A method for rapid generation of soil nutrient distribution maps

Abstract

This application relates to a rapid method for generating soil nutrient distribution maps. The method includes: generating nutrient change directions at preset angle intervals, starting from the main variation direction, to obtain a direction sequence; clustering sampling points according to their nutrient changes in the corresponding nutrient change directions for the same index, resulting in a set of corresponding cluster regions; analyzing the dispersion of nutrient change directions corresponding to sampling points in each cluster region and the dispersion of nutrient change directions corresponding to sampling points in adjacent cluster regions to obtain the direction variation coefficient of the cluster region; determining the interpolation coefficient corresponding to the target nutrient for each prediction point based on the boundary curvature of the cluster region to which each prediction point belongs and the nutrient distribution within that cluster region; and assigning an interpolation scale to each prediction point according to the corresponding interpolation coefficient to generate a soil nutrient distribution map. This method can significantly improve the spatial accuracy and reliability of soil nutrient maps.

Description

Technical Field

[0001] This application relates to the field of soil mapping technology, and in particular to a method for rapidly generating soil nutrient distribution maps. Background Technology

[0002] As modern agriculture develops towards precision and efficiency, understanding and mastering the spatial distribution of nutrients (such as nitrogen, phosphorus, potassium, and organic matter) in farmland soil has become a crucial foundation for precision fertilization, improving crop yield and quality, and protecting arable land resources. Traditional soil nutrient survey methods rely on manual sampling, sample collection, and laboratory testing, which are time-consuming, labor-intensive, and difficult to apply in large-scale agricultural production. In recent years, with the development of technologies such as portable soil sensors, geographic information systems (GIS), remote sensing image analysis, and spatial interpolation modeling, spectral analysis can be used to quickly determine nutrient content in planned soil sampling. GIS can be used to analyze the relationship between nutrients and spatial location to construct distribution models, and spatial interpolation algorithms can be used to quickly generate continuous soil nutrient distribution maps, visually displaying the distribution of soil nutrients.

[0003] Kriging interpolation is typically used to quickly generate a continuous nutrient distribution surface based on planned discrete soil sampling point nutrient values. However, because the spatial autocorrelation index of the Kriging algorithm is sensitive to extreme values, its core mechanism is based on weighted averaging of spatial autocorrelation, meaning that the closer the spatial distance, the more similar the attributes. For example, near the edges of non-agricultural areas such as ditches and roads, soil nutrients often change drastically due to erosion, washout, or human intervention. If standard Kriging interpolation is used to process this area, the algorithm will simultaneously extract soil sampling points from the surrounding areas (high-value and low-value areas) for modeling at the interpolation points near the boundary, averaging the nutrient values ​​on both sides. The interpolation point will present a transitional value between high and low, forming a trend smoothing rather than an actual abrupt change. In soil areas with drastic nutrient changes, the originally clear abrupt changes may be blurred or even flattened. In areas with abrupt changes in nutrient aberrations, such as farmland boundaries, ditches, and roads, the Kriging algorithm will model them together with cultivated land data, resulting in distorted nutrient maps with blurred boundaries or abnormal diffusion. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a rapid method for generating soil nutrient distribution maps, the specific technical solution of which is as follows:

[0005] Firstly, a method for rapidly generating soil nutrient distribution maps is provided, the method comprising:

[0006] Starting with the principal variation direction corresponding to each sampling point in the target area, a nutrient change direction corresponding to that sampling point is generated at preset angle intervals to obtain a sequence of directions for each sampling point arranged by number; the principal variation direction of each sampling point is determined based on the difference in nutrient value of the target nutrient between that sampling point and its adjacent sampling points; the target nutrient is any nutrient in the target area;

[0007] For the same index in each directional sequence, each sampling point is clustered according to the nutrient change in the direction of nutrient change corresponding to that index, and the set of cluster regions corresponding to that index is obtained.

[0008] The dispersion of nutrient change direction corresponding to each sampling point in each cluster region of the same cluster region set, and the dispersion of nutrient change direction corresponding to each sampling point in adjacent cluster regions of the same cluster region set are analyzed to obtain the directional variation coefficient of the cluster region.

[0009] Based on the boundary tortuosity of the cluster region to which each prediction point belongs in each cluster region set and the nutrient distribution within that cluster region, the interpolation coefficient corresponding to the target nutrient for each prediction point is determined; the prediction point indicates the point in the target region other than the sampling point;

[0010] The interpolation scale is assigned to each prediction point based on the corresponding interpolation coefficient to generate a soil nutrient distribution map.

[0011] Optionally, the analysis of the dispersion of nutrient change direction corresponding to each sampling point in each cluster region of the same cluster region set, and the dispersion of nutrient change direction corresponding to each sampling point in adjacent cluster regions of the same cluster region set, to obtain the directional variation coefficient of the cluster region, includes:

[0012] The directional difference coefficient of a cluster region is determined based on the average value of the nutrient change direction corresponding to all sampling points in the same cluster region set, the number of multiple sampling points in each cluster region of the same cluster region set, and the nutrient change direction corresponding to each sampling point in the cluster region. The directional difference coefficient characterizes the dispersion of the nutrient change direction corresponding to each sampling point in the cluster region.

[0013] The directional variation coefficient of a cluster region is obtained by analyzing the directional difference coefficient of each cluster region in the same cluster region set and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set.

[0014] Optionally, the step of analyzing the directional difference coefficients of each cluster region in the same cluster region set and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set to which the cluster region belongs, to obtain the directional variation coefficient of the cluster region, includes:

[0015] The directional variation coefficient of a cluster region is determined by the number of multiple adjacent cluster regions in each cluster region within the same cluster region set, the directional variation coefficient of each adjacent cluster region, the directional variation coefficient of the cluster region, and the average value of the comprehensive value of the directional variation coefficients. The comprehensive value of the directional coefficients indicates the sum of the directional variation coefficients of the cluster region and the directional variation coefficients of its multiple adjacent cluster regions.

[0016] Optionally, the tortuosity of the boundary of the cluster region to which each predicted point belongs in each cluster region set is obtained according to the following steps:

[0017] Based on the boundary length of the cluster region to which each prediction point belongs in each cluster region set, the area of ​​the cluster region, and the variance of the nutrient values ​​of the target nutrients of all sampling points in the cluster region, the boundary tortuosity coefficient of the cluster region is determined; the boundary tortuosity coefficient characterizes the boundary tortuosity of the cluster region.

[0018] Optionally, the tortuosity of the boundary of the cluster region to which each predicted point belongs in each cluster region set is obtained according to the following steps:

[0019] The boundary tortuosity coefficient of the cluster region is determined by multiplying the tortuosity ratio and the nutrient fluctuation value. The tortuosity ratio indicates the ratio of the boundary length of the cluster region to which each predicted point belongs in each cluster region set to the area of ​​that cluster region. The nutrient fluctuation value indicates the variance of the target nutrient value of all sampling points in the cluster region.

[0020] Optionally, determining the interpolation coefficients for each prediction point corresponding to the target nutrient based on the boundary tortuosity of the cluster region to which each prediction point belongs in each cluster region set and the nutrient distribution within that cluster region includes:

[0021] Based on the boundary tortuosity coefficient of the cluster region to which each prediction point belongs in each cluster region set, and the average of the boundary tortuosity coefficients of all cluster regions in that cluster region set, the heterogeneity coefficient of each prediction point is determined; the heterogeneity coefficient indicates the degree of abrupt change in the nutrient value of the target nutrient at each prediction point in multiple nutrient change directions.

[0022] Based on the heterogeneity coefficient of each prediction point and the nutrient distribution within the cluster region to which the prediction point belongs in each cluster region set, the interpolation coefficient corresponding to the target nutrient for the prediction point is determined.

[0023] Optionally, determining the interpolation coefficient corresponding to the target nutrient for each predicted point based on the heterogeneity coefficient of each predicted point and the nutrient distribution within the cluster region to which the predicted point belongs in each cluster region set includes:

[0024] Based on the directional difference coefficient of the cluster region to which each prediction point belongs in each cluster region set and the heterogeneity coefficient of the prediction point, the heterogeneity score corresponding to the target nutrient of the prediction point is determined.

[0025] Based on the heterogeneity score of each prediction point and the nutrient distribution within the cluster region to which the prediction point belongs in each cluster region set, the interpolation coefficient corresponding to the target nutrient for the prediction point is determined.

[0026] Optionally, determining the interpolation coefficient corresponding to the target nutrient for each predicted point based on the heterogeneity score of each predicted point and the nutrient distribution within the cluster region to which the predicted point belongs in each cluster region set includes:

[0027] Based on the nutrient value sequence of the target nutrient measured at all sampling points in the target area and the nutrient value sequence of various other nutrients measured at all sampling points in the target area, calculate the Pearson correlation coefficient between the target nutrient and each other nutrient.

[0028] Based on the Pearson correlation coefficient between the target nutrient and each other nutrient, the heterogeneity score of each prediction point corresponding to the target nutrient, and the heterogeneity score of each prediction point corresponding to each other nutrient, the interpolation coefficient corresponding to the target nutrient for that prediction point is determined.

[0029] Optionally, the step of assigning interpolation scales to each prediction point according to the corresponding interpolation coefficients to generate a soil nutrient distribution map includes:

[0030] The interpolation scale for each prediction point corresponding to the target nutrient is determined based on the initial interpolation scale and the interpolation coefficient for each prediction point corresponding to the target nutrient; the initial interpolation scale for each prediction point corresponding to the target nutrient is calculated using the Kriging algorithm.

[0031] Soil nutrient distribution maps are generated based on the interpolation scale corresponding to the target nutrients at each prediction point.

[0032] Optionally, generating a soil nutrient distribution map based on the interpolation scale corresponding to the target nutrient at each predicted point includes:

[0033] Based on the interpolation scale of each prediction point corresponding to the target nutrient, multiple sampling points adjacent to each prediction point are selected from the target region.

[0034] Based on the variation characteristics of the target substance in multiple sampling points adjacent to each prediction point, the corresponding semi-variogram model is determined.

[0035] Based on the corresponding semivariogram model, determine the spatial semivariogram value between each prediction point and each adjacent sampling point, and construct the Kriging equation system;

[0036] Solve the Kriging equations to obtain the interpolation weights of each sampling point for the corresponding prediction point;

[0037] Based on the nutrient value of the target nutrient at each sampling point and the interpolation weight of each sampling point relative to the corresponding prediction point, the nutrient interpolation estimate of the corresponding prediction point is obtained.

[0038] A soil nutrient distribution map is generated based on the interpolated nutrient estimates from multiple prediction points.

[0039] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this application.

[0040] This application offers the following advantages: By using the main variation direction of each sampling point in the target area as the starting direction, generating a direction sequence corresponding to each sampling point at a preset angle, and clustering the sampling points based on nutrient changes in the same direction within the sequence, a multi-directional refined analysis of the spatial variation characteristics of soil nutrients is achieved. Introducing directional information for spatial clustering of soil can more realistically reflect changes in the spatial structure of nutrients, especially capturing local high heterogeneity areas caused by factors such as topography, fertilization, and drainage. By analyzing the directional dispersion of sampling points in the same cluster area and its adjacent areas, the directional variation coefficient of each cluster area is accurately calculated, effectively identifying high heterogeneity areas in nutrient distribution. Analyzing the directional variation coefficient of nutrients in each direction accurately identifies areas with prominent variation in a certain direction, revealing the directional distribution pattern. By dynamically determining the interpolation coefficient based on the boundary tortuosity and internal nutrient distribution of the cluster area to which the predicted point belongs, and allocating a multi-scale interpolation strategy accordingly, a high-precision preservation of local details and an organic unity of overall trend continuity are achieved, significantly improving the spatial representation accuracy and reliability of soil nutrient maps. By refining the directional clustering structure to the scale of prediction points, the differences in environmental background of each point in multiple directions are fully considered, and a local heterogeneity index is established. This allows the interpolation process to adaptively adjust the scale according to the level of heterogeneity, which avoids the oversmoothing problem in highly heterogeneous regions and improves the computational efficiency and robustness in low-heterogeneous regions. Ultimately, this achieves the rapid generation of high-efficiency, high-quality nutrient composition maps, providing a scientific basis and technical support for precision fertilization and precision agricultural management. Attached Figure Description

[0041] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of a method for rapidly generating a soil nutrient distribution map in one embodiment;

[0043] Figure 2 This is a schematic diagram of the structure of a rapid soil nutrient distribution map generation system in one embodiment;

[0044] Figure 3 This is a schematic diagram of the structure of an electronic device in one embodiment. Detailed Implementation

[0045] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for rapidly generating soil nutrient distribution maps according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0047] The following, with reference to the accompanying drawings, details a specific scheme for a rapid method of generating soil nutrient distribution maps provided in this application. For example... Figure 1 As shown, the method includes:

[0048] S11. Taking the main variation direction corresponding to each sampling point in the target area as the starting direction, generate a nutrient change direction corresponding to the sampling point at preset angle intervals to obtain the direction sequence of each sampling point arranged by number.

[0049] This application delineates target soil areas based on research objectives or agricultural production zoning, then divides these target areas into several sub-regions, and samples are collected within each sub-region. GPS (Global Positioning System) measuring instruments can be used to record the accurate latitude and longitude coordinates of each sampling point, forming a spatial database. Vehicle-mounted multi-parameter soil sensors are employed, including soil conductivity sensors, near-infrared spectroscopy sensors, and pH sensors (acidity / alkalinity sensors). The sensors are mounted on vehicles (tractors, ATVs, etc.), connected to a GPS and a data acquisition device or tablet. The vehicle travels along a predetermined route, and the sensors automatically collect data and location information at a set frequency (e.g., 1-5 times per second). The collected raw data (sensor readings + coordinates) is normalized and imported into a computer, for example, through max-min normalization, and then undergoes basic cleaning, such as removing obvious outliers (e.g., signal loss, GPS jump points) and checking coordinate accuracy.

[0050] The preset angle can be set according to the actual situation, for example, it can be 15°. Each sampling point corresponds to a direction sequence. Each direction sequence includes 24 nutrient change directions obtained by increasing the azimuth angle every 15° starting from the main variation direction of the sampling point. The direction sequence can be [θ, θ+15°, θ+30°, ..., θ+345°], where θ is the angle of the main variation direction. The direction sequence is an ordered list, and its elements are arranged according to a specific rule. The index in the direction sequence starts from 0. The index m=0 corresponds to the direction θ (main variation direction), the index m=1 corresponds to the direction θ+15, the index m=2 corresponds to the direction θ+30, ..., and so on.

[0051] The direction can be defined as 0° in a two-dimensional geographic coordinate system, rotating clockwise and increasing sequentially. First, determine the angle of the principal variation direction for each sampling point, such as 20°, 39°, etc. Then, based on the principal variation direction of each sampling point, determine multiple nutrient change directions for that sampling point. Each nutrient change direction in the direction sequence is represented in the form of angles, such as 25°, 54°, etc.

[0052] The principal variation direction of each sampling point is determined based on the difference in nutrient values ​​of the target nutrient between that sampling point and its adjacent sampling points. The target nutrient is any nutrient in the target area, such as nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, iron, zinc, copper, etc. The nutrient values ​​in this application are all calculated based on the nutrient values ​​of the target nutrient to generate a soil nutrient distribution map of the target nutrient. Specifically, taking the sampling point as the center, multiple adjacent sampling points are selected within the target area along each nutrient variation direction. The differences in nutrient values ​​between adjacent sampling points and the target nutrient at that sampling point, as well as the corresponding directional information, are analyzed. Methods such as gradient calculation, principal component analysis, or local semivariogram analysis are used to fit the maximum variation trend of the nutrient value in a two-dimensional geographic coordinate system. The direction corresponding to this trend is the principal variation direction of that sampling point.

[0053] S12. For the same index in each direction sequence, based on the nutrient change of each sampling point in the direction of nutrient change corresponding to that index, cluster each sampling point to obtain the set of cluster regions corresponding to that index.

[0054] In the rapid generation of soil nutrient distribution maps, only nutrient measurement data from discrete sampling points in the soil are available. Generating the distribution map requires continuous soil nutrient data, necessitating the use of spatial interpolation algorithms to construct a complete nutrient distribution picture. However, factors such as topography and hydrology, fertilization and tillage operations, and soil type boundaries can lead to high heterogeneity in nutrient distribution in some soil regions. During spatial interpolation, the interpolated data in highly heterogeneous areas becomes overly smooth, resulting in distorted nutrient distribution maps. To analyze soil regional heterogeneity, spatial clustering of the soil can be introduced using directional information to more accurately reflect changes in the spatial structure of nutrients, especially capturing locally highly heterogeneous areas caused by factors such as topography, fertilization, and drainage. This lays the foundation for subsequent quantification of directional heterogeneity and refined interpolation.

[0055] Spatial variability of soil nutrients often fails to accurately reveal its complex structure when analyzed in a single direction. If soil nutrients exhibit strong variability in only one direction but are stable in other directions, the heterogeneity is not necessarily high, but rather strong in direction. By analyzing the variability characteristics in multiple directions, the nutrient fluctuations in local areas can be more comprehensively identified. When a region shows significant variability in multiple directions, it often indicates a complex spatial structure, diverse controlling factors, and that it belongs to a highly heterogeneous region.

[0056] Therefore, for the same index in multiple directional sequences corresponding to multiple sampling points, the nutrient change direction corresponding to that index in the directional sequence of each sampling point can be obtained. Then, gradient calculation is used to analyze the nutrient change of each sampling point in the nutrient change direction corresponding to that index. Based on the analysis results, direction-aware K-means clustering is performed on multiple sampling points to obtain a set of cluster regions corresponding to each index. Each set of cluster regions contains multiple cluster regions, and each cluster region is a spatial grouping composed of multiple sampling points that exhibit similar spatial variation patterns in the nutrient change direction corresponding to the same index.

[0057] Soil nutrients exhibit spatial anisotropy, with inconsistent trends of variation in different directions. Each sampling point has its principal direction of variation. If the principal direction of variation of sampling points in a certain cluster region differs significantly from the directional pattern of the target region, it indicates that the cluster region may be affected by local special factors, leading to a complex nutrient distribution.

[0058] S13. Analyze the dispersion of the nutrient change direction corresponding to each sampling point in each cluster region of the same cluster region set, and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set to which the cluster region belongs, to obtain the directional variation coefficient of the cluster region.

[0059] The coefficient of directional variation indicates the relative strength of the difference in directional consistency between each cluster and its neighboring clusters within the same set of clusters. A larger coefficient of directional variation indicates a more significant abrupt change between the cluster and its neighboring clusters, and thus higher spatial heterogeneity.

[0060] In one embodiment, the dispersion of the nutrient change direction corresponding to each sampling point in each cluster region of the same cluster region set, and the dispersion of the nutrient change direction corresponding to each sampling point in adjacent cluster regions of the same cluster region set, are analyzed to obtain the directional variation coefficient of the cluster region, including:

[0061] The directional difference coefficient of a cluster region is determined based on the average value of the nutrient change direction corresponding to all sampling points in the same cluster region set, the number of multiple sampling points in each cluster region of the same cluster region set, and the nutrient change direction corresponding to each sampling point in the cluster region. The directional difference coefficient characterizes the dispersion of the nutrient change direction corresponding to each sampling point in the cluster region.

[0062] The directional variation coefficient of a cluster region is obtained by analyzing the directional difference coefficient of each cluster region in the same cluster region set and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set.

[0063] The directional difference coefficient for each cluster region is obtained by comparing the nutrient change direction corresponding to each sampling point in each cluster region with the nutrient change directions corresponding to all sampling points in the cluster region set to which the cluster region belongs. The nutrient change direction corresponding to each sampling point is obtained through the following steps: obtaining the index of the cluster region set to which the sampling point belongs, and retrieving the corresponding nutrient change direction in the direction sequence corresponding to the sampling point based on the index, thus obtaining the nutrient change direction corresponding to the sampling point.

[0064] The formula for calculating the directional difference coefficient is as follows: ;in, Indicates the first The directional difference coefficient of each cluster region Indicates the first The number of multiple sampling points within a cluster region Indicates the first The first cluster region One sampling point, Indicates the first Within the _ cluster region, the _ The direction of nutrient changes at each sampling point Indicates the first The average value of the nutrient change direction corresponding to all sampling points in the set of cluster regions to which a cluster region belongs.

[0065] In one embodiment, the directional variation coefficient of each cluster region in the same cluster region set and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set are analyzed to obtain the directional variation coefficient of the cluster region, including:

[0066] The directional variation coefficient of a cluster region is determined by the number of multiple adjacent cluster regions in each cluster region within the same cluster region set, the directional variation coefficient of each adjacent cluster region, the directional variation coefficient of the cluster region, and the average value of the comprehensive value of the directional variation coefficients. The comprehensive value of the directional coefficients indicates the sum of the directional variation coefficients of the cluster region and the directional variation coefficients of its multiple adjacent cluster regions.

[0067] If a cluster differs significantly in direction from its surrounding clusters, it indicates a strong spatial uniqueness and high heterogeneity. Conversely, if the directional trends are continuous, they may belong to the same geomorphic unit or work area, indicating low heterogeneity. The directional variation coefficient of each cluster can be obtained by comparing the differences in nutrient change directions between adjacent clusters. The directional variation coefficient of each cluster region The calculation formula is:

[0068] ;

[0069] in, Instruction No. The directional variation coefficient of each cluster region Indicates the relationship with the first The number of adjacent clusters belonging to the same set of clusters of a given cluster region, where adjacent clusters indicate the number of regions in the set of clusters of the given cluster region. The boundaries of each cluster region are adjacent to the cluster regions. Indicates the first The first cluster region Adjacent cluster regions, Indicates the first The first cluster region The directional difference coefficient of adjacent cluster regions Indicates the first The directional difference coefficient of each cluster region This represents the average value of the comprehensive value of the directional difference coefficients. The comprehensive value of the directional difference coefficients indicates the first... The directional difference coefficient of each cluster region and the sum of the directional difference coefficients of all its adjacent cluster regions.

[0070] S14. Based on the boundary tortuosity of the cluster region to which each prediction point belongs in each cluster region set and the nutrient distribution in that cluster region, determine the interpolation coefficient corresponding to the target nutrient for each prediction point.

[0071] The prediction points indicate points in the target area other than the sampling points. The resolution and accuracy can be set by the user, and a dense grid covering the target area is generated regularly. Each intersection node in the grid is a prediction point.

[0072] In one embodiment, the tortuosity of the boundary of the cluster region to which each predicted point belongs in each cluster region set is obtained according to the following steps:

[0073] Based on the boundary length of the cluster region to which each prediction point belongs in each cluster region set, the area of ​​the cluster region, and the variance of the nutrient values ​​of the target nutrients of all sampling points in the cluster region, the boundary tortuosity coefficient of the cluster region is determined; the boundary tortuosity coefficient characterizes the boundary tortuosity of the cluster region.

[0074] Furthermore, the tortuosity of the boundary of the cluster region to which each predicted point belongs in each cluster region set is obtained according to the following steps:

[0075] The boundary tortuosity coefficient of the cluster region is determined by multiplying the tortuosity ratio and the nutrient fluctuation value. The tortuosity ratio indicates the ratio of the boundary length of the cluster region to which each predicted point belongs in each cluster region set to the area of ​​that cluster region. The nutrient fluctuation value indicates the variance of the target nutrient value of all sampling points in the cluster region.

[0076] In highly heterogeneous regions, soil nutrient values ​​fluctuate drastically over short spatial distances (e.g., sudden increases or decreases). When performing clustering, the cluster boundaries need to be frequently switched to accommodate these dramatic changes, leading to highly heterogeneous regions being easily clustered into isolated areas with extremely tortuous boundaries, while nutrient values ​​also fluctuate dramatically. A better approach is to first determine the cluster region to which each predicted point belongs in each cluster set, and then obtain the tortuosity of that cluster region's boundary based on its boundary behavior and nutrient distribution.

[0077] The curvature of the boundary of the cluster region to which each predicted point belongs in each cluster region set The calculation formula is: ,in, The tortuosity of the boundary of the cluster region to which the predicted point belongs. This is the boundary length of the cluster region to which the predicted point belongs. The area of ​​the cluster region to which the predicted point belongs. This represents the variance of the target nutrient values ​​across all sampling points within the cluster to which the prediction point belongs, i.e., the nutrient fluctuation of the target nutrient within the cluster to which the prediction point belongs. The larger the size, the more obvious the nutrient changes. This represents the ratio of the boundary of the cluster region to which the predicted point belongs to to the area. The larger the value, the more tortuous the boundaries of the clustered regions.

[0078] If a highly heterogeneous region has a similar directional structure to a normally distributed region along the main variation direction, the heterogeneity contrast obtained by clustering only along the main variation direction may not be significant. To improve the accuracy of heterogeneity expression, clustering results from multiple directional gradients should be introduced for joint analysis. Normal regions typically exhibit low heterogeneity (stable directional structure) under other directional gradients, while regions with truly complex nutrient structures show high heterogeneity and significant structural fluctuations across multiple directional gradients. Therefore, the reliability of heterogeneity identification can be improved by considering the degree of consistency between directions. The heterogeneity coefficient of each predicted point within the multi-directional clustering region is obtained through the clustering results from multiple directional gradients.

[0079] Therefore, in one embodiment, based on the boundary tortuosity of the cluster region to which each prediction point belongs in each cluster region set and the nutrient distribution within that cluster region, the interpolation coefficient corresponding to the target nutrient for each prediction point is determined, including:

[0080] Based on the boundary tortuosity coefficient of the cluster region to which each prediction point belongs in each cluster region set, and the average of the boundary tortuosity coefficients of all cluster regions in that cluster region set, the heterogeneity coefficient of each prediction point is determined; the heterogeneity coefficient indicates the degree of abrupt change in the nutrient value of the target nutrient at each prediction point in multiple nutrient change directions.

[0081] Based on the heterogeneity coefficient of each prediction point and the nutrient distribution within the cluster region to which the prediction point belongs in each cluster region set, the interpolation coefficient corresponding to the target nutrient for the prediction point is determined.

[0082] No. Heterogeneity coefficient of each prediction point The calculation formula is:

[0083] ;

[0084] in, For the first Heterogeneity coefficient of each prediction point Let be the number of multiple cluster regions, and 'a' be the a-th cluster region set. For the first The tortuosity of the boundary of the cluster region to which the predicted point belongs in the a-th cluster region set. Let be the mean of the boundary tortuosity of all cluster regions in the a-th cluster region set. Represents the i-th cluster in any set of clustered regions The contrast between the heterogeneity of the cluster region to which each predicted point belongs and other regions in that cluster region set. The larger the number, the higher the number. The greater the heterogeneity contrast of the cluster region to which a prediction point belongs, the better.

[0085] If a prediction point is located in multiple cluster regions, and the nutrient variability of its cluster region is high, while the heterogeneity contrast with adjacent cluster regions is also generally high, it indicates that the prediction point is in a spatial location with complex structure and strong trend changes, and its spatial heterogeneity is high. It should be identified as a key heterogeneous point.

[0086] Therefore, in one embodiment, the interpolation coefficient corresponding to the target nutrient for each predicted point is determined based on the heterogeneity coefficient of each predicted point and the nutrient distribution within the cluster region to which the predicted point belongs in each cluster region set, including:

[0087] Based on the directional difference coefficient of the cluster region to which each prediction point belongs in each cluster region set and the heterogeneity coefficient of the prediction point, the heterogeneity score corresponding to the target nutrient of the prediction point is determined.

[0088] Based on the heterogeneity score of each prediction point and the nutrient distribution within the cluster region to which the prediction point belongs in each cluster region set, the interpolation coefficient corresponding to the target nutrient for the prediction point is determined.

[0089] No. Each prediction point corresponds to a heterogeneity score for the target nutrient. The calculation formula is: ,in, For the first Each prediction point corresponds to a heterogeneity score for the target nutrient. Let be the number of multiple cluster regions, and 'a' be the a-th cluster region set. For the first The directional variation coefficient of the cluster region to which the predicted point belongs in the a-th cluster region set. For the first Heterogeneity coefficients for each prediction point.

[0090] In soil, there are certain correlations, synergies, or antagonistic mechanisms among different nutrient elements. When a predicted point in a certain nutrient distribution already shows high spatial heterogeneity, if the predicted point also shows high heterogeneity in other nutrient distributions that are significantly correlated with it, then the point is more likely to be in a true structural abrupt region or functional heterogeneous region, and therefore should be judged as a high-confidence high-heterogeneity point.

[0091] Therefore, in one embodiment, the interpolation coefficient corresponding to the target nutrient for each predicted point is determined based on the heterogeneity score of each predicted point and the nutrient distribution within the cluster region to which the predicted point belongs in each cluster region set, including:

[0092] Based on the nutrient value sequence of the target nutrient measured at all sampling points in the target area and the nutrient value sequence of various other nutrients measured at all sampling points in the target area, calculate the Pearson correlation coefficient between the target nutrient and each other nutrient.

[0093] Based on the Pearson correlation coefficient between the target nutrient and each other nutrient, the heterogeneity score of each prediction point corresponding to the target nutrient, and the heterogeneity score of each prediction point corresponding to each other nutrient, the interpolation coefficient corresponding to the target nutrient for that prediction point is determined.

[0094] Based on the heterogeneity score of each predicted point for the target nutrient and the relevant nutrient distribution at that predicted point, the interpolation coefficient corresponding to the target nutrient can be determined. , The calculation formula is:

[0095] ;

[0096] in, For the first Each prediction point corresponds to an interpolation coefficient for the target nutrient. For the first Nutrients The quantity of various nutrients in the target area. The sequence of nutrient values ​​for the target nutrients measured at all sampling points in the target area and the sequence of the target nutrients measured at all sampling points in the target area. The Pearson correlation coefficient between nutrient value sequences of seed nutrient, For the first Nutrients in the first Heterogeneity score of each prediction point For the first Each prediction point corresponds to a heterogeneity score for the target nutrient. Indicates the first The stronger the correlation between the seed nutrient and the target nutrient, the greater the spatial heterogeneity of the same prediction point, which indicates that the spatial heterogeneity of the prediction point may be higher.

[0097] S15. Assign interpolation scales to each prediction point according to the corresponding interpolation coefficients to generate a soil nutrient distribution map.

[0098] In the process of spatial mapping of soil nutrients, due to the significant differences in spatial heterogeneity across different regions, applying a uniform interpolation scale to all prediction points can easily lead to over-smoothing of highly heterogeneous areas, resulting in the loss of local details or misjudgment. Therefore, an appropriate interpolation scale can be dynamically selected based on the heterogeneity intensity of each prediction point. Small-scale interpolation can be used for points with high heterogeneity to capture local fluctuations more precisely, thus reflecting the true distribution characteristics of soil nutrients more accurately while balancing prediction accuracy and spatial continuity.

[0099] In one embodiment, an interpolation scale is assigned to each prediction point according to the corresponding interpolation coefficient to generate a soil nutrient distribution map, including:

[0100] The interpolation scale for each prediction point corresponding to the target nutrient is determined based on the initial interpolation scale and the interpolation coefficient for each prediction point corresponding to the target nutrient; the initial interpolation scale for each prediction point corresponding to the target nutrient is calculated using the Kriging algorithm.

[0101] Soil nutrient distribution maps are generated based on the interpolation scale corresponding to the target nutrients at each prediction point.

[0102] In the process of using the Kriging algorithm for spatial modeling and prediction interpolation of soil nutrients, to prevent over-smoothing, the interpolation scale needs to be dynamically reduced as the heterogeneity of each prediction point increases. The interpolation scale for each prediction point can be obtained based on the interpolation coefficients of each prediction point in the target region. Interpolation scale for each prediction point The calculation formula is: ,in, For the first Interpolation scale for each prediction point For the first Each prediction point corresponds to an initial interpolation scale for the target nutrient, which is calculated using the Kriging algorithm. For the first Each prediction point corresponds to an interpolation coefficient for the target nutrient.

[0103] After assigning an interpolation scale based on spatial heterogeneity analysis to each prediction point, the soil nutrient composition map can be quickly generated by combining the Kriging interpolation algorithm.

[0104] In one embodiment, generating a soil nutrient distribution map based on the interpolation scale corresponding to the target nutrient at each predicted point includes:

[0105] Based on the interpolation scale of each prediction point corresponding to the target nutrient, multiple sampling points adjacent to each prediction point are selected from the target region.

[0106] Based on the variation characteristics of the target substance in multiple sampling points adjacent to each prediction point, the corresponding semi-variogram model is determined.

[0107] Based on the corresponding semivariogram model, determine the spatial semivariogram value between each prediction point and each adjacent sampling point, and construct the Kriging equation system;

[0108] Solve the Kriging equations to obtain the interpolation weights of each sampling point for the corresponding prediction point;

[0109] Based on the nutrient value of the target nutrient at each sampling point and the interpolation weight of each sampling point relative to the corresponding prediction point, the nutrient interpolation estimate of the corresponding prediction point is obtained.

[0110] A soil nutrient distribution map is generated based on the interpolated nutrient estimates from multiple prediction points.

[0111] The interpolation scale determines the search range of sample points around each prediction point and the parameters of the semivariogram used. First, based on the interpolation scale corresponding to the target nutrient at each prediction point, multiple sampling points adjacent to each prediction point are selected from the target region. Then, based on the variation characteristics of the target substance in these adjacent sampling points, the corresponding semivariogram model is determined. Subsequently, using the corresponding semivariogram model, the spatial semivariogram value between each prediction point and each adjacent sampling point is determined, and a Kriging equation system is constructed to solve for the interpolation weight of each sampling point relative to the prediction point. Based on the target nutrient value at each sampling point and the interpolation weight of each sampling point relative to the corresponding prediction point, the nutrient interpolation estimate for the corresponding prediction point is obtained. Based on the nutrient interpolation estimates of multiple prediction points, a soil nutrient distribution map is generated.

[0112] The above process is repeated for all prediction points, and kriging interpolation predictions are completed independently based on their respective interpolation scales and structural characteristics. Batch or parallel computation can be used to accelerate the generation efficiency. Finally, all prediction results are stitched together to form a complete continuous nutrient space layer, forming a high-resolution, structure-sensitive soil nutrient distribution map. Compared with the traditional globally uniform interpolation method, this significantly improves the prediction accuracy of highly heterogeneous areas while taking into account the computational efficiency of low-heterogeneous areas, achieving an organic unity between rapid nutrient map generation and refined management.

[0113] This application generates a directional sequence for each sampling point at a preset angle, using the main variation direction of each sampling point in the target area as the starting direction. Based on nutrient changes along the same direction in the sequence, the sampling points are clustered, achieving multi-directional refined analysis of the spatial variation characteristics of soil nutrients. Introducing directional information for spatial clustering of soil more accurately reflects the spatial structure changes of nutrients, especially capturing localized high heterogeneity areas caused by factors such as topography, fertilization, and drainage. By analyzing the directional dispersion of sampling points within the same cluster and its adjacent regions, the directional variation coefficient of each cluster is accurately calculated, effectively identifying highly heterogeneous nutrient distribution areas. Analyzing the directional variation coefficient of nutrients in each direction accurately identifies areas with prominent variation in a certain direction, revealing directional distribution patterns. By dynamically determining interpolation coefficients based on the boundary tortuosity and internal nutrient distribution of the cluster to which the predicted point belongs, and allocating multi-scale interpolation strategies accordingly, a high-precision preservation of local details and an organic unity of overall trend continuity are achieved, significantly improving the spatial accuracy and reliability of soil nutrient maps. By refining the directional clustering structure to the scale of prediction points, the differences in environmental background of each point in multiple directions are fully considered, and a local heterogeneity index is established. This allows the interpolation process to adaptively adjust the scale according to the level of heterogeneity, which avoids the oversmoothing problem in highly heterogeneous regions and improves the computational efficiency and robustness in low-heterogeneous regions. Ultimately, this achieves the rapid generation of high-efficiency, high-quality nutrient composition maps, providing a scientific basis and technical support for precision fertilization and precision agricultural management.

[0114] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0115] This application also provides a system for rapidly generating soil nutrient distribution maps, such as... Figure 2 As shown, the system includes:

[0116] The first generation module 21 is used to generate a nutrient change direction corresponding to each sampling point in the target area, starting from the main variation direction corresponding to each sampling point, and generating a nutrient change direction corresponding to each sampling point at preset angle intervals, so as to obtain a sequence of directions arranged by number for each sampling point; the main variation direction of each sampling point is determined according to the difference in nutrient value of the target nutrient between the sampling point and its adjacent sampling points; the target nutrient is any nutrient in the target area;

[0117] Clustering module 22 is used to cluster each sampling point according to the nutrient change in the nutrient change direction corresponding to the directional sequence for the same index, and obtain the set of clustering regions corresponding to the index.

[0118] Analysis module 23 is used to analyze the dispersion of the nutrient change direction corresponding to each sampling point in each cluster region of the same cluster region set, and the dispersion of the nutrient change direction corresponding to each sampling point in the adjacent cluster regions of the same cluster region set to which the cluster region belongs, and to obtain the direction variation coefficient of the cluster region.

[0119] The determination module 24 is used to determine the interpolation coefficients corresponding to the target nutrients for each prediction point based on the boundary tortuosity of the cluster region to which each prediction point belongs in each cluster region set and the nutrient distribution within that cluster region; the prediction points indicate points in the target region other than the sampling points;

[0120] The second generation module 25 is used to assign interpolation scales to each prediction point according to the corresponding interpolation coefficients in order to generate a soil nutrient distribution map.

[0121] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs.

[0122] Figure 3 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the method described in any of the above embodiments. Figure 3The electronic device 30 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0123] like Figure 3 As shown, the electronic device 30 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including memory 32 and processor 31).

[0124] Bus 33 includes a data bus, an address bus, and a control bus.

[0125] The memory 32 may include volatile memory, such as random access memory (RAM) 321 and / or cache memory 322, and may further include read-only memory (ROM) 323.

[0126] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) program module 324, such program module 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0127] The processor 31 executes various functional applications and data processing, such as the methods provided in any of the above embodiments, by running computer programs stored in the memory 32.

[0128] Electronic device 30 can also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 35. Furthermore, electronic device 30 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 36. As shown, network adapter 36 communicates with other modules of electronic device 30 via bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0129] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0130] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0131] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0132] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0133] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above embodiments.

[0134] The program code for executing the computer program product of this application can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

[0137] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for rapid generation of soil nutrient profile, characterized in that, The method comprises: Taking a main variation direction corresponding to each sampling point in the target region as a starting direction, generating a nutrient variation direction corresponding to the sampling point every interval of a preset angle to obtain a direction sequence corresponding to each sampling point in sequence number; the main variation direction of each sampling point is determined according to the difference in the nutrient value of the target nutrient between the sampling point and its adjacent sampling points; the target nutrient is any nutrient in the target region; For the same sequence number in each direction sequence, clustering the sampling points according to the nutrient variation of the sampling points in the nutrient variation direction corresponding to the sequence number to obtain a clustering region set corresponding to the sequence number; Analyzing the dispersion degree of the nutrient variation direction corresponding to each sampling point in each clustering region of the same clustering region set and the dispersion degree of the nutrient variation direction corresponding to each sampling point in the adjacent clustering region belonging to the same clustering region set of the clustering region to obtain the direction variation coefficient of the clustering region; Based on the boundary tortuosity of the clustering region to which each prediction point belongs in each clustering region set and the nutrient distribution in the clustering region, determining the interpolation coefficient of the target nutrient corresponding to each prediction point; the prediction point indicates a point in the target region other than the sampling point; Assigning an interpolation scale to each prediction point according to the corresponding interpolation coefficient to generate a soil nutrient distribution map; The analysis of the dispersion degree of the nutrient variation direction corresponding to each sampling point in each clustering region of the same clustering region set and the dispersion degree of the nutrient variation direction corresponding to each sampling point in the adjacent clustering region belonging to the same clustering region set of the clustering region to obtain the direction variation coefficient of the clustering region comprises: Determining the direction difference coefficient of the clustering region according to the average value of the nutrient variation direction corresponding to all sampling points in the same clustering region set, the number of sampling points in each clustering region of the same clustering region set, and the nutrient variation direction corresponding to each sampling point in the clustering region; the direction difference coefficient represents the dispersion degree of the nutrient variation direction corresponding to each sampling point in the clustering region; Analyzing the direction difference coefficient of each clustering region in the same clustering region set and the dispersion degree of the nutrient variation direction corresponding to each sampling point in the adjacent clustering region belonging to the same clustering region set of the clustering region to obtain the direction variation coefficient of the clustering region; The boundary tortuosity of the clustering region to which each prediction point belongs in each clustering region set is obtained according to the following steps: Based on the boundary length of the clustering region to which each prediction point belongs in each clustering region set, the area of the clustering region, and the variance of the nutrient value of the target nutrient of all sampling points in the clustering region, determining the boundary tortuosity coefficient of the clustering region; the boundary tortuosity coefficient represents the boundary tortuosity of the clustering region.

2. A method of rapid generation of soil nutrient profile as claimed in claim 1 wherein, The analysis of the direction difference coefficient of each clustering region in the same clustering region set and the dispersion degree of the nutrient variation direction corresponding to each sampling point in the adjacent clustering region belonging to the same clustering region set of the clustering region to obtain the direction variation coefficient of the clustering region comprises: The direction variation coefficient of the cluster region is determined according to the number of the plurality of adjacent cluster regions of the cluster region in the same set of cluster regions, the direction difference coefficient of each adjacent cluster region of the cluster region, the direction difference coefficient of the cluster region, and the average of the comprehensive values of the direction difference coefficients.

3. The method of claim 1, wherein the soil nutrient profile is generated rapidly. The boundary tortuosity of the cluster region to which each prediction point belongs in the set of cluster regions is obtained according to the following steps: The boundary tortuosity coefficient of the cluster region is determined based on the product of the tortuosity ratio and the nutrient fluctuation value; the tortuosity ratio indicates the ratio of the boundary length of the cluster region to which each prediction point belongs in each set of cluster regions to the area of the cluster region, and the nutrient fluctuation value indicates the variance of the nutrient values of the target nutrient of all sampling points in the cluster region.

4. The method of claim 1, wherein the soil nutrient profile is generated rapidly. The interpolation coefficient of each prediction point corresponding to the target nutrient is determined based on the boundary tortuosity of the cluster region to which each prediction point belongs in the set of cluster regions and the nutrient distribution in the cluster region, including: The heterogeneity coefficient of each prediction point is determined based on the boundary tortuosity coefficient of the cluster region to which each prediction point belongs in the set of cluster regions and the average of the boundary tortuosity coefficients of all cluster regions in the set of cluster regions; the heterogeneity coefficient indicates the degree of mutation of the nutrient value of the target nutrient of each prediction point in multiple nutrient change directions; The interpolation coefficient of each prediction point corresponding to the target nutrient is determined according to the heterogeneity coefficient of each prediction point and the nutrient distribution in the cluster region to which the prediction point belongs in the set of cluster regions.

5. A method of rapid generation of soil nutrient profile as claimed in claim 4 wherein, The interpolation coefficient of each prediction point corresponding to the target nutrient is determined according to the heterogeneity coefficient of each prediction point and the nutrient distribution in the cluster region to which the prediction point belongs in the set of cluster regions, including: The heterogeneity score of each prediction point corresponding to the target nutrient is determined according to the direction difference coefficient of the cluster region to which each prediction point belongs in the set of cluster regions and the heterogeneity coefficient of the prediction point; The interpolation coefficient of each prediction point corresponding to the target nutrient is determined according to the heterogeneity score of each prediction point and the nutrient distribution in the cluster region to which the prediction point belongs in the set of cluster regions.

6. A method of rapid generation of soil nutrient profile as claimed in claim 5 wherein, The interpolation coefficient of each prediction point corresponding to the target nutrient is determined according to the heterogeneity score of each prediction point and the nutrient distribution in the cluster region to which the prediction point belongs in the set of cluster regions, including: Pearson correlation coefficients between the target nutrient and each other nutrient are calculated respectively based on the sequence of nutrient values of the target nutrient measured by all sampling points in the target region and the sequence of nutrient values of the plurality of other nutrients measured by all sampling points in the target region; The interpolation coefficient of each prediction point corresponding to the target nutrient is determined according to the Pearson correlation coefficient of the target nutrient and each other nutrient, the heterogeneity score of the prediction point corresponding to the target nutrient, and the heterogeneity score of the prediction point corresponding to each other nutrient.

7. The method of claim 1, wherein the soil nutrient profile is generated rapidly. The interpolation scale is assigned to each prediction point according to the corresponding interpolation coefficient to generate a soil nutrient distribution map, including: According to the initial interpolation scale corresponding to the target nutrient of each prediction point, the interpolation coefficient corresponding to the target nutrient of each prediction point, the interpolation scale corresponding to the target nutrient of each prediction point is determined; the initial interpolation scale corresponding to the target nutrient of each prediction point is calculated according to the Kriging algorithm; According to the interpolation scale corresponding to the target nutrient of each prediction point, a soil nutrient distribution map is generated.

8. A method of rapid generation of soil nutrient profile as claimed in claim 7 wherein, The soil nutrient distribution map generated according to the interpolation scale corresponding to the target nutrient of each prediction point comprises: According to the interpolation scale corresponding to the target nutrient of each prediction point, a plurality of sampling points adjacent to each prediction point are selected from the target area; According to the change characteristics of the target substance in the plurality of sampling points adjacent to each prediction point, a corresponding semi-variogram function model is determined; According to the corresponding semi-variogram function model, the spatial semi-variogram value between each prediction point and each adjacent sampling point is determined, and a Kriging equation set is constructed; Solving the Kriging equation set, the interpolation weight of each sampling point for the corresponding prediction point is obtained; According to the nutrient value of the target nutrient of each sampling point and the interpolation weight of each sampling point for the corresponding prediction point, the nutrient interpolation estimation value of the corresponding prediction point is obtained; According to the nutrient interpolation estimation values of the plurality of prediction points, a soil nutrient distribution map is generated.

Images