GIS pattern spot-based peasant household portrait generation method and system

Abstract

The invention relates to the technical field of digital agriculture, in particular to a farmer portrait generation method and system based on GIS pattern spots. According to the farmer portrait generation method based on the GIS pattern spots, a farmland plot is locked in real time through a satellite map with the precision of 0.5 m, and farmer production behaviors are collected and analyzed in real time so as to generate a digital farmer portrait. According to the farmer portrait generation method and system based on the GIS pattern spots, decision-making precision can be accurate to a window period of 3 days, meanwhile, early warning of plant diseases and insect pests can be achieved 14 days in advance, the fuel cost of agricultural machinery is reduced, meanwhile, environmental pollution is reduced, supervision efficiency is improved, and great convenience is provided for modern agricultural development.

Description

Technical Field

[0001] The present invention relates to the technical field of digital agriculture, and particularly relates to a method and system for generating farmer portraits based on GIS patches. Background Art

[0002] Traditional methods for generating farmer portraits mainly rely on means such as on-site investigations, expert experience judgments, and simple data analysis. However, with the increasing diversification and complexity of agricultural production, these traditional methods have gradually revealed many deficiencies.

[0003] First, the data collection method is single. The data collection for traditional farmer portraits mostly relies on on-site investigations, such as questionnaire surveys, interviews, etc. Although this method can obtain first-hand information, it is inefficient and greatly restricted by factors such as region and time.

[0004] Second, the analysis method is simple. In data processing and analysis, traditional methods mostly use basic statistical methods such as descriptive statistics and simple regression, and it is difficult to discover the deep-seated laws and relationships behind the data.

[0005] Third, there is a lack of dynamic update. The production conditions, technical levels, market demands, etc. of farmers are dynamically changing, but traditional portrait methods often have difficulty reflecting these changes in a timely manner, resulting in lagging or inaccurate portrait results.

[0006] Fourth, the personalization is insufficient. Due to the limitations of technology and methods, traditional farmer portraits can often only provide relatively rough descriptions of group characteristics and are difficult to meet the needs of precise and personalized services.

[0007] Fifth, there are accuracy issues. Due to the limitations of data collection and analysis methods, traditional farmer portraits may not accurately reflect the true characteristics and needs of farmers, thus affecting the effectiveness of decision-making.

[0008] Sixth, there are timeliness issues: Agricultural production has seasonality and periodicity, and the production and operation conditions of farmers are also constantly changing. Traditional methods are difficult to update portrait information in real time, resulting in outdated or invalid portrait results.

[0009] Seventh, there are comprehensiveness issues. Traditional methods often focus on collecting and analyzing the economic data of farmers while ignoring non-economic factors such as the social culture, psychological behavior, etc. of farmers, resulting in one-sided or incomplete portrait results.

[0010] Eighth, there are application issues. Due to the accuracy and timeliness issues of traditional farmer portraits, their actual application effects are often unsatisfactory and it is difficult to provide strong support for agricultural production, rural development, etc.

[0011] In summary, the traditional methods for generating farmer portraits have many deficiencies in aspects such as data collection, analysis methods, dynamic updates, and personalized services. These problems restrict their application effects in modern agricultural production management. Therefore, it is necessary to introduce more advanced technologies and methods to improve and perfect the process of generating farmer portraits.

[0012] Aiming at problems in the prior art, such as the traditional farmer portraits relying on questionnaire data (precision attenuation rate > 40%), lacking spatial data support, insufficient deep association application of GIS and farmer behavior, and technical bottlenecks in the fusion of multi-source heterogeneous data (spatio-temporal alignment error > 15%), the present invention proposes a method and system for generating farmer portraits based on GIS patches. Summary of the Invention

[0013] In order to make up for the defects of the prior art, the present invention provides a simple and efficient method and system for generating farmer portraits based on GIS patches.

[0014] The present invention is realized through the following technical solutions:

[0015] A method for generating farmer portraits based on GIS patches, characterized in that: real-time locking of farmland plots through satellite maps with a precision of 0.5 meters, real-time collection and analysis of farmers' production behaviors to generate digital farmer portraits;

[0016] Including the following steps:

[0017] Step S1, real-time collection of 8 types of data sources, namely satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy strategy data, market price and meteorological disaster warning data;

[0018] Among them, the satellite image data is satellite remote sensing plot boundary data with a precision of 0.5 meters;

[0019] Step S2, establishing a spatio-temporal mapping relationship between GIS patches and farmer behaviors;

[0020] Dynamically updating the growth status of crops through time-series remote sensing data, and establishing a mapping relationship with farmers' production record (fertilization, irrigation) data;

[0021] Step S3, developing a multi-modal data dynamic fusion algorithm (error rate < 5%);

[0022] Step S3.1, data preprocessing and feature mapping

[0023] Using drones or handheld devices to correct the GIS patch boundaries, eliminating geometric distortions caused by terrain undulations, then uniformly processing the data through coordinate transformation, and then custom extracting key information;

[0024] The key information includes the area and shape of the plot, the spectral characteristics of the crops, and the planting inputs and outputs of the crops.

[0025] Step S3.2, Spatiotemporal Feature Decoupling and Fusion

[0026] Use the Delaunay algorithm to cut the plot into several triangular meshes, then use the confidence evaluation matrix to adaptively weight the spatial and temporal feature information, and introduce a time decay function to improve the accuracy of the data.

[0027] Step S4, Farmer Portrait Modeling and Optimization

[0028] Analyze the impact of soil type and climate on planting decisions through the Geographically Weighted Regression (GWR) model, extract multimodal features, and generate a digital farmer portrait that includes five dimensions: basic geographical information dimension, agricultural production dimension, economic and social dimension, environment and ecology dimension, and dynamic time series dimension.

[0029] Step S5, Output Application Decision

[0030] Step S5.1, Combine the real-time collected information, and generate planting suggestions, pest and disease warnings, and agricultural machinery path optimization plans based on the Geographically Weighted Regression (GWR) model; among them, the planting suggestions are accurate to a 3-day window period, and the pest and disease warnings predict pests and diseases at least 14 days in advance.

[0031] Step S5.2, Display the association relationship between the plot and the farmer through a visualization map, mark the inefficient operation areas (detour rate > 15%), and push the generated planting suggestions and pest and disease warnings to the corresponding farmers via text message or network.

[0032] In the said step S3.2, after cutting the plot into several triangular meshes, use the spatiotemporal attention module and the time decay model to dynamically calibrate the data of different timestamps and align the data.

[0033] Using the confidence evaluation matrix to adaptively weight the spatial and temporal feature information includes the following steps:

[0034] Step S3.2.1, Data Source Quality Assessment

[0035] First, define 5-level quality indicators, including integrity, accuracy, timeliness, consistency, and relevance. Use data probes to detect the quality status of 8 types of data sources, and then calculate the index weights using the entropy weight method.

[0036] Step S3.2.2, Cross-Source Credibility Calculation

[0037] Customize and establish a data source conflict check and verification mechanism. When the data source conflict check and verification mechanism is triggered, update the confidence level through a Bayesian network and dynamically adjust the weights of various source data in the matrix;

[0038] Step S3.2.3, Matrix visualization construction

[0039] First, generate an N*N symmetric matrix and mark the custom-selected abnormal data sources in red.

[0040] In the said step S4, the basic geographic information dimension includes plot area and shape index, elevation and slope, adjacency relationship, infrastructure accessibility, soil type and texture, and tillage layer thickness;

[0041] The agricultural production dimension includes planting structure, crop yield and output value, disaster risk index, irrigation coverage rate, fertilization intensity, mechanization level, rotation cycle, and biodiversity;

[0042] The economic and social dimension includes the composition of farmers' income, farmers' debt ratio, market access index, social security coverage, labor quality, consumption structure, and farmers' risk resistance ability;

[0043] The environment and ecology dimension includes the average NDVI value, soil organic matter content, water pollution index, and carbon sequestration capacity;

[0044] The dynamic time series dimension includes the sensitivity of market price fluctuations, climate response coefficient, and lag period of technology adoption.

[0045] In the said step S5, when generating planting suggestions, the real-time monitoring of the traveling speed and operation coverage rate of agricultural machinery is carried out through a positioning system, and in combination with the soil compaction sensing data, the areas of missed tillage, repeated tillage, or abnormal operation are identified, and whether there are continuous operation obstacles or operation deviations caused by mechanical wear is judged by comparing the historical operation trajectory with the current operation mode;

[0046] Then, the agricultural machinery trajectory is superimposed and analyzed with the satellite image to verify the matching degree of the cultivated land boundary. If an uncovered area is found, a replanting suggestion is generated.

[0047] A farmer portrait generation system based on GIS patches for implementing the above method, including a data acquisition module, a multi-source data fusion module, a farmer portrait generation module, and a decision engine module;

[0048] The data acquisition module is used to collect satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy policy data, market price, and meteorological disaster warning data in real time;

[0049] Among them, the satellite image data is satellite remote sensing plot boundary data with a precision of 0.5 meters;

[0050] A multi-source data fusion module, which is used to integrate 8 types of data sources, including satellite images, agricultural machinery trajectories, soil sensors, crop growth monitoring data, farmer behavior data, policy subsidy data, market prices, and meteorological disaster warnings;

[0051] A farmer portrait generation module, which is used to real-time associate farmer behavior with plot topological features, analyze the impact of soil type and climate on planting decisions using the Geographically Weighted Regression (GWR) model, and generate a digital farmer portrait containing five dimensions;

[0052] A decision engine module, which is used to generate suggestions for planting windows, pest and disease warnings, and optimization schemes for agricultural machinery paths.

[0053] A device for generating a farmer portrait based on GIS patches, characterized in that it includes a memory and a processor; the memory is used to store computer programs, and the processor is used to implement the method steps as described above when executing the computer programs.

[0054] A readable storage medium, characterized in that a computer program is stored on the readable storage medium, and the computer program implements the method steps as described above when executed by a processor.

[0055] The beneficial effects of the present invention are as follows: The method and system for generating a farmer portrait based on GIS patches integrate 8 types of data sources simultaneously and in real-time dynamic update in terms of data dimensions, can be accurate to a 3-day window period in terms of decision-making accuracy, can issue pest and disease warnings 14 days in advance, reduce the fuel cost of agricultural machinery, reduce environmental pollution at the same time, improve the supervision efficiency, and provide great convenience for the development of modern agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in 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 some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0057] Attached Figure 1 It is a schematic diagram of the method for generating a farmer portrait based on GIS patches of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] To enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work shall fall within the protection scope of the present invention.

[0059] The method for generating a farmer portrait based on GIS patches locks the farmland plots in real time through satellite maps with a precision of 0.5 meters, and collects and analyzes the production behaviors of farmers in real time to generate a digital farmer portrait;

[0060] It includes the following steps:

[0061] Step S1: Collect 8 types of data sources in real time, namely satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy strategy data, market price and meteorological disaster warning data;

[0062] Among them, the satellite image data is satellite remote sensing plot boundary data with a precision of 0.5 meters;

[0063] Step S2: Establish a spatio-temporal mapping relationship between GIS patches and farmer behaviors;

[0064] Dynamically update the growth status of crops through time-series remote sensing data, and establish a mapping relationship with farmer production record (fertilization, irrigation) data;

[0065] Step S3: Develop a multi-modal data dynamic fusion algorithm (error rate < 5%);

[0066] Step S3.1: Data preprocessing and feature mapping

[0067] Use drones or handheld devices to correct the GIS patch boundaries, eliminate geometric distortions caused by terrain undulations, then uniformly process the data using coordinate transformation, and then custom extract key information;

[0068] The key information includes the area, shape, spectral characteristics of crops, planting inputs and outputs of plots.

[0069] Step S3.2: Spatio-temporal feature decoupling and fusion

[0070] Use the Delaunay algorithm to cut the plot into several triangular meshes, then use the confidence evaluation matrix to adaptively weight the spatial and time features, and introduce a time decay function to improve the accuracy of the data;

[0071] Step S4: Farmer portrait modeling and optimization

[0072] Analyze the impact of soil type and climate on planting decisions through the Geographically Weighted Regression (GWR) model, extract multi-modal features, and generate a digital farmer portrait that includes five dimensions: the basic geographical information dimension, the agricultural production dimension, the economic and social dimension, the environmental and ecological dimension, and the dynamic time series dimension.

[0073] Step S5: Output application decisions

[0074] Step S5.1: Combine the real-time collected information, and generate planting suggestions, pest and disease warnings, and optimized agricultural machinery path plans based on the Geographically Weighted Regression (GWR) model; among them, the planting suggestions are accurate to a 3-day window period, and the pest and disease warnings predict pests and diseases at least 14 days in advance.

[0075] Step S5.2: Display the association relationship between plots and farmers through a visualization map, mark the inefficient operation areas (detour rate > 15%), and push the generated planting suggestions and pest and disease warnings to the corresponding farmers via text messages or the Internet.

[0076] In the said step S3.2, after cutting the plot into several triangular grids, use the spatio-temporal attention module and the time decay model to dynamically calibrate the data with different timestamps and align the data.

[0077] Adaptive weight information of spatial features and time features using a confidence evaluation matrix includes the following steps:

[0078] Step S3.2.1: Data source quality assessment

[0079] First, define 5-level quality indicators, including integrity, accuracy, timeliness, consistency, and relevance. Use data probes to detect the quality status of 8 types of data sources, and then calculate the indicator weights using the entropy weight method.

[0080] Step S3.2.2: Cross-source credibility calculation

[0081] First, customize and establish a data source conflict check and verification mechanism. For example, when the difference between the rainfall of the weather station and the satellite data > 30%, trigger the verification process; when the data source conflict check and verification mechanism is triggered, update the confidence through the Bayesian network and dynamically adjust the weights of various source data in the matrix.

[0082] Step S3.2.3: Matrix visualization construction

[0083] First, generate an N*N symmetric matrix, and mark the custom-selected abnormal data sources in red. For example, when the matching rate of the agricultural machinery trajectory data and the satellite image cultivated land range < 70%, trigger an alarm.

[0084] The application of the evaluation matrix is mainly reflected in the dynamic weight allocation in multi-modal dynamic fusion. For example, the weight of high-confidence data is increased to 0.8, and the weight of low-confidence data (manually filled data) is reduced to 0.2.

[0085] In step S4, the basic geographic information dimensions include plot area and shape index, elevation and slope, adjacency relationship, infrastructure accessibility, soil type and texture, and plough layer thickness.

[0086] The agricultural production dimensions include planting structure, crop yield and output value, disaster risk index, irrigation coverage rate, fertilization intensity, level of mechanization, rotation cycle, and biodiversity.

[0087] The economic and social dimensions include the composition of farmers' income, farmers' debt ratio, market access index, social security coverage, labor quality, consumption structure, and farmers' risk resistance ability.

[0088] The environmental and ecological dimensions include the average NDVI value, soil organic matter content, water pollution index, and carbon sequestration capacity.

[0089] The dynamic time series dimensions include the sensitivity of market price fluctuations, climate response coefficient, and lag period of technology adoption.

[0090] In step S5, when generating planting suggestions, the traveling speed and operation coverage rate of agricultural machinery are monitored in real time through a positioning system. Combining with the soil compaction sensing data, areas of missed tillage, repeated tillage, or abnormal operation are identified. By comparing the historical operation trajectory with the current operation mode, it is judged whether there are continuous operation obstacles or operation deviations caused by mechanical wear.

[0091] Then, the agricultural machinery trajectory is superimposed and analyzed with the satellite image to verify the matching degree of the cultivated land boundary. If an uncovered area is found, a replanting suggestion is generated. For example: when the trajectory coverage rate < 95%, it is recommended to replant the missed tillage area. When the operation efficiency decreases (speed < 20%), it is recommended to maintain the agricultural machinery.

[0092] The farmer portrait generation system based on GIS patches is used to implement the above method. The technical architecture includes a data input layer, an intelligent analysis layer, and a decision support layer. Among them, the data input layer is equipped with a data collection module, the intelligent analysis layer is equipped with a multi-source data fusion module and a farmer portrait generation module, and the decision support layer is equipped with a decision engine module.

[0093] The data collection module is used to collect satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy policy data, market price, and meteorological disaster warning data in real time.

[0094] Among them, the satellite image data is satellite remote sensing plot boundary data (GIS patches) with a precision of 0.5 meters.

[0095] The multi-source data fusion module integrates 8 types of data sources through spatial modeling and data alignment means, including satellite images, agricultural machinery trajectories, soil sensors, crop growth monitoring data, farmer behavior data, policy subsidy data, market prices, and meteorological disaster warnings.

[0096] The farmer portrait generation module is used to real-time associate farmer behavior with plot topological features (response delay ≤ 5 minutes), analyze the impact of soil type and climate on planting decisions using the Geographically Weighted Regression model GWR, extract multi-modal features, and generate a digital farmer portrait containing five dimensions.

[0097] The decision engine module is used to generate suggestions for the planting window period (error ± 3 days), pest and disease warnings (14 days in advance), and an optimized agricultural machinery path plan (fuel consumption reduced by 23%).

[0098] The farmer portrait generation device based on GIS patches includes a memory and a processor; the memory is used to store computer programs, and the processor is used to implement the above method steps when executing the computer programs.

[0099] The readable storage medium stores a computer program, and the computer program implements the above method steps when executed by a processor.

[0100] The farmer portrait generation method and system based on GIS patches lock farmland plots through satellite maps (0.5-meter accuracy), integrate 8 types of data such as agricultural machinery operations, soil monitoring, and market conditions, and analyze farmer production behavior in real time; it can accurately match farmers' operations such as tilling and fertilizing to the corresponding plots within 5 minutes; warn of pests and diseases 14 days in advance and recommend the best sowing / fertilizing time (error not exceeding 3 days).

[0101] Compared with traditional methods, this technology upgrades agricultural decision-making from relying on empirical estimates to real-time data-driven, increases the per-acre yield by 12% - 15%, improves the supervision efficiency by more than 3 times, and reduces environmental pollution at the same time.

[0102] The above-described embodiments are only one of the specific implementation manners of the present invention, and the general changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a farmer portrait based on GIS patches, characterized in that: Real-time lock farmland plots through satellite maps with 0.5-meter accuracy, collect and analyze farmers' production behaviors in real time to generate digital farmer portraits; It includes the following steps: Step S1: Real-time collect 8 types of data sources, namely satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy strategy data, market price and meteorological disaster warning data; Among them, the satellite image data is satellite remote sensing plot boundary data with 0.5-meter accuracy; Step S2: Establish the spatio-temporal mapping relationship between GIS patches and farmer behaviors; Dynamically update the growth status of crops through time-series remote sensing data, and develop a multi-modal data dynamic fusion algorithm with farmers' production records. Step S3.1: Data preprocessing and feature mapping Use drones or handheld devices to correct the GIS patch boundaries, eliminate geometric distortions caused by terrain undulations, then uniformly process the data using coordinate transformation, and then custom extract key information; Step S3.2: Spatio-temporal feature decoupling and fusion Use the Delaunay algorithm to cut the plot into several triangular meshes, then use the confidence evaluation matrix to adaptively weight the spatial and temporal features, and introduce a time decay function to improve the accuracy of the data; Step S4: Farmer portrait modeling and optimization Analyze the impact of soil type and climate on planting decisions through the Geographically Weighted Regression model GWR, extract multi-modal features, and generate a digital farmer portrait including five dimensions: basic geographic information dimension, agricultural production dimension, economic and social dimension, environment and ecology dimension, and dynamic time series dimension; Step S5: Output application decisions Step S5.1: Combine the information collected in real time, and generate planting suggestions, pest and disease warnings, and agricultural machinery path optimization plans based on the Geographically Weighted Regression model GWR; among them, the planting suggestions are accurate to a 3-day window period, and the pest and disease warnings predict pests and diseases at least 14 days in advance; Step S5.2: Display the association relationship between plots and farmers through a visualization map, mark inefficient operation areas, and push the generated planting suggestions and pest and disease warnings to the corresponding farmers via text messages or the Internet.

2. The method for generating a farmer portrait based on GIS patches according to claim 1, wherein: The key information includes the area and shape of the plot, the spectral characteristics of the crops, the planting inputs and outputs of the crops.

3. The method for generating a farmer portrait based on GIS patches according to claim 1, wherein: In step S3.2, after cutting the plot into several triangular meshes, use the spatio-temporal attention module and the time decay model to dynamically calibrate the data with different timestamps and align the data.

4. The method for generating a farmer portrait based on GIS patches according to claim 1, characterized in that: In step S3.2, using the confidence evaluation matrix to adaptively weight the spatial and temporal features includes the following steps: Step S3.2.1: Data source quality assessment First, define 5-level quality indicators, including integrity, accuracy, timeliness, consistency, and relevance. Use data probes to detect the quality status of 8 types of data sources, and then use the entropy weight method to calculate the index weights; Step S3.2.2: Cross-source credibility calculation Customize and establish a data source conflict check and verification mechanism. When the data source conflict check and verification mechanism is triggered, update the confidence through a Bayesian network and dynamically adjust the weights of various source data in the matrix; Step S3.2.3: Matrix visualization construction First, generate an N*N symmetric matrix and mark the custom-selected abnormal data sources in red.

5. The method for generating a farmer portrait based on GIS patches according to claim 1, wherein: In step S4, the basic geographic information dimensions include plot area and shape index, elevation and slope, adjacency relationship, infrastructure accessibility, soil type and texture, and tillage layer thickness; The agricultural production dimensions include planting structure, crop yield and output value, disaster risk index, irrigation coverage rate, fertilization intensity, mechanization level, rotation cycle, and biodiversity; The economic and social dimensions include the composition of farmers' income, farmers' debt ratio, market access index, social security coverage, labor quality, consumption structure, and farmers' risk resistance ability; The environmental and ecological dimensions include the mean NDVI, soil organic matter content, water pollution index, and carbon sequestration capacity; The dynamic time series dimensions include the sensitivity of market price fluctuations, climate response coefficient, and lag period of technology adoption.

6. The method for generating a farmer portrait based on GIS patches according to claim 1, characterized in that: In step S5, when generating planting suggestions, the real-time monitoring of the traveling speed and operation coverage rate of agricultural machinery is carried out through a positioning system. Combining the soil compaction sensing data, identify areas with missed tillage, repeated tillage, or abnormal operations, and compare the historical operation trajectory with the current operation mode to determine whether there are continuous operation obstacles or operation deviations caused by mechanical wear; Then, overlay and analyze the agricultural machinery trajectory with the satellite image to verify the matching degree of the cultivated land boundary. If an uncovered area is found, a reseeding suggestion is generated.

7. A farmer portrait generation system based on GIS patches, characterized in that: Used to implement the method described in any one of claims 1 to 6, including a data acquisition module, a multi-source data fusion module, a farmer portrait generation module, and a decision engine module; The data acquisition module is used to collect satellite image data, agricultural machinery trajectory data, soil sensor data, crop growth monitoring data, farmer behavior data, subsidy policy data, market price, and meteorological disaster warning data in real time; Among them, the satellite image data is satellite remote sensing plot boundary data with a precision of 0.5 meters; The multi-source data fusion module is used to integrate 8 types of data sources, including satellite images, agricultural machinery trajectories, soil sensors, crop growth monitoring data, farmer behavior data, policy subsidy data, market price, and meteorological disaster warnings; The farmer portrait generation module is used to real-time associate farmer behavior with plot topological features, and use the geographically weighted regression model GWR to analyze the influence of soil type and climate on planting decisions, and generate a digital farmer portrait containing five dimensions; The decision engine module is used to generate planting window period suggestions, pest and disease warnings, and agricultural machinery path optimization plans.

8. An equipment for generating a portrait of a rural household based on GIS patches, characterized in that: Includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method described in any one of claims 1 to 6 when executing the computer program.

9. A readable storage medium, characterized in that: A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, it implements the method described in any one of claims 1 to 6.

Images