Rural land planning and construction method fusing multi-source data

Abstract

The invention discloses a rural land planning and construction method fusing multi-source data, and the method comprises the steps: obtaining rural land planning data and a building distribution diagram, and obtaining a cultivated land region and a population gathering region; calculating the distance between the cultivated land area and the population gathering area, distributing a nearest cultivated land area for each population gathering area according to the distance, and outputting the nearest cultivated land area as a cultivated land area to be planned; obtaining remote sensing image data of a to-be-planned cultivated land area, extracting farmland and ridges based on the remote sensing image data, and constructing a shortest path from a population gathering area to each farmland through the ridges; physical parameters and micro-ecological parameters of the ridge are obtained, a labor-saving index of the shortest path is calculated based on the physical parameters, and a planting assistance index of the ridge on the farmland side is calculated based on the micro-ecological parameters; based on the labor-saving index and the planting assistance index, combining with the peasant household body function data of the population gathering area, dynamically distributing the farmland; adaptive farmlands can be recommended for planting according to different body functions of farmers, and reasonable resource allocation is realized.

Description

Technical Field

[0001] This invention relates to the field of land planning technology, and in particular to a method for rural land planning and construction that integrates multi-source data. Background Technology

[0002] Existing rural land planning methods suffer from several problems, including a single data source, a one-sided planning dimension, and insufficient dynamic adaptability. They rely heavily on static data such as remote sensing imagery and land use records, neglecting dynamic information such as farmer behavior and micro-ecological changes. They focus on macro-level aspects like farmland protection and infrastructure layout, paying insufficient attention to specific scenarios such as age-friendly working spaces and the micro-ecological functions of field ridges and ditches. Planning schemes are mostly static outputs, failing to respond to dynamic needs such as population structure changes (e.g., aging) and fluctuations in ecological benefits. Furthermore, data silos are significant, with ineffective integration of agricultural, ecological, and social data, resulting in poor alignment between planning schemes and actual production needs and ecological protection requirements. This easily leads to problems such as land abandonment, resource waste, and ecological degradation, hindering the sustainable use of rural land. Summary of the Invention

[0003] In view of this, the present invention proposes a rural land planning and construction method that integrates multi-source data, which can recommend suitable farmland for planting based on the different physical functions of farmers.

[0004] The technical solution of this invention is implemented as follows:

[0005] A method for rural land planning and construction that integrates multi-source data includes the following steps:

[0006] Step S1: Obtain rural land planning data and building distribution map; determine cultivated land areas based on land planning data; and determine population concentration areas based on building distribution map.

[0007] Step S2: Calculate the distance between cultivated land areas and densely populated areas, assign the nearest cultivated land area to each densely populated area based on the distance, and output it as the cultivated land area to be planned.

[0008] Step S3: Obtain remote sensing image data of the cultivated area to be planned, extract farmland and field ridges based on the remote sensing image data, and construct the shortest path from the population gathering area to each farmland via the field ridges;

[0009] Step S4: Obtain the physical parameters and micro-ecological parameters of the field ridges, calculate the labor-saving index of the shortest path based on the physical parameters, and calculate the planting assistance index of the field ridges on the farmland side based on the micro-ecological parameters.

[0010] Step S5: Dynamically allocate farmland based on labor-saving indicators and planting assistance indicators combined with farmers' physical function data in densely populated areas.

[0011] Preferably, step S1 includes the following steps:

[0012] Step S11: Retrieve land planning data from the database of the natural resources department, and perform format conversion and standardization processing on the land planning data;

[0013] Step S12: Extract the labeled cultivated land type from the preprocessed land planning type, and combine it with the geographic information system to transform the spatial coordinates of the cultivated land into vector boundaries to obtain the cultivated land area;

[0014] Step S13: Obtain a building distribution map, identify residential concentration areas through the building distribution map, and determine population concentration areas.

[0015] Preferably, in step S13, after identifying a residential area, nighttime remote sensing data of the residential area is obtained, nighttime light intensity values ​​are extracted from the nighttime remote sensing data, a regression model between the nighttime light intensity values ​​and the actual resident population is constructed, and the population concentration area is determined based on the regression model.

[0016] Preferably, step S2 includes the following specific steps:

[0017] Step S21: Unify the cultivated land area and the population concentration area into the same coordinate system;

[0018] Step S22: Calculate the actual distance between the center coordinates of each population cluster area and the center coordinates of each cultivated land area using a geographic information system;

[0019] Step S23: For each population cluster area, compare its actual distance with all cultivated land areas, and select the cultivated land area with the smallest distance as the cultivated land area to be planned.

[0020] Preferably, step S3 includes the following specific steps:

[0021] Step S31: Obtain high-resolution remote sensing image data of the cultivated area to be planned, and perform radiometric correction and image enhancement;

[0022] Step S32: Use an image segmentation algorithm to perform pixel-level classification on the remote sensing image data, identify and extract the boundary information of each farmland, and generate a farmland vector layer;

[0023] Step S33: Identify linear regions between farmlands as field ridges and construct a field ridge vector network;

[0024] Step S34: Based on the population cluster area, farmland vector layer, and field ridge vector network, construct the shortest path from the population cluster area to each farmland via the field ridge.

[0025] Preferably, the specific steps of step S34 are as follows: taking the population gathering area as the starting point and the entrance and exit of each farmland as the ending point, constructing several initial paths from the starting point to each ending point through the field ridge vector network, and optimizing the initial paths based on the shortest path using the Grey Wolf optimization algorithm to obtain the shortest path from the population gathering area to each farmland through the field ridge.

[0026] Preferably, step S4 includes the following specific steps:

[0027] Step S41: Collect the width, slope, road surface smoothness, obstacle density, and material hardness of each section of the shortest path as physical parameters, and standardize the physical parameters.

[0028] Step S42: Calculate the single-segment labor-saving index of a single-segment field ridge based on the physical parameters of each ridge, and sum the single-segment labor-saving indices to obtain the labor-saving index of the shortest path.

[0029] Step S43: Obtain soil fertility, biodiversity, and water purification capacity of surrounding ditches from the field ridges on one side of each farmland as micro-ecological parameters, and standardize the micro-ecological parameters.

[0030] Step S44: Use the analytic hierarchy process (AHP) to determine the weights of each microecological parameter, assign them to the microecological parameters, and then sum them to obtain the planting assistance index of the field ridge on the side of the farmland.

[0031] Preferably, step S5 includes the following specific steps:

[0032] Step S51: Assess the physical function data of each farmer in the densely populated area and obtain the physical fitness coefficient;

[0033] Step S52: Obtain the crop preferences of each farmer in the densely populated area, and obtain the crop matching coefficient with the field ridge based on the crop preferences;

[0034] Step S53: Add the product of the body fit coefficient and the labor saving index to the product of the crop fit coefficient and the planting assistance index to obtain the comprehensive score.

[0035] Step S54: Assign the farmland with the highest comprehensive score to farmers for planting.

[0036] Preferably, the steps for obtaining the body fit coefficient and crop fit coefficient are as follows: inputting the body function data and crop preference into the constructed neural network respectively, and obtaining the body fit coefficient and crop fit coefficient through neural network processing.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] This invention presents a rural land planning and construction method that integrates multi-source data. By combining data from land planning, building distribution, and remote sensing images, it achieves precise and dynamic rural land planning. Based on distance, it allocates the nearest arable land to densely populated areas, shortening farmers' round-trip distances and reducing time and physical labor costs. By extracting field ridges to construct the shortest path and calculating labor-saving indicators, and combining this with farmers' physical function data to allocate farmland, it significantly improves the convenience of labor for elderly and other special groups, reducing the risk of land abandonment. Simultaneously, based on field ridge micro-ecological parameters, it calculates planting assistance indicators, incorporating the ecological value of field ridges into the planning, promoting the stability of the farmland ecosystem and crop growth, and achieving synergistic optimization of production and ecology. Furthermore, the dynamic allocation mechanism can respond to changes in farmers' physical functions and land parameters, ensuring that the planning scheme continuously adapts to actual needs and providing strong support for the efficient use of rural land. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating a rural land planning and construction method that integrates multi-source data according to the present invention.

[0041] Figure 2 This is a flowchart of step S1 of a rural land planning and construction method that integrates multi-source data according to the present invention;

[0042] Figure 3 This is a flowchart of step S2 of a rural land planning and construction method that integrates multi-source data according to the present invention;

[0043] Figure 4 This is a flowchart of step S3 of a rural land planning and construction method that integrates multi-source data according to the present invention;

[0044] Figure 5 This is a flowchart of step S4 of a rural land planning and construction method that integrates multi-source data according to the present invention;

[0045] Figure 6 This is a flowchart of step S5 of a rural land planning and construction method that integrates multi-source data according to the present invention; Detailed Implementation

[0046] To better understand the technical content of this invention, a specific embodiment is provided below, and the invention will be further described in conjunction with the accompanying drawings.

[0047] See Figures 1 to 6 This invention provides a method for rural land planning and construction that integrates multi-source data, comprising the following steps:

[0048] Step S1: Obtain rural land planning data and building distribution map; determine cultivated land areas based on land planning data; and determine population concentration areas based on building distribution map.

[0049] Step S2: Calculate the distance between cultivated land areas and densely populated areas, assign the nearest cultivated land area to each densely populated area based on the distance, and output it as the cultivated land area to be planned.

[0050] Step S3: Obtain remote sensing image data of the cultivated area to be planned, extract farmland and field ridges based on the remote sensing image data, and construct the shortest path from the population gathering area to each farmland via the field ridges;

[0051] Step S4: Obtain the physical parameters and micro-ecological parameters of the field ridges, calculate the labor-saving index of the shortest path based on the physical parameters, and calculate the planting assistance index of the field ridges on the farmland side based on the micro-ecological parameters.

[0052] Step S5: Dynamically allocate farmland based on labor-saving indicators and planting assistance indicators combined with farmers' physical function data in densely populated areas.

[0053] This invention discloses a rural land planning and construction method that integrates multi-source data. Its main purpose is to allocate farmland that is close to, easily accessible, and easy to cultivate to elderly farmers with poor physical function. First, it is necessary to determine the arable land areas in the rural area. Rural land planning data can quickly determine the specific locations of several arable land areas in the rural area. Then, a rural building distribution map is obtained, and population clusters are determined by the clustering of buildings. Since population clusters and arable land areas may be scattered, and the distance from the same population cluster to different arable land areas varies significantly, the method involves calculating the distances between population clusters. This invention can assign the nearest arable land area to each population cluster, which can be designated as a planned arable land area. Each arable land area contains several plots of farmland, which are separated by field ridges. When assessing whether farmland is suitable for people with poor physical function, such as the elderly, in addition to the distance between farmland plots, this invention also considers whether the hardness and width of the field ridges are suitable for the elderly to walk on, whether it is convenient and labor-saving, and whether the microbial parameters of the field ridges can have a positive impact on the planting of crops. Therefore, after obtaining the planned arable land area, the farmland and field ridges can be quickly extracted from the remote sensing image data of the planned arable land area.

[0054] When selecting and evaluating farmland based on field ridges, the first step is to determine the distance from densely populated areas to each plot of farmland in the planned cultivation area via the field ridges. Based on this distance, the shortest path can be determined, which is the minimum distance from the densely populated area to each plot of farmland. However, farmland selection is not solely based on distance. Farmers need to walk along the field ridges on the shortest path, so the labor-saving aspect of the ridges and their contribution to crop cultivation are also factors to consider. By obtaining the physical and micro-ecological parameters of the field ridges, the labor-saving index of the shortest path for each plot of farmland can be calculated using the physical parameters. Simultaneously, the cultivation assistance index of the field ridges on one side of each plot can be calculated using the micro-ecological parameters. Finally, based on the labor-saving and cultivation assistance indices, combined with the physical function data of farmers in densely populated areas, farmland can be dynamically allocated. When farmers reach a plot of farmland via the shortest path, it requires less effort compared to other plots of farmland. Furthermore, during the cultivation process, the micro-ecological parameters on the field ridges can contribute to crop growth and increase farmland yield.

[0055] Preferably, step S1 includes the following steps:

[0056] Step S11: Retrieve land planning data from the database of the natural resources department, and perform format conversion and standardization processing on the land planning data;

[0057] Step S12: Extract the labeled cultivated land type from the preprocessed land planning type, and combine it with the geographic information system to transform the spatial coordinates of the cultivated land into vector boundaries to obtain the cultivated land area;

[0058] Step S13: Obtain a building distribution map, identify residential concentration areas through the building distribution map, obtain nighttime remote sensing data of residential concentration areas, extract nighttime light intensity values ​​through nighttime remote sensing data, construct a regression model between nighttime light intensity values ​​and actual resident population, and determine population concentration areas based on the regression model.

[0059] The database of the natural resources department stores land planning data for each rural area. After obtaining the land planning data for the corresponding rural area, the land planning data can be formatted and standardized. Then, the marked cultivated land types can be extracted from the processed land planning data. Finally, the cultivated land area can be vectorized through a geographic information system (GIS) to obtain cultivated land areas with clear boundaries.

[0060] When assessing densely populated areas, the first step is to obtain a building distribution map. This map reveals concentrated residential areas, which can be initially identified as densely populated areas. However, rural areas may have many migrant workers, requiring further assessment of these areas. Specifically, this involves obtaining nighttime remote sensing data of the corresponding residential areas, extracting nighttime light intensity values ​​from the data, and then constructing a regression model. Based on the specific magnitude of the nighttime light intensity values, the actual resident population can be estimated, thus determining whether the residential area has a large actual resident population and identifying it as a densely populated area.

[0061] Preferably, step S2 includes the following specific steps:

[0062] Step S21: Unify the cultivated land area and the population concentration area into the same coordinate system;

[0063] Step S22: Calculate the actual distance between the center coordinates of each population cluster area and the center coordinates of each cultivated land area using a geographic information system;

[0064] Step S23: For each population cluster area, compare its actual distance with all cultivated land areas, and select the cultivated land area with the smallest distance as the cultivated land area to be planned.

[0065] After determining the arable land areas and population concentration areas, it is necessary to determine the arable land area closest to each population concentration area. First, the arable land areas and population concentration areas are unified under the same coordinate system to facilitate measurement. Then, a GIS system is introduced to calculate the actual distance from each population concentration area to each arable land area. Finally, the arable land area with the smallest distance is output as the arable land area to be planned.

[0066] Preferably, step S3 includes the following specific steps:

[0067] Step S31: Obtain high-resolution remote sensing image data of the cultivated area to be planned, and perform radiometric correction and image enhancement;

[0068] Step S32: Use an image segmentation algorithm to perform pixel-level classification on the remote sensing image data, identify and extract the boundary information of each farmland, and generate a farmland vector layer;

[0069] Step S33: Identify linear regions between farmlands as field ridges and construct a field ridge vector network;

[0070] Step S34: Based on the population cluster area, farmland vector layer, and field ridge vector network, construct the shortest path from the population cluster area to each farmland via the field ridge.

[0071] After identifying the cultivated areas to be planned, remote sensing image data of these areas can be acquired and preprocessed. The processed remote sensing image data is then classified at the pixel level using an image segmentation algorithm to obtain the boundary information of each farmland, thus generating a vector layer of the farmland. The areas between farmlands are called field ridges. Each cultivated area to be planned consists of multiple farmlands, and the field ridges are used to separate them, forming a field ridge network. After identifying the linear regions between farmlands, the field ridge vector network can be obtained, and finally, the shortest path can be planned.

[0072] Preferably, the specific steps of step S34 are as follows: taking the population gathering area as the starting point and the entrance and exit of each farmland as the ending point, constructing several initial paths from the starting point to each ending point through the field ridge vector network, and optimizing the initial paths based on the shortest path using the Grey Wolf optimization algorithm to obtain the shortest path from the population gathering area to each farmland through the field ridge.

[0073] After determining the boundaries of farmland and the vector network of field ridges, path planning can be carried out with the population gathering area as the starting point and the entrance / exit of each farmland as the ending point. Since the field ridges are in a network shape, there are multiple paths from the population gathering area to the farmland via the field ridges. That is, there are multiple paths between each population gathering area and each farmland. This invention introduces the Grey Wolf Optimization Algorithm to find the shortest path and determine the shortest path. The shortest path represents the shortest distance from the population gathering area to the entrance / exit of a certain farmland via the field ridges.

[0074] Preferably, step S4 includes the following specific steps:

[0075] Step S41: Collect the width, slope, road surface smoothness, obstacle density, and material hardness of each section of the shortest path as physical parameters, and standardize the physical parameters.

[0076] Step S42: Calculate the single-segment labor-saving index of a single-segment field ridge based on the physical parameters of each ridge, and sum the single-segment labor-saving indices to obtain the labor-saving index of the shortest path.

[0077] Step S43: Obtain soil fertility, biodiversity, and water purification capacity of surrounding ditches from the field ridges on one side of each farmland as micro-ecological parameters, and standardize the micro-ecological parameters.

[0078] Step S44: Use the analytic hierarchy process (AHP) to determine the weights of each microecological parameter, assign them to the microecological parameters, and then sum them to obtain the planting assistance index of the field ridge on the side of the farmland.

[0079] After determining the shortest path, it is necessary to obtain the physical and micro-ecological parameters of the field ridges along the shortest path. The physical parameters include data such as the width and slope of the field ridges. Based on the physical parameters, the labor-saving index of a single field ridge segment can be calculated. Then, by combining the labor-saving indexes of the single segments, the labor-saving index of the shortest path can be obtained. The greater the width, the gentler the slope, the smoother the road surface, and the higher the density of obstacles, the less effort farmers need to put into walking on the field ridges.

[0080] Similarly, soil fertility, biodiversity, and water purification capacity of surrounding ditches on one side of the farmland are collected as micro-ecological parameters. After obtaining the micro-ecological parameters, the weights are determined by the analytic hierarchy process (AHP), and then the planting assistance index of the farmland ridge is obtained by weighted summation. For a single farmland, the shape can be approximated as a quadrilateral, so the number of ridges on the farmland ridge is 4. The sum of the planting assistance indexes of the 4 ridges is used as the total evaluation index.

[0081] Preferably, step S5 includes the following specific steps:

[0082] Step S51: Assess the physical function data of each farmer in the densely populated area and obtain the physical fitness coefficient;

[0083] Step S52: Obtain the crop preferences of each farmer in the densely populated area, and obtain the crop matching coefficient with the field ridge based on the crop preferences;

[0084] Step S53: Add the product of the body fit coefficient and the labor saving index to the product of the crop fit coefficient and the planting assistance index to obtain the comprehensive score.

[0085] Step S54: Assign the farmland with the highest comprehensive score to farmers for planting.

[0086] When allocating farmland, it is also necessary to consider the physical fitness data of farmers in densely populated areas. The physical fitness data can be used to obtain the physical fitness coefficient, which is related to the labor-saving index. At the same time, the farmers' planting preferences can be obtained, which are related to the planting assistance index. After assigning the physical fitness coefficient and crop fitness coefficient as weights to the labor-saving index and the planting assistance index, the results are summed to obtain a comprehensive score. Finally, the farmland with the highest comprehensive score can be allocated to the corresponding farmers.

[0087] Preferably, the steps for obtaining the body fit coefficient and crop fit coefficient are as follows: inputting the body function data and crop preference into the constructed neural network respectively, and obtaining the body fit coefficient and crop fit coefficient through neural network processing.

[0088] The body fit coefficient and crop fit coefficient can be obtained through a trained neural network. After training the neural network with a large amount of historical data, it can automatically identify the accurate body fit coefficient and crop fit coefficient, so as to be used for the calculation of the comprehensive score.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for rural land planning and construction that integrates multi-source data, characterized in that, Includes the following steps: Step S1: Obtain rural land planning data and building distribution map; determine cultivated land areas based on land planning data; and determine population concentration areas based on building distribution map. Step S2: Calculate the distance between cultivated land areas and densely populated areas, assign the nearest cultivated land area to each densely populated area based on the distance, and output it as the cultivated land area to be planned. Step S3: Obtain remote sensing image data of the cultivated area to be planned, extract farmland and field ridges based on the remote sensing image data, and construct the shortest path from the population gathering area to each farmland via the field ridges; Step S4: Obtain the physical parameters and micro-ecological parameters of the field ridges, calculate the labor-saving index of the shortest path based on the physical parameters, and calculate the planting assistance index of the field ridges on the farmland side based on the micro-ecological parameters. Step S5: Dynamically allocate farmland based on labor-saving indicators and planting assistance indicators combined with farmers' physical function data in densely populated areas.

2. The rural land planning and construction method integrating multi-source data according to claim 1, characterized in that, The specific steps of step S1 include: Step S11: Retrieve land planning data from the database of the natural resources department, and perform format conversion and standardization processing on the land planning data; Step S12: Extract the labeled cultivated land type from the preprocessed land planning type, and combine it with the geographic information system to transform the spatial coordinates of the cultivated land into vector boundaries to obtain the cultivated land area; Step S13: Obtain a building distribution map, identify residential concentration areas through the building distribution map, and determine population concentration areas.

3. The rural land planning and construction method integrating multi-source data according to claim 2, characterized in that, In step S13, after identifying a concentrated residential area, nighttime remote sensing data of the concentrated residential area is obtained. Nighttime light intensity values ​​are extracted from the nighttime remote sensing data, and a regression model between the nighttime light intensity values ​​and the actual resident population is constructed. Based on the regression model, the population concentration area is determined.

4. The rural land planning and construction method integrating multi-source data according to claim 1, characterized in that, The specific steps of step S2 include: Step S21: Unify the cultivated land area and the population concentration area into the same coordinate system; Step S22: Calculate the actual distance between the center coordinates of each population cluster area and the center coordinates of each cultivated land area using a geographic information system; Step S23: For each population cluster area, compare its actual distance with all cultivated land areas, and select the cultivated land area with the smallest distance as the cultivated land area to be planned.

5. A method for rural land planning and construction that integrates multi-source data according to claim 1, characterized in that, The specific steps of step S3 include: Step S31: Obtain high-resolution remote sensing image data of the cultivated area to be planned, and perform radiometric correction and image enhancement; Step S32: Use an image segmentation algorithm to perform pixel-level classification on the remote sensing image data, identify and extract the boundary information of each farmland, and generate a farmland vector layer; Step S33: Identify linear regions between farmlands as field ridges and construct a field ridge vector network; Step S34: Based on the population cluster area, farmland vector layer, and field ridge vector network, construct the shortest path from the population cluster area to each farmland via the field ridge.

6. The rural land planning and construction method integrating multi-source data according to claim 5, characterized in that, The specific steps of step S34 are as follows: take the population gathering area as the starting point and the entrance and exit of each farmland as the ending point, construct several initial paths from the starting point to each ending point through the field ridge vector network, optimize the initial paths based on the shortest path using the Grey Wolf optimization algorithm, and obtain the shortest path from the population gathering area to each farmland through the field ridge.

7. A method for rural land planning and construction that integrates multi-source data according to claim 1, characterized in that, The specific steps of step S4 include: Step S41: Collect the width, slope, road surface smoothness, obstacle density, and material hardness of each section of the shortest path as physical parameters, and standardize the physical parameters. Step S42: Calculate the single-segment labor-saving index of a single-segment field ridge based on the physical parameters of each ridge, and sum the single-segment labor-saving indices to obtain the labor-saving index of the shortest path. Step S43: Obtain soil fertility, biodiversity, and water purification capacity of surrounding ditches from the field ridges on one side of each farmland as micro-ecological parameters, and standardize the micro-ecological parameters. Step S44: Use the analytic hierarchy process (AHP) to determine the weights of each microecological parameter, assign them to the microecological parameters, and then sum them to obtain the planting assistance index of the field ridge on the side of the farmland.

8. A method for rural land planning and construction that integrates multi-source data according to claim 1, characterized in that, The specific steps of step S5 include: Step S51: Assess the physical function data of each farmer in the densely populated area and obtain the physical fitness coefficient; Step S52: Obtain the crop preferences of each farmer in the densely populated area, and obtain the crop matching coefficient with the field ridge based on the crop preferences; Step S53: Add the product of the body fit coefficient and the labor saving index to the product of the crop fit coefficient and the planting assistance index to obtain the comprehensive score. Step S54: Assign the farmland with the highest comprehensive score to farmers for planting.

9. A method for rural land planning and construction that integrates multi-source data according to claim 8, characterized in that, The steps for obtaining the body fit coefficient and crop fit coefficient are as follows: inputting the body function data and crop preference into the constructed neural network, and obtaining the body fit coefficient and crop fit coefficient through neural network processing.

Images