In some optional implementations, the raster data of the target land arability evaluation results and the target evaluation index raster data are obtained through the following steps: Obtain raster data of land suitability assessment results and assessment index raster data for the target area; The raster data of land arability evaluation results and the raster data of evaluation indicators are processed by coordinate unification to obtain the processed raster data of land arability evaluation results and the raster data of evaluation indicators. Spatial scale unification processing is performed on the processed raster data of land arability evaluation results and evaluation indexes to obtain raster data of target land arability evaluation results and target evaluation indexes.

[0075] This embodiment also provides a land suitability assessment device, such as... Figure 8 As shown, the device includes: The second acquisition module 801 is used to acquire the post-verification data set of the field to be evaluated. The post-verification data set is determined based on the field experience data set of the field to be evaluated. The field experience data set is used to characterize the empirical values ​​of multiple evaluation indicators of the field to be evaluated. The calculation module 802 is used to input the post-validation data dataset into the pre-built land arability level prediction model so that the land arability level prediction model can calculate the land arability probability distribution and confidence interval of the field to be evaluated. The land arability level prediction model is constructed according to the land arability prediction model construction method of the above embodiment. The second determining module 803 is used to determine the arability assessment result of the field to be assessed based on the probability distribution of land arability and the confidence interval of the field to be assessed.

[0076] In some alternative embodiments, the above-described apparatus further includes: The third acquisition module is used to acquire the land arability probability distribution of each assessment dimension if the arability level of the field to be assessed is determined to be lower than the preset level based on the arability assessment results of the field to be assessed. The land arability probability distribution of each assessment dimension is calculated using the land arability level prediction model. The third determination module is used to determine the suitability level of the corresponding assessment dimension based on the probability distribution of land arability of each assessment dimension; The generation module is used to generate structured management texts based on the suitability levels of each assessment dimension.

[0077] In some alternative embodiments, the above-described apparatus further includes: The display module is used to send the structured management text of the field to be evaluated and the arability evaluation results of the field to be evaluated to the display terminal for display.

[0078] The land arability prediction model construction device provided in this embodiment of the invention can execute the land arability prediction model construction method provided in any embodiment of the invention, and the land arability assessment device provided in this embodiment of the invention can execute the land arability assessment method provided in any embodiment of the invention, possessing the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0079] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0080] The following is a detailed reference. Figure 9This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 901, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 902 or a program loaded from memory 908 into random access memory (RAM) 903. The RAM 903 also stores various programs and data required for the operation of the electronic device. The processor 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0081] Typically, the following devices can be connected to I / O interface 905: input devices 906 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 907 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 908 including, for example, magnetic tapes, hard disks, etc.; and communication devices 909. Communication device 909 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 9 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0082] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 909, or installed from a memory 908, or installed from a ROM 902. When the computer program is executed by the processor 901, it performs the functions defined in the land arability prediction model construction method of the embodiments of the present invention, or performs the functions defined in the land arability assessment method.

[0083] Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0084] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the land arability prediction model construction method shown in the above embodiments is implemented, or the land arability assessment method shown in the above embodiments is implemented.

[0085] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0086] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for constructing a land arability prediction model, characterized in that, The method includes: The target area is obtained by acquiring raster data of the target land arability evaluation results, target evaluation index raster data, and land arability index system. The land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators. A Bayesian network structure is constructed based on the land arability index system. In the Bayesian network structure, each root node is a multiple evaluation index in the land arability index system, the middle nodes are evaluation dimensions, and the leaf nodes are the land arability evaluation results. The raster data of the target land arability evaluation results and the target evaluation index are converted into a sample data table. The sample data table is then input into a Bayesian network structure for parameter learning to obtain a land arability prediction model.

2. The method according to claim 1, characterized in that, The raster data of the target land suitability evaluation results and the target evaluation index raster data are obtained through the following steps: Obtain raster data of land suitability assessment results and assessment index raster data for the target area; The raster data of the land arability evaluation results and the raster data of the evaluation indicators are subjected to coordinate unification processing to obtain the processed raster data of the land arability evaluation results and the raster data of the evaluation indicators. The processed land arability evaluation result raster data and evaluation index raster data are subjected to spatial scale unification processing to obtain target land arability evaluation result raster data and target evaluation index raster data.

3. A method for assessing land suitability, characterized in that, The method includes: Obtain a post-validation data set of the field to be evaluated, which is determined based on the field experience dataset of the field to be evaluated, and the field experience dataset is used to characterize the empirical values ​​of multiple evaluation indicators of the field to be evaluated. The post-validation dataset is input into a pre-built land arability grade prediction model so that the land arability grade prediction model can calculate the probability distribution and confidence interval of the land arability of the field to be evaluated. The land arability grade prediction model is constructed according to the land arability prediction model construction method according to claim 1 or 2. The land suitability assessment results of the plots to be assessed are determined based on the probability distribution of land suitability and the confidence interval.

4. The method according to claim 3, characterized in that, The method further includes: If the arability assessment results of the field to be assessed determine that the arability level of the field to be assessed is lower than the preset level, the probability distribution of land arability for each assessment dimension is obtained. The probability distribution of land arability for each assessment dimension is calculated using a land arability level prediction model. The suitability level for each assessment dimension is determined based on the probability distribution of land arability for each assessment dimension. Structured management texts are generated based on the suitability levels of each assessment dimension.

5. The method according to claim 4, characterized in that, Following the step of generating structured management text based on suitability levels of each assessment dimension, the method further includes: The structured management text of the field to be evaluated, along with the arability assessment results of the field to be evaluated, is sent to the display terminal for display.

6. A device for constructing a land arability prediction model, characterized in that, The device includes: The first acquisition module is used to acquire raster data of the target land arability evaluation results, target evaluation index raster data, and land arability index system of the target area. The land arability index system includes multiple evaluation dimensions, and each evaluation dimension includes multiple evaluation indicators. The construction module is used to construct a Bayesian network structure based on the land arability index system. In the Bayesian network structure, each root node is a multiple evaluation index in the land arability index system, the middle nodes are evaluation dimensions, and the leaf nodes are the land arability evaluation results. The first determining module is used to convert the raster data of the target land arability evaluation results and the target evaluation index raster data into a sample data table, and input the sample data table into a Bayesian network structure for parameter learning to obtain a land arability prediction model.

7. A land suitability assessment device, characterized in that, The device includes: The second acquisition module is used to acquire the post-verification data set of the field to be evaluated. The post-verification data set is determined based on the field experience data set of the field to be evaluated. The field experience data set is used to characterize the empirical values ​​of multiple evaluation indicators of the field to be evaluated. The calculation module is used to input the post-validation data dataset into a pre-constructed land arability level prediction model, so that the land arability level prediction model can calculate the probability distribution and confidence interval of land arability of the field to be evaluated. The land arability level prediction model is constructed according to the land arability prediction model construction method according to claim 1 or 2. The second determination module is used to determine the arability assessment result of the field to be assessed based on the probability distribution of land arability and the confidence interval of the field to be assessed.

8. An electronic device, characterized in that, include: The system includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the land arability prediction model construction method according to any one of claims 1 or 2, or the land arability assessment method according to any one of claims 3 to 5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the land arability prediction model construction method according to any one of claims 1 or 2, or the land arability assessment method according to any one of claims 3 to 5.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the land arability prediction model construction method according to any one of claims 1 or 2, or to execute the land arability assessment method according to any one of claims 3 to 5.

Images