Farmland quality spatial heterogeneity layering method and device, equipment and storage medium

Abstract

The application provides a farmland quality spatial heterogeneity layering method, device, equipment and storage medium, and belongs to the technical field of data processing, and comprises the following steps: performing classification principal component analysis on obtained monitoring index data to determine index weights of monitoring indexes; marking importance labels of the monitoring indexes according to the index weights and index importance degree threshold values; respectively layering and fusing the monitoring indexes marked with the same type of importance labels, performing spatial overlay analysis on all layering and fusion results, and obtaining farmland quality spatial heterogeneity layering results. The farmland quality spatial heterogeneity layering method provided by the application highlights the contribution of important indexes to farmland quality by calculating index weights through classification principal component analysis and then dividing important indexes and general indexes, reduces the interference of general indexes on farmland quality, and guarantees the rationality and accuracy of farmland quality spatial layering.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, equipment and storage medium for stratifying spatial heterogeneity of cultivated land quality. Background Art

[0002] Arable land quality is a combination of soil fertility, soil health, and field infrastructure, defining its ability to sustainably produce and maintain the quality and safety of agricultural products. The arable land quality grading system uses a comprehensive index method to evaluate the ability of these factors, including soil fertility, soil health, and field infrastructure, to sustainably produce and maintain the quality and safety of agricultural products, from the perspective of agricultural production. Arable land quality monitoring is fundamental to arable land protection and utilization, providing timely insights into changing trends in arable land quality and playing a vital role in guiding agricultural production and protecting arable land resources. Establishing arable land quality monitoring points provides basic data on arable land quality. However, due to limited human and material resources, it is not possible to monitor every arable land patch within the monitoring area. Therefore, the spatial distribution of arable land quality monitoring points will affect the accuracy of arable land quality monitoring.

[0003] Given the heterogeneity of cultivated land quality in its spatial distribution, randomly placing cultivated land quality monitoring points may result in ineffective coverage of heterogeneous areas, and monitoring data may not fully and accurately reflect the cultivated land quality status of the entire monitored area. Spatial heterogeneity refers to the unevenness and complexity of the spatial distribution of cultivated land quality. Spatial stratification can divide the monitored area into different layers or regions (referred to as spatial stratification) based on the heterogeneity of cultivated land quality. Based on the characteristics of each spatial stratum, cultivated land quality monitoring points can be targeted to ensure the scientific and effective monitoring of cultivated land quality.

[0004] In this field, research on spatial stratification uses land use types and administrative divisions as basic units of spatial stratification. While this approach is simple and easy to implement, and stratification data is readily available, it cannot fully and accurately reflect the complex internal variation within the region under investigation and is unsuitable for areas with high levels of variation. Spatial stratification of regions under investigation based on prior knowledge, environmental factors, auxiliary data, and historical data relies heavily on data quality and methodological rationality. This approach suffers from insufficient capture of intra-stratum variation, reducing the accuracy and reliability of the spatial stratification results. Summary of the Invention

[0005] The present invention provides a method, device, equipment and storage medium for stratifying spatial heterogeneity of cultivated land quality. The method calculates indicator weights through classification principal component analysis and then divides important indicators and general indicators. After stratifying the two separately, spatial heterogeneity stratification is performed according to the stratification fusion rules, which can effectively improve the accuracy and reliability of the spatial stratification results.

[0006] The application provides a farmland quality spatial heterogeneity layering method, comprising the following steps:

[0007] All monitoring indexes reflecting the farmland quality level of the to-be-tested area are determined, and monitoring index data corresponding to each monitoring index is obtained;

[0008] The index weight of each monitoring index is determined based on the classification principal component analysis of the monitoring index data of the monitoring indexes;

[0009] The importance label of each monitoring index is labeled according to the comparison result of the index weight of each monitoring index and the index importance threshold value; the index importance threshold value is determined based on the mean and standard deviation of all index weights, and the importance label has at least two types;

[0010] The monitoring indexes labeled with the same type of importance label are respectively subjected to layering fusion to obtain layering fusion results;

[0011] The spatial superposition analysis is performed on all layering fusion results to obtain the farmland quality spatial heterogeneity layering result of the to-be-tested area.

[0012] According to the farmland quality spatial heterogeneity layering method provided by the application, the index weight of each monitoring index is determined based on the classification principal component analysis of the monitoring index data of the monitoring indexes, and the specific steps include:

[0013] The monitoring index data is subjected to dimensionless normalization processing;

[0014] The classification principal component analysis is performed on the monitoring index data subjected to dimensionless normalization processing to obtain the principal component load corresponding to each monitoring index, the characteristic value of each principal component and the variance contribution rate;

[0015] The linear combination coefficient of each principal component corresponding to each monitoring index is calculated according to the principal component load corresponding to the monitoring index and the characteristic value of each principal component;

[0016] The score coefficient corresponding to each monitoring index is calculated based on the linear combination coefficient of each principal component corresponding to all monitoring indexes and the variance contribution rate of each principal component;

[0017] The score coefficient corresponding to each monitoring index is subjected to normalization processing to determine the index weight of each monitoring index.

[0018] According to the farmland quality spatial heterogeneity layering method provided by the application, the first function expression for calculating the linear combination coefficient of each principal component corresponding to each monitoring index according to the principal component load corresponding to the monitoring index and the characteristic value of each principal component is:

[0019] ;

[0020] in, For the i Monitoring indicators j The linear combination coefficients of the principal components; For the i Monitoring indicators j The principal component loadings of the principal components; For the j the eigenvalues ​​of the principal components; n is the number of monitoring indicators, ; m is the number of principal components, .

[0021] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, the second function expression for calculating the score coefficient corresponding to each monitoring indicator based on the linear combination coefficient of each principal component corresponding to all the monitoring indicators and the variance contribution rate of each principal component is:

[0022] ;

[0023] in, For the i The score coefficient corresponding to each monitoring indicator; For the j The variance contribution rate of the principal components.

[0024] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, the score coefficient corresponding to each monitoring indicator is normalized to determine the third function expression of the indicator weight of each monitoring indicator:

[0025] ;

[0026] in, For the i The indicator weight corresponding to each monitoring indicator.

[0027] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, the importance label of each monitoring indicator is marked based on the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold, including:

[0028] Marking the importance label of the monitoring indicator whose indicator weight is greater than the indicator importance threshold as an important indicator label;

[0029] The importance label of the monitoring indicator whose indicator weight is not greater than the indicator importance threshold is marked as a general indicator label.

[0030] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, the monitoring indicators marked with the same type of importance labels are stratified and fused to obtain stratified fusion results, including:

[0031] Performing indicator stratification on each monitoring indicator whose importance label is marked as the important indicator label, and obtaining an indicator stratification result corresponding to each monitoring indicator; performing spatial overlay analysis on the indicator stratification results corresponding to all the monitoring indicators, and obtaining spatial combinations of various indicator stratification results to form a plurality of different first-level combinations; performing fusion processing on all the first-level combinations according to a preset stratification fusion rule to determine an important indicator stratification result;

[0032] According to the preset number of clusters, cluster analysis is performed on each monitoring indicator whose importance label is marked as the general indicator label, and each monitoring indicator marked as the general indicator label is divided into multiple cluster categories; the average value of the comprehensive index of cultivated land quality of each cluster category is determined; according to the size of the average value of the comprehensive index of cultivated land quality, according to the preset level division rules, the cluster categories are divided into multiple preset levels to obtain the general indicator stratification results.

[0033] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, performing spatial overlay analysis on all the stratified fusion results to obtain the stratified results of spatial heterogeneity of cultivated land quality in the measured area includes:

[0034] Performing spatial overlay analysis on the important indicator stratification results and the general indicator stratification results to obtain spatial combinations of various important indicator stratification results and various general indicator stratification results, thereby forming a variety of different second-level combinations;

[0035] According to the hierarchical fusion rule, all the second-level combinations are fused to obtain the stratified results of the spatial heterogeneity of cultivated land quality in the measured area.

[0036] According to a method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, after obtaining the spatial heterogeneity stratification results of cultivated land quality in the test area, based on the geographic detector q The rationality of the spatial heterogeneity stratification results of cultivated land quality is measured by the value, including:

[0037] Select the stratification results of the spatial heterogeneity of cultivated land quality as the independent variable and the cultivated land quality evaluation data as the dependent variable;

[0038] According to the factor detector in the geographic detector, a spatial differentiation coefficient of the dependent variable in the spatial heterogeneity stratification is calculated.

[0039] According to the spatial differentiation coefficient, it is judged whether the spatial heterogeneity stratification result of the cultivated land quality is reasonable.

[0040] According to the cultivated land quality spatial heterogeneity stratification method provided by the application, the fourth function expression of the spatial differentiation coefficient of the dependent variable in the spatial heterogeneity stratification according to the factor detector in the geographic detector is:

[0041] ;

[0042] Among them, q The spatial differentiation coefficient is; L The classification number of the spatial heterogeneity stratification result of the cultivated land quality is represented; k The classification index is; The number of spatial units contained in the classification k The total number of all spatial units in the to-be-tested region is represented; N The total number of all spatial units in the to-be-tested region is represented; The variance of the dependent variable in the classification k The variance of the dependent variable in the to-be-tested region is represented. The variance of the dependent variable in the to-be-tested region is represented.

[0043] The application also provides a cultivated land quality spatial heterogeneity stratification device, comprising the following modules:

[0044] An index data retrieval unit is configured to determine all monitoring indexes reflecting the cultivated land quality level of a to-be-tested region and obtain monitoring index data corresponding to each monitoring index;

[0045] An index weight calculation unit is configured to determine the index weight of each monitoring index based on a classification principal component analysis of the monitoring index data of the monitoring indexes;

[0046] A monitoring index classification unit is configured to label the importance label of each monitoring index according to the comparison result of the index weight of each monitoring index and an index importance degree threshold value; the index importance degree threshold value is determined based on the mean and standard deviation of all index weights, and the importance label has at least two types;

[0047] An index stratification fusion unit is configured to perform stratification fusion on monitoring indexes labeled with the same type of importance label respectively, to obtain stratification fusion results;

[0048] A stratification result output unit is configured to perform spatial overlay analysis on all stratification fusion results, to obtain the spatial heterogeneity stratification result of the cultivated land quality of the to-be-tested region.

[0049] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-described methods for stratifying spatial heterogeneity of cultivated land quality is implemented.

[0050] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for stratifying spatial heterogeneity of cultivated land quality.

[0051] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for stratifying spatial heterogeneity of cultivated land quality.

[0052] The present invention provides a method, device, equipment and storage medium for stratifying spatial heterogeneity of cultivated land quality, which takes into account indicator weights. The method calculates indicator weights through classification principal component analysis and then divides important indicators and general indicators. After stratifying the two separately, spatial heterogeneity stratification is performed according to the stratification fusion rules. This highlights the contribution of important indicators to cultivated land quality, reduces the interference of general indicators on cultivated land quality, and ensures the rationality and accuracy of spatial stratification of cultivated land quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 This is one of the flow charts of the method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention.

[0055] Figure 2 This is the second flow chart of the method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention.

[0056] Figure 3 It is a schematic diagram of the stratification results of spatial heterogeneity of cultivated land quality provided by the present invention.

[0057] Figure 4 It is a structural schematic diagram of the cultivated land quality spatial heterogeneity stratification device provided by the present invention.

[0058] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0059] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0060] It should be noted that, in the description of the present invention, the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, the phrase "comprises a..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising the elements. Terms such as "upper" and "lower" indicate positions or relationships based on those shown in the accompanying drawings and are intended solely to facilitate the description of the present invention and simplify the description. They are not intended to indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation, and are therefore not to be construed as limitations on the present invention. Unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be broadly construed, for example, to mean fixed, removable, or integral; mechanical or electrical; direct or indirect through an intermediary; or internal communication between two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] The terms "first," "second," and so forth, used herein are used to distinguish similar objects, not to describe a specific order or precedence. It should be understood that such terms are interchangeable where appropriate, allowing embodiments of the present invention to be implemented in an order other than that illustrated or described herein. Furthermore, the terms "first," "second," and so forth generally distinguish objects of a single type, and do not limit the number of objects. For example, the first object may be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the connected objects.

[0062] The following combination Figure 1-Figure 5The present invention describes the method, device, equipment and storage medium for stratifying the spatial heterogeneity of cultivated land quality. The method aims to calculate the indicator weights of each monitoring indicator reflecting the quality level of cultivated land in the measured area, label the importance of the monitoring indicators according to their importance, and then use different methods to spatially stratify and fuse important indicators and general indicators of different importance levels, so as to achieve efficient, fast and accurate stratification of the spatial heterogeneity of cultivated land quality.

[0063] Figure 1 This is one of the flow charts of the method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps:

[0064] Step 101: determine all monitoring indicators that reflect the quality level of cultivated land in the area to be measured, and obtain monitoring indicator data corresponding to each monitoring indicator.

[0065] The area to be tested refers to a specific geographical area where a stratified analysis of spatial heterogeneity of cultivated land quality is required, which can be a piece of farmland, cultivated land within a certain administrative division, etc.

[0066] The quality level of arable land refers to the comprehensive degree to which arable land can meet the sustainable output and quality safety of agricultural products, which is usually determined by factors such as arable land fertility, soil health and field infrastructure.

[0067] Monitoring indicators refer to various factors or parameters used to characterize the quality level of cultivated land, such as soil organic matter content, pH value, soil bulk density, effective soil layer thickness, plow layer texture, obstacle factors, total nitrogen content, available phosphorus, available potassium, irrigation capacity, drainage capacity, etc.

[0068] The monitoring indicator data corresponding to each monitoring indicator refers to the specific value of each monitoring indicator obtained through measurement, statistics, etc. These monitoring indicator data can be stored in a table format, for example, each row represents a monitoring point or sample, and each column represents a monitoring indicator.

[0069] Specifically, in step 101, the scope and characteristics of the area to be measured can be determined based on the research objectives and needs. For example, a specific area of ​​cultivated land within a certain administrative region can be selected based on the "Grades of Cultivated Land Quality" (GB / T 33469-2016), the actual conditions of the cultivated land in the study area, and existing research results. For example, 13 indicators can be selected, including the degree of farmland forest network, slope, effective soil layer thickness, cultivated layer texture, soil barriers, soil bulk density, pH, organic matter, total nitrogen, available phosphorus, available potassium, irrigation capacity, and drainage capacity.

[0070] Finally, through field sampling, statistical surveys, remote sensing monitoring, or consulting relevant data, the specific data values ​​of each monitoring indicator at different locations or on different cultivated land patches within the measured area are obtained. For example, in the cultivated land quality monitoring of a certain administrative region, the corresponding monitoring indicator data is collected and saved as an Excel spreadsheet or CSV file and stored in a database or file system.

[0071] Step 102: Determine the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator.

[0072] Categorical principal component analysis is a multivariate statistical analysis method used to extract the main components of data while taking into account categorical variables, in order to achieve the purpose of data dimensionality reduction and feature extraction. This method converts categorical variables into numerical scores by introducing optimal quantization technology, allowing them to participate in analysis like numerical variables, ultimately revealing the data structure, relationships between variables, and case distribution patterns in a low-dimensional space. In the spatial heterogeneity stratification method of cultivated land quality provided by the present invention, categorical principal component analysis is used to extract the main components of monitoring indicator data. Its specific implementation steps may include:

[0073] (1) Standardize the monitoring indicator data to eliminate the differences in dimensions and magnitudes between different monitoring indicator data. The standardization calculation formula used can be:

[0074] (1)

[0075] in, For the i The first sample j Monitoring indicator data of each monitoring indicator; For the j The mean value of the monitoring indicator data of each monitoring indicator; For the j The standard deviation of the monitoring indicator data of each monitoring indicator; The first i The monitoring indicator data of the jth monitoring indicator of samples.

[0076] (2) Statistical Product and Service Solutions (SPSS) software can be used to perform categorical principal component analysis on the standardized monitoring indicator data to obtain the results of the categorical principal component analysis, such as the load of each monitoring indicator on each principal component, the eigenvalue of each principal component, and the variance contribution rate.

[0077] (3) Based on the results of the classification principal component analysis, the indicator weight of each monitoring indicator is calculated. The indicator weight reflects the relative importance of each monitoring indicator to cultivated land quality. The larger the indicator weight, the greater the impact of the monitoring indicator corresponding to the indicator weight on cultivated land quality. When calculating the indicator weight, the variance contribution rate of each principal component and the load of the monitoring indicator on each principal component are usually considered.

[0078] Step 103: label each monitoring indicator with an importance label based on the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold, wherein the indicator importance threshold is determined based on the mean and standard deviation of all indicator weights.

[0079] First, the mean of the indicator weights of all monitoring indicators can be calculated And standard deviation σ, the specific calculation formula is:

[0080] (2)

[0081] (3)

[0082] in, n is the number of monitoring indicators; For the i The indicator weight of each monitoring indicator.

[0083] Then, we can calculate the mean of the indicator weights and standard deviation Set the indicator importance threshold, for example, , k Is a positive integer.

[0084] Finally, the indicator weight of each monitoring indicator is compared with the set indicator importance threshold. If the indicator weight is greater than the indicator importance threshold, the monitoring indicator corresponding to the indicator weight is marked with an important indicator label; otherwise, it is marked with a general indicator label.

[0085] For example, in a study of cultivated land in a certain administrative region, the mean of the indicator weights is calculated , standard deviation . Then when setting k = 1, the indicator importance threshold is 0.0769 + 1 × 0.0194 = 0.0963. If the weights of organic matter and total nitrogen exceed this indicator importance threshold, they will be labeled as important indicators (i.e., identified as important indicators). However, the weights of available phosphorus, available potassium, irrigation capacity, and drainage capacity, which do not exceed this indicator importance threshold, will be labeled as general indicators (i.e., identified as general indicators).

[0086] In step 104, monitoring indicators marked with the same type of importance labels are fused in layers to obtain layered fusion results. Generally speaking, layered fusion refers to integrating data of different types or layers according to certain rules to form a more comprehensive and representative data layer.

[0087] In the present invention, there are at least two types of importance labels. For example, importance labels include important indicator labels and general indicator labels. After all monitoring indicators are divided into important indicators and general indicators in step 103, the present invention provides implementation steps for hierarchical fusion of monitoring indicators of different importance levels, including but not limited to:

[0088] For monitoring indicators labeled as important indicators, each important indicator is graded according to relevant standards, specifications, or professional knowledge to obtain important indicator stratification results. For example, the grades of organic matter and total nitrogen are divided into three levels: high, medium, and low.

[0089] For monitoring indicators labeled with general indicator labels, appropriate clustering methods are used to perform spatial clustering and stratification to obtain general indicator stratification results. The choice of clustering method can be determined based on the characteristics of the data and the purpose of analysis, such as second-order clustering, k-means clustering, etc.

[0090] Step 105: Perform spatial overlay analysis on all the stratified fusion results to obtain stratified results of spatial heterogeneity of cultivated land quality in the measured area. Spatial overlay analysis is a core spatial analysis method in Geographic Information Systems (GIS). It refers to the process of fusing the geometric and attribute information of two or more spatial layers into a new layer through overlay operations and extracting new information based on spatial positional relationships. Essentially, it reveals the interaction patterns between different geographic elements through spatial logical operations (such as intersection, union, and difference) or attribute association calculations.

[0091] Specifically, the stratification results of the important indicators and the stratification results of the general indicators obtained in step 104 can be spatially overlaid according to a preset stratification fusion rule to obtain a variety of different level combinations. Then, all level combinations are fused according to the preset stratification fusion rule to obtain the stratification results of the spatial heterogeneity of cultivated land quality in the measured area. For example, a combination of high and high can be determined as the highest level, a combination of low and low can be determined as the lowest level, and other combinations can be determined as medium levels based on specific circumstances.

[0092] The method for stratifying the spatial heterogeneity of cultivated land quality provided by the present invention provides a method for stratifying the spatial heterogeneity of cultivated land quality that takes into account indicator weights. The method calculates the indicator weights through classification principal component analysis and then divides the important indicators and general indicators. After the two are stratified separately, spatial heterogeneity stratification is performed according to the stratification fusion rules. This highlights the contribution of important indicators to cultivated land quality, reduces the interference of general indicators on cultivated land quality, and ensures the rationality and accuracy of the spatial stratification of cultivated land quality.

[0093] After obtaining the monitoring indicator data of all monitoring indicators reflecting the quality level of cultivated land in the measured area based on the method provided in the above embodiment, the indicator weight of each monitoring indicator is determined based on the classification principal component analysis of the monitoring indicator data, which may specifically include but not be limited to:

[0094] Carry out dimension normalization processing on monitoring indicator data;

[0095] Perform classification principal component analysis on the monitoring indicator data after dimension normalization to obtain the principal component load, characteristic value and variance contribution rate of each principal component corresponding to each monitoring indicator;

[0096] According to the principal component loads and eigenvalues ​​of the principal components corresponding to the monitoring indicators, the linear combination coefficients of the principal components corresponding to each monitoring indicator are calculated;

[0097] Based on the linear combination coefficients of the principal components corresponding to all monitoring indicators and the variance contribution rates of the principal components, the score coefficient corresponding to each monitoring indicator is calculated;

[0098] The score coefficient corresponding to each monitoring indicator is normalized to determine the indicator weight of each monitoring indicator.

[0099] Figure 2 This is the second flow chart of the method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention, such as Figure 2 As shown in the figure, after obtaining all the monitoring indicator data of the cultivated land quality level in the measured area, the monitoring indicator data can be dimensionally normalized to eliminate the impact of the differences in dimensions and magnitudes between different monitoring indicators. The dimensional normalization methods used may include but are not limited to Z-score normalization, minimum-maximum normalization, or decimal scaling normalization.

[0100] For example, the transformation formula for transforming data into a specified range (usually [0, 1]) using minimum-maximum standardization is:

[0101] (4)

[0102] in, and Respectivelyj The minimum and maximum values ​​of the monitoring indicator data for each monitoring indicator; For the i The first sample j Monitoring indicator data of each monitoring indicator; is the first dimension obtained after dimension normalization. i The first sample j Monitoring indicator data of each monitoring indicator.

[0103] Categorical principal component analysis is based on principal component analysis. It considers the impact of categorical variables (such as different land use types and different administrative divisions) on the quality of cultivated land in the measured area and extracts the principal components that can reflect the main characteristics of the monitoring indicator data. The specific steps are as follows:

[0104] (1) Constructing a classification principal component analysis model: Select appropriate classification variables and incorporate them into the classification principal component analysis model.

[0105] (2) Use statistical analysis software (such as SPSS, R language, etc.) to perform classified principal component analysis on the monitoring indicator data after dimension normalization to obtain the principal component load, characteristic value and variance contribution rate of each principal component corresponding to each monitoring indicator.

[0106] (3) According to the principal component loads and eigenvalues ​​of the principal components corresponding to the monitoring indicators, the linear combination coefficients of the principal components corresponding to each monitoring indicator are calculated. Specifically, the following first function expression can be used for calculation:

[0107] (5)

[0108] in, For the i Monitoring indicators j The linear combination coefficients of the principal components; For the i Monitoring indicators j The principal component loadings of the principal components; For the j the eigenvalues ​​of the principal components; n is the number of monitoring indicators, ; m is the number of principal components, .

[0109] (4) Based on the linear combination coefficients of the principal components corresponding to the monitoring indicators and the variance contribution rates of the principal components, the score coefficient corresponding to each monitoring indicator is calculated. Specifically, the following second function expression can be used for calculation:

[0110] (6)

[0111] in, For the i The score coefficient corresponding to each monitoring indicator; For the j The variance contribution rate of the principal components.

[0112] (5) Normalize the score coefficients corresponding to each monitoring indicator to determine the indicator weight of each monitoring indicator. Normalization can be performed using a variety of methods, such as maximum-minimum normalization, sum normalization, and softmax normalization, which are not specifically limited in the present invention.

[0113] The indicator weight reflects the relative importance of each monitoring indicator to cultivated land quality. The larger the indicator weight, the greater the impact of the corresponding monitoring indicator on cultivated land quality. The normalized indicator weight reflects the relative importance of each monitoring indicator to cultivated land quality. The larger the normalized indicator weight, the greater the impact of the corresponding monitoring indicator on cultivated land quality.

[0114] The present invention provides a method for normalizing the score coefficient corresponding to each monitoring indicator and determining the indicator weight corresponding to each monitoring indicator. Specifically, the following third function expression can be used for calculation:

[0115] (7)

[0116] in, For the i The indicator weight corresponding to each monitoring indicator.

[0117] The method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention can achieve systematic analysis and processing of monitoring indicator data by determining the indicator weights of monitoring indicators through dimensional normalization processing, classification principal component analysis, calculation of linear combination coefficients, score coefficients and normalization processing. This scheme can effectively eliminate the influence caused by dimensional and order of magnitude differences between different monitoring indicators, making data processing more scientific and reasonable. At the same time, the use of classification principal component analysis combined with relevant calculation steps can fully explore the main features in the data and improve the accuracy and reliability of indicator weight calculation. In addition, this systematic analysis process has wide applicability, provides a more accurate weight basis for subsequent stratification of spatial heterogeneity of cultivated land quality, and enhances the scientific nature and effectiveness of the entire cultivated land quality monitoring system.

[0118] As an optional embodiment, the implementation steps of marking the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold include but are not limited to:

[0119] The importance labels of monitoring indicators whose indicator weights are greater than the above indicator importance thresholds are marked as important indicator labels;

[0120] The importance labels of monitoring indicators whose indicator weights are not greater than the indicator importance threshold are marked as general indicator labels.

[0121] Specifically, the present invention labels the importance of monitoring indicators by comparing their weights with their importance thresholds. Specifically, monitoring indicators whose weights exceed the importance threshold are labeled as important, while those whose weights do not exceed the importance threshold are labeled as general. This approach effectively distinguishes the degree of impact of monitoring indicators on cultivated land quality.

[0122] The indicator importance threshold is determined based on the mean and standard deviation of the indicator weights of all monitored indicators, usually set as the mean plus one standard deviation. This setting makes the threshold statistically reasonable and can dynamically adapt to different data distributions.

[0123] The labeled monitoring indicators are divided into two categories: important indicators and general indicators. Important indicators generally have a significant impact on cultivated land quality. For example, in the cultivated land quality monitoring of a certain administrative region, organic matter and total nitrogen are labeled as important indicators due to their high weights, indicating that they have a significant impact on cultivated land quality; while soil bulk density and the degree of farmland forest network are labeled as general indicators due to their low weights.

[0124] This weighted indicator labeling method has significant advantages. First, it effectively highlights the core role of key indicators in arable land quality evaluation, allowing subsequent analysis to focus on key factors, improving analytical efficiency and the accuracy of results. Second, by rationally categorizing indicators, it reduces the interference of general indicators in the analytical process, making the analysis more targeted.

[0125] It should be noted that in this paper, different stratification and fusion strategies are employed for key indicators and general indicators in the spatial heterogeneity stratification of cultivated land quality. Key indicators are stratified in detail according to professional standards, while general indicators are integrated and stratified using methods such as cluster analysis. Finally, the stratification results of the two types of indicators are spatially overlaid and fused to obtain the stratification results of spatial heterogeneity of cultivated land quality.

[0126] This approach not only improves the scientific nature of the placement of farmland quality monitoring points within the strata, but also enhances the efficiency and effectiveness of farmland quality monitoring, providing more accurate data support for farmland protection and utilization. By precisely distinguishing the importance of monitoring indicators, farmland quality assessment becomes more targeted and effective.

[0127] As an optional embodiment, after the labeled monitoring indicators are divided into two categories: important indicators and general indicators, the monitoring indicators labeled with the same importance labels are hierarchically integrated, specifically including but not limited to:

[0128] (1) Performing indicator stratification on each monitoring indicator whose importance label is marked as the important indicator label, and obtaining the indicator stratification result corresponding to each monitoring indicator; performing spatial overlay analysis on the indicator stratification results corresponding to all the monitoring indicators, and obtaining the spatial combination of various indicator stratification results, forming a plurality of different first-level combinations; performing fusion processing on all the first-level combinations according to the preset stratification fusion rules, and determining the important indicator stratification result.

[0129] (2) According to the preset number of clusters, cluster analysis is performed on each monitoring indicator whose importance label is marked as a general indicator label, and each monitoring indicator marked as a general indicator label is divided into multiple cluster categories; the average value of the comprehensive index of cultivated land quality of each cluster category is determined; according to the size of the average value of the comprehensive index of cultivated land quality, the cluster categories are divided into multiple preset levels according to the preset level division rules to obtain the general indicator stratification results.

[0130] refer to Figure 2 As shown, this invention utilizes distinct stratification and fusion approaches for key indicators and general indicators. The fusion and stratification of key indicators places greater emphasis on specialized domain knowledge and experience. This is because key indicators, such as soil fertility indicators (organic matter, total nitrogen, etc.), play a decisive role in arable land quality. Changes in these monitored indicators are directly related to the core characteristics of arable land quality. By leveraging specialized domain knowledge and experience to stratify these indicators, we can accurately capture their impact on arable land quality, ensuring that these key indicators occupy a central position in arable land quality evaluation and resulting in more reasonable and accurate evaluation results.

[0131] While general indicators are relatively less important, they still have a certain impact on cultivated land quality. By analyzing the data characteristics of these general indicators, we can objectively present their distribution and variation across different spatial units. This approach more accurately reflects the actual status of general indicators and provides supplementary information for comprehensive assessments of cultivated land quality. Therefore, the stratification method for general indicators is based more on analysis of their own data characteristics.

[0132] Using different stratification methods for important and general indicators can fully leverage their respective strengths, accurately characterizing key factors while also considering the influence of other factors, thereby comprehensively improving the accuracy of stratification of spatial heterogeneity in cultivated land quality. This stratification and integration approach, which combines professional knowledge and experience with data feature analysis, makes the analysis process more scientific and reasonable. It not only respects the natural laws and professional theories of cultivated land quality formation, but also fully utilizes the information inherent in the data, enhancing the scientific nature and reliability of the stratification results.

[0133] Specifically, for important indicators marked with important indicator labels, each important indicator can be graded based on relevant standards or professional field knowledge, including the following steps:

[0134] Step 1: Obtain the stratified results for each key indicator. For each monitoring indicator labeled "key indicator" (i.e., key indicator), determine its grading based on predefined grading criteria. For example, for the key indicator of organic matter, cultivated land is categorized as high (>20 g / kg), medium (10-20 g / kg), and low (<10 g / kg) based on its organic matter content.

[0135] Step 2: Perform a spatial overlay analysis on the stratified indicator results corresponding to all the monitoring indicators to determine the spatial combinations of the various stratified indicator results, forming a variety of first-level combinations. Specifically, Geographic Information System (GIS) software can be used to perform a spatial overlay analysis on the stratified indicator results for each important indicator to analyze the spatial relationships and interactions between different geographic spatial layers. This spatial overlay analysis can determine the spatial combinations of the various stratified indicator results, forming a variety of different level combinations (herein referred to as first-level combinations).

[0136] Step 3: According to the preset hierarchical fusion rules, the multiple first-level combinations obtained from the spatial overlay analysis are fused to obtain the final cultivated land quality levels corresponding to different first-level combinations as the important indicator stratification results. For example, the important indicator stratification results can include: high (all indicator stratification results are medium and above, and include at least one high), low (all indicator stratification results are medium and below, and include at least one low), and medium (other cases except high and low).

[0137] As another optional embodiment, for general indicators marked with general indicator labels, appropriate clustering methods are used to perform spatial clustering and stratification. The choice of clustering method can be determined according to the characteristics of the data and the purpose of analysis, such as using second-order clustering, k-means clustering, etc.

[0138] For example, second-order clustering is an exploratory cluster analysis method that can handle both continuous and categorical variables, is suitable for large datasets, and is robust to outliers. The second-order clustering analysis process is divided into pre-clustering and formal clustering. For each sample in the dataset, a cluster feature tree is constructed based on its variable values. The set of samples in the leaf nodes (the lowest level of the tree) forms preliminary subclusters. Based on the selected number of clusters, the subclusters formed by the pre-clustering are merged using the Bayesian Information Criterion to obtain the final clustering result.

[0139] Specifically in this embodiment, the number of clusters can be set to 3, and the clustering function of statistical analysis software (such as SPSS) is used to perform cluster analysis on general indicators, and the average value of the comprehensive cultivated land quality index corresponding to each cluster result is statistically analyzed.

[0140] Among them, the comprehensive index of cultivated land quality is a comprehensive indicator used to evaluate the quality level of cultivated land. It comprehensively considers the fertility of cultivated land, soil health, and field infrastructure. The comprehensive index of cultivated land quality is generally calculated by weighting the scores of each individual monitoring indicator to reflect the comprehensive degree to which cultivated land can meet the sustainable output and quality safety of agricultural products. The specific calculation process can be:

[0141] First, based on the "Grades of Cultivated Land Quality" (GB / T 33469-2016), the weights of each monitoring indicator are determined. This weighting is typically determined using methods such as the Analytic Hierarchy Process (AHP) or the Delphi method to consider the impact of each monitoring indicator on cultivated land quality.

[0142] Then, the membership degree of each monitoring indicator is calculated based on its measured value. The membership degree reflects the contribution of each monitoring indicator to the quality of cultivated land. The calculation methods include upward function, downward function, peak function, linear function and conceptual function.

[0143] Finally, the comprehensive index of cultivated land quality is calculated by weighted accumulation, that is, the sum of the weighted values ​​of all monitoring indicators (weight multiplied by the degree of membership) is calculated to obtain the corresponding comprehensive index of cultivated land quality.

[0144] In conjunction with the solution of the present invention, the three cluster categories are associated with the corresponding farmland quality comprehensive index data. By averaging the farmland quality comprehensive index within each cluster category, the average farmland quality comprehensive index corresponding to each cluster result can be obtained. Finally, the cluster results can be divided into three levels: high, medium, and low according to the average farmland quality comprehensive index, thus obtaining the general indicator stratification results.

[0145] The method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention can realize the stratified fusion of monitoring indicators marked with the same importance label, specifically including indicator stratification and spatial overlay analysis of important indicators, as well as cluster analysis and grade division of general indicators. This processing method can effectively distinguish monitoring indicators of different importance levels and adopt appropriate stratification strategies for processing them respectively. Detailed indicator stratification and spatial overlay analysis of important indicators can more accurately reflect their impact on cultivated land quality; cluster analysis and calculation of the average value of the cultivated land quality comprehensive index for general indicators can improve processing efficiency while ensuring analysis accuracy. By processing important indicators and general indicators separately, the interference of general indicators on cultivated land quality can be reduced, and the contribution of important indicators can be more highlighted, thereby improving the rationality and accuracy of the spatial heterogeneity stratification of cultivated land quality. This helps to more scientifically arrange cultivated land quality monitoring points within the stratification, improve the efficiency and accuracy of cultivated land quality monitoring, and provide more accurate data support for cultivated land protection and utilization.

[0146] After completing the hierarchical fusion of important indicators and general indicators to obtain the important indicator stratification results and the general indicator stratification results, the present invention obtains the spatial heterogeneity stratification results of cultivated land quality in the measured area by performing spatial overlay analysis on all the stratified fusion results, specifically including:

[0147] The important indicator stratification results and the general indicator stratification results are spatially overlaid and analyzed to obtain the spatial combination of various important indicator stratification results and various general indicator stratification results, forming a variety of different level combinations (herein referred to as second level combinations).

[0148] It should be noted that after performing the spatial overlay analysis of the important indicator stratification results and the general indicator stratification results, there is a certain possibility that the cultivated land patch fragmentation problem will occur, which will seriously affect the spatial heterogeneity stratification results, or the area of ​​the second-level combination will be particularly small, which is not conducive to spatial heterogeneity stratification.

[0149] In view of this, the present invention further performs fusion processing on all second-level combinations according to preset hierarchical fusion rules to obtain the final stratification result of the spatial heterogeneity of cultivated land quality in the measured area.

[0150] The results of layered fusion are generally presented in the form of spatial layers, each of which contains different levels of regional divisions. The results of the important indicator layer reflect the spatial distribution differences of key monitoring indicators, while the results of the general indicator layer show the combined impact of other monitoring indicators.

[0151] In this embodiment, a spatial overlay analysis is performed on the two stratified fusion results, that is, they are superimposed under the same geographic coordinate system for spatial analysis. The specific operation can be achieved through geographic information system (GIS) software. In GIS, the important indicator stratification results and the general indicator stratification results are used as two input layers, and the spatial overlay analysis tool is used for analysis. During the analysis, the software will generate a new combined layer based on the attribute value of each pixel in the two input layers, where the attribute value of each pixel is a combination of the attribute values ​​of the corresponding pixels in the two input layers. For example, if a certain area in the important indicator stratification results is high, and the same area in the general indicator stratification results is medium, then the combined attribute value of the area after the overlay analysis is "high-medium".

[0152] After spatial overlay analysis, a variety of different second-level combinations are obtained. These second-level combinations reflect the comprehensive performance of different regions in important indicators and general indicators. These second-level combinations can be fused according to the preset hierarchical fusion rules. The hierarchical fusion rules are formulated based on research objectives and practical experience, and are used to determine the final stratification results of cultivated land quality corresponding to different level combinations. For example, for the second-level combination "high-high", it is fused into the highest level; for the second-level combination "low-low", it is fused into the lowest level; and second-level combinations such as "high-medium" or "medium-high" are fused into the medium level. In this way, multiple second-level combinations are simplified and fused into a comprehensive stratification result of spatial heterogeneity of cultivated land quality.

[0153] The resulting spatial heterogeneity of cultivated land quality stratification intuitively demonstrates the spatial distribution differences in cultivated land quality within the measured region. High-grade areas indicate good cultivated land quality, while low-grade areas indicate areas where quality needs improvement. This information can be used to guide agricultural production and land management. For example, in high-grade areas, planting density can be appropriately increased to improve yields, while in low-grade areas, targeted soil improvement measures are needed to enhance cultivated land quality.

[0154] The method for stratifying the spatial heterogeneity of cultivated land quality provided by the present invention achieves an accurate characterization of the spatial heterogeneity of cultivated land quality by spatially overlaying the stratification results of important indicators and general indicators, and fusing them according to preset rules, providing a scientific basis for agricultural production layout and land resource management.

[0155] Based on the content of the above embodiment, as an optional embodiment, after obtaining the spatial heterogeneity stratification results of the cultivated land quality in the test area, the present invention can also be based on the geographic detector. q The rationality of the spatial heterogeneity stratification results of cultivated land quality is measured by the value, which specifically includes but is not limited to the following steps:

[0156] The results of spatial heterogeneity of cultivated land quality were selected as independent variables, and cultivated land quality evaluation data were selected as dependent variables. The independent variables were factors used to explain spatial heterogeneity, while the dependent variables were indicators used to measure spatial heterogeneity.

[0157] According to the factor detector in the geographic detector, the spatial differentiation coefficient of the dependent variable in the spatial heterogeneity layer is calculated. Specifically, the following fourth function expression can be used to perform this calculation:

[0158] (8)

[0159] in, q is the spatial differentiation coefficient; L The number of categories representing the stratification results of spatial heterogeneity of cultivated land quality; k For classification index; Representation classification k The number of spatial units contained in ; N Represents the total number of all spatial units in the area to be measured; Representation classification k The variance of the dependent variable in ; Represents the variance of the dependent variable in the measured area.

[0160] According to the spatial differentiation coefficient, it is judged whether the results of spatial heterogeneity of cultivated land quality are reasonable, that is, according to the calculated spatial differentiation coefficient ( q value), to judge whether the results of spatial heterogeneity stratification of cultivated land quality are reasonable. q The larger the value, the higher the spatial differentiation of the cultivated land quality spatial heterogeneity stratification results and the better the stratification effect. q The value is 0.74, indicating that the spatial heterogeneity stratification results of cultivated land quality have significant spatial differentiation, so it is determined that the spatial heterogeneity stratification results of cultivated land quality are reasonable.

[0161] The present invention can obtain the spatial heterogeneity stratification results of cultivated land quality based on the geographic detector. q The rationality of the stratification method was assessed using the appropriate values. As an effective spatial analysis tool, the GeoDetector can objectively quantify the spatial heterogeneity of stratification results. This scheme ensures the scientific and reliable nature of the stratification results and avoids errors in farmland quality assessment caused by unreasonable stratification. Through rationality assessment, the stratification method can be further optimized, improving the accuracy and efficiency of farmland quality monitoring and providing more reliable data support for farmland protection and utilization.

[0162] Taking a certain area to be measured as an example, the specific implementation of the method for stratifying spatial heterogeneity of cultivated land quality provided by the present invention is described in detail and completely, including but not limited to the following steps:

[0163] Step 1: Calculate indicator weights and divide indicators (i.e., label importance labels).

[0164] According to the "Quality Grades of Cultivated Land" (GB / T 33469-2016), based on the cultivated land conditions and existing research results in the tested area, and taking into account the availability of data, the degree of farmland forest network, slope, effective soil layer thickness, plow layer texture, obstacle factors, soil bulk density, pH, organic matter, total nitrogen, available phosphorus, available potassium, irrigation capacity and drainage capacity can be selected to reflect the quality level of cultivated land in the tested area.

[0165] SPSS software was used to perform principal component analysis on the standardized monitoring indicators to obtain the eigenvalues, component loadings and variance contribution rates of each principal component.

[0166] According to the method provided in the above embodiment, the indicator weight of each monitoring indicator is calculated, and the relevant data is recorded in Table 1 below.

[0167] Table 1 Index weights and index division results

[0168]

[0169] For example, as shown in Table 1, organic matter and total nitrogen have relatively high weights, at 0.106 and 0.105, respectively, indicating that organic matter and total nitrogen have a greater impact on arable land quality. Soil bulk density and the degree of farmland forest network have relatively low weights, at 0.043 and 0.048, respectively, indicating a relatively small impact on arable land quality.

[0170] According to the indicator weight calculation results in Table 1, the importance label of each monitoring indicator is marked using the method provided in the above embodiment:

[0171] Since the mean and standard deviation of all indicator weights are calculated to be 0.0769 and 0.0194 respectively, if we set k =1, the calculated indicator importance threshold is 0.0963. The indicator weights of organic matter and total nitrogen exceed the indicator importance threshold; if the k =2, the calculated indicator importance threshold is 0.1157, and all monitoring indicators do not exceed the indicator importance threshold. Therefore, we can select k = 1, then the two monitoring indicators, organic matter and total nitrogen, that exceed the indicator importance threshold are designated as important indicators, and the remaining indicators are general indicators, as detailed in Table 1. Furthermore, the two monitoring indicators, organic matter and total nitrogen, reflect the nutrient status of cultivated land and are therefore considered important indicators of cultivated land quality, consistent with professional knowledge.

[0172] Step 2: Perform hierarchical fusion of important indicators and general indicators.

[0173] For example, referring to the "Soil Nutrient Classification Standards for the Second National Soil Survey," combined with the cultivated land conditions in the surveyed area, the classification standards for two important indicators, organic matter and total nitrogen, can be determined as follows: the organic matter indicator stratification results are high (>20 g / kg), medium (10-20 g / kg), and low (<10 g / kg); the total nitrogen indicator stratification results are high (>1 g / kg), medium (0.75-1 g / kg), and low (<0.75 g / kg). Then, a spatial overlay analysis is performed on the organic matter and total nitrogen indicator stratification results. According to the important indicator stratification fusion rules, the important indicators are divided into high, medium, and low, resulting in the important indicator stratification results.

[0174] In addition, in this embodiment, general indicators include the degree of farmland forest network, slope, effective soil layer thickness, plough layer texture, obstacle factors, soil bulk density, pH, available phosphorus, available potassium, irrigation capacity and drainage capacity. General indicators can be divided into high, medium and low based on second-order clustering. Using the second-order clustering of SPSS software, the number of clusters is set to 3 categories, the clustering results are obtained, and the average value of the cultivated land quality comprehensive index of each category is calculated (the calculated average values ​​of the cultivated land quality comprehensive index of the three clusters are 0.744, 0.666 and 0.585 respectively). Then, according to the average value of the cultivated land quality comprehensive index, it can be divided into three levels of high, medium and low from large to small (that is, the general indicator stratification results are obtained).

[0175] Step 3: stratify the spatial heterogeneity of cultivated land quality.

[0176] Figure 3 is a schematic diagram of the stratification results of spatial heterogeneity of cultivated land quality provided by the present invention, such as Figure 3 As shown in the figure, the stratified results of important indicators and general indicators were spatially overlaid and analyzed. Based on the constructed hierarchical fusion rules, the nine different level combinations were fused into five levels: high (high-high), medium-high (including medium-high, high-medium), medium (including medium-medium, high-low, low-high), medium-low (including medium-low, low-medium), and low (including low-low).

[0177] Step 4: Conduct a rationality assessment of the results of spatial heterogeneity stratification of cultivated land quality.

[0178] Mainly based on geographic detectors q The value measures the differentiation of the dependent variable in the spatial heterogeneity layer. Geographic Detector is a statistical method used to detect and utilize spatial heterogeneity, which can reveal the causes and driving forces behind geographical phenomena. The factor detector in Geographic Detector uses q The value detects the spatial heterogeneity of the dependent variable and evaluates the extent to which the independent variables explain this heterogeneity.

[0179] In this embodiment, the cultivated land quality evaluation data of the tested area is used as a reference, and the q The rationality of quantitatively detecting the spatial heterogeneity of cultivated land quality by using the stratification of the three values.

[0180] The cultivated land quality evaluation data is based on the 13 monitoring indicators used in this study. The cultivated land quality grades are classified according to the "Cultivated Land Quality Grades" (GB / T 33469-2016) to obtain cultivated land quality grades from Grade 1 to Grade 10. Considering that the spatial heterogeneity of cultivated land quality in this example is stratified into five grades, the 10 cultivated land quality grades are sequentially merged two by two to obtain five cultivated land quality grades, designated Grade 1 to Grade 5, with lower grades indicating higher cultivated land quality.

[0181] The spatial heterogeneity of cultivated land quality is calculated based on the geographic detector (Formula (8) provided in the above embodiment), and the calculated q The value is 0.74, corresponding to p The value is 0, which is significant. The results show that the stratification results of the spatial heterogeneity of cultivated land quality in the tested area have obvious differentiation and good stratification effect.

[0182] Optionally, a comparative experiment of stratification of spatial heterogeneity of cultivated land quality and stratification of clustering of all indicators can be set up, in which the stratification of clustering of all indicators is based on the second-order clustering of all monitoring indicators, divided into 5 categories, and the average value of the comprehensive index of cultivated land quality corresponding to each cluster result is statistically analyzed. According to the numerical value of the average value of the comprehensive index of cultivated land quality, the cluster results are divided into 5 levels: high, medium-high, medium, medium-low and low. The cultivated land quality evaluation data of the tested area are selected as a reference, and the results of clustering of all indicators are quantitatively analyzed. The results of clustering of all indicators calculated based on the geographic detector are analyzed. q The value is 0.63 (corresponding to p The value is 0), which is higher than that of spatial heterogeneity stratification. q The value of 0.74 was reduced by 0.11. Therefore, it is fully proved that the stratification effect of spatial heterogeneity of cultivated land quality provided by the present invention is significantly better than the full index clustering stratification method.

[0183] According to the content of the above embodiment, a method for stratifying spatial heterogeneity of cultivated land quality taking into account indicator weights provided by the present invention is exemplified. The method calculates indicator weights through classification principal component analysis, uses the indicator importance threshold determined by the mean and standard deviation of all indicator weights, and compares the indicator weights of each monitoring indicator with the indicator importance threshold to divide all monitoring indicators into important indicators and general indicators. After the important indicators and general indicators are stratified separately, spatial heterogeneity stratification is performed according to the stratification fusion rules, which further highlights the contribution of important indicators to cultivated land quality, reduces the interference of general indicators on cultivated land quality, and ensures the rationality and accuracy of spatial stratification of cultivated land quality.

[0184] Figure 4 This is a schematic diagram of the structure of the device for stratifying spatial heterogeneity of cultivated land quality provided by the present invention. Figure 4 As shown, mainly including but not limited to:

[0185] The indicator data retrieving unit 41 is mainly used to determine all monitoring indicators reflecting the quality level of cultivated land in the area to be measured, and obtain the monitoring indicator data corresponding to each monitoring indicator;

[0186] The indicator weight calculation unit 42 is mainly used to determine the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator;

[0187] The monitoring indicator classification unit 43 is mainly used to label the importance label of each monitoring indicator based on the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold; the indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance label has at least two types;

[0188] The indicator hierarchical fusion unit 44 is mainly used to perform hierarchical fusion on monitoring indicators marked with the same type of importance labels to obtain hierarchical fusion results;

[0189] The stratification result output unit 45 is mainly used to perform spatial overlay analysis on all the stratification fusion results to obtain the stratification results of the spatial heterogeneity of the cultivated land quality in the area to be measured.

[0190] It should be noted that the device for stratifying spatial heterogeneity of cultivated land quality provided by the present invention can execute the method for stratifying spatial heterogeneity of cultivated land quality described in any of the above embodiments during specific operation, which will not be described in detail in this embodiment.

[0191] The spatial heterogeneity stratification device for cultivated land quality provided by the present invention calculates the indicator weights through classification principal component analysis and then divides the important indicators and general indicators. After stratifying the two separately, spatial heterogeneity stratification is performed according to the stratification fusion rules. This highlights the contribution of important indicators to cultivated land quality, reduces the interference of general indicators on cultivated land quality, and ensures the rationality and accuracy of the spatial stratification of cultivated land quality.

[0192] Figure 5 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 510 , a communications interface 520 , a memory 530 and a communication bus 540 , wherein the processor 510 , the communications interface 520 and the memory 530 communicate with each other via the communication bus 540 . The processor 510 can call the logic instructions in the memory 530 to execute the method for stratifying the spatial heterogeneity of cultivated land quality, which includes: determining all monitoring indicators reflecting the quality level of cultivated land in the area to be measured, and obtaining monitoring indicator data corresponding to each monitoring indicator; determining the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator; marking the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold; the indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance labels are of at least two types; performing layered fusion on the monitoring indicators marked with the same type of importance labels to obtain layered fusion results; performing spatial overlay analysis on all the layered fusion results to obtain the stratified results of the spatial heterogeneity of cultivated land quality in the area to be measured.

[0193] Furthermore, the logic instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0194] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the method for stratifying the spatial heterogeneity of cultivated land quality provided by the above-mentioned embodiments, the method including: determining all monitoring indicators reflecting the quality level of cultivated land in the measured area, and obtaining monitoring indicator data corresponding to each monitoring indicator; determining the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator; marking the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold; the indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance label has at least two types; performing layered fusion on the monitoring indicators marked with the same type of importance labels to obtain layered fusion results; performing spatial overlay analysis on all the layered fusion results to obtain the stratified results of the spatial heterogeneity of cultivated land quality in the measured area.

[0195] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the method for stratifying the spatial heterogeneity of cultivated land quality provided by the above-mentioned embodiments, the method comprising: determining all monitoring indicators reflecting the quality level of cultivated land in the area to be measured, and obtaining monitoring indicator data corresponding to each monitoring indicator; determining the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator; marking the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold; the indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance label has at least two types; performing layered fusion on the monitoring indicators marked with the same type of importance labels to obtain layered fusion results; performing spatial overlay analysis on all the layered fusion results to obtain the stratified results of the spatial heterogeneity of cultivated land quality in the area to be measured.

[0196] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0197] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0198] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for stratifying spatial heterogeneity of cultivated land quality, characterized by: include: Determine all monitoring indicators that reflect the quality level of cultivated land in the area to be tested, and obtain the monitoring indicator data corresponding to each monitoring indicator; Determining an indicator weight for each of the monitoring indicators based on a classification principal component analysis of the monitoring indicator data of the monitoring indicators; According to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold, an importance label of each monitoring indicator is marked; The indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance labels are of at least two types; Perform hierarchical fusion on monitoring indicators marked with the same type of importance labels to obtain hierarchical fusion results; Performing spatial overlay analysis on all the stratified fusion results to obtain stratified results of spatial heterogeneity of cultivated land quality in the tested area; The step of marking the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold includes: Marking the importance label of the monitoring indicator whose indicator weight is greater than the indicator importance threshold as an important indicator label; Marking the importance label of the monitoring indicator whose indicator weight is not greater than the indicator importance threshold as a general indicator label; The monitoring indicators marked with the same type of importance labels are respectively subjected to hierarchical fusion to obtain hierarchical fusion results, including: Performing indicator stratification on each monitoring indicator whose importance label is marked as the important indicator label, and obtaining an indicator stratification result corresponding to each monitoring indicator; performing spatial overlay analysis on the indicator stratification results corresponding to all the monitoring indicators, and obtaining spatial combinations of various indicator stratification results to form a plurality of different first-level combinations; performing fusion processing on all the first-level combinations according to a preset stratification fusion rule to determine an important indicator stratification result; According to the preset number of clusters, cluster analysis is performed on each monitoring indicator whose importance label is marked as the general indicator label, and each monitoring indicator marked as the general indicator label is divided into multiple cluster categories; the average value of the comprehensive index of cultivated land quality of each cluster category is determined; according to the size of the average value of the comprehensive index of cultivated land quality, according to the preset level division rules, the cluster categories are divided into multiple preset levels to obtain the general indicator stratification results.

2. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 1, characterized in that: The determining of the indicator weight of each monitoring indicator based on the classification principal component analysis of the monitoring indicator data of the monitoring indicator specifically includes: Performing dimension normalization processing on the monitoring indicator data; Performing a principal component analysis on the dimensionally normalized monitoring indicator data to obtain the principal component loads, eigenvalues, and variance contribution rates of each principal component corresponding to each monitoring indicator; Calculate the linear combination coefficient of each principal component corresponding to each monitoring indicator according to the principal component loads and the eigenvalues ​​of each principal component corresponding to the monitoring indicator; Calculate the score coefficient corresponding to each monitoring indicator based on the linear combination coefficient of each principal component corresponding to all the monitoring indicators and the variance contribution rate of each principal component; The score coefficient corresponding to each monitoring indicator is normalized to determine the indicator weight of each monitoring indicator.

3. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 2, characterized in that: The first function expression for calculating the linear combination coefficient of each principal component corresponding to each monitoring indicator according to the principal component loads and the eigenvalues ​​of each principal component corresponding to the monitoring indicator is: ； in, For the i Monitoring indicators j The linear combination coefficients of the principal components; For the i Monitoring indicators j The principal component loadings of the principal components; For the j the eigenvalues ​​of the principal components; n is the number of monitoring indicators, ; m is the number of principal components, .

4. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 3, characterized in that: The second function expression for calculating the score coefficient corresponding to each monitoring indicator based on the linear combination coefficient of each principal component corresponding to all the monitoring indicators and the variance contribution rate of each principal component is: ； in, For the i The score coefficient corresponding to each monitoring indicator; For the j The variance contribution rate of the principal components.

5. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 4, characterized in that: The score coefficient corresponding to each monitoring indicator is normalized to determine the indicator weight of each monitoring indicator. The third function expression is: ； in, For the i The indicator weight corresponding to each monitoring indicator.

6. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 1, characterized in that: The performing of spatial overlay analysis on all the stratified fusion results to obtain the stratified results of spatial heterogeneity of cultivated land quality in the tested area includes: Performing spatial overlay analysis on the important indicator stratification results and the general indicator stratification results to obtain spatial combinations of various important indicator stratification results and various general indicator stratification results, thereby forming a variety of different second-level combinations; According to the hierarchical fusion rule, all the second-level combinations are fused to obtain the stratified results of the spatial heterogeneity of cultivated land quality in the measured area.

7. The method for stratifying spatial heterogeneity of cultivated land quality according to any one of claims 1 to 6, characterized in that: After obtaining the spatial heterogeneity stratification results of the cultivated land quality in the test area, the geographical detector is used to q The rationality of the spatial heterogeneity stratification results of cultivated land quality is measured by the value, including: Select the stratification results of the spatial heterogeneity of cultivated land quality as the independent variable and the cultivated land quality evaluation data as the dependent variable; Calculating the spatial differentiation coefficient of the dependent variable in the spatial heterogeneity layer according to the factor detector in the geographic detector; According to the spatial differentiation coefficient, it is judged whether the result of the spatial heterogeneity stratification of cultivated land quality is reasonable.

8. The method for stratifying spatial heterogeneity of cultivated land quality according to claim 7, characterized in that: The fourth functional expression for calculating the spatial differentiation coefficient of the dependent variable in the spatial heterogeneity layer according to the factor detector in the geographic detector is: ； in, q is the spatial differentiation coefficient; L The number of categories representing the stratification results of spatial heterogeneity of cultivated land quality; k For classification index; Representation classification k The number of spatial units contained in ; N represents the total number of all spatial units in the area to be measured; Representation classification k The variance of the dependent variable in ; Represents the variance of the dependent variable in the measured area.

9. A device for stratifying spatial heterogeneity of cultivated land quality, characterized in that: include: An indicator data retrieval unit is used to determine all monitoring indicators reflecting the quality level of cultivated land in the area to be measured, and obtain monitoring indicator data corresponding to each monitoring indicator; An indicator weight calculation unit, configured to determine an indicator weight of each monitoring indicator based on a classification principal component analysis of the monitoring indicator data of the monitoring indicator; A monitoring indicator classification unit, configured to mark the importance label of each monitoring indicator according to a comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold; The indicator importance threshold is determined based on the mean and standard deviation of all indicator weights, and the importance labels are of at least two types; The indicator hierarchical fusion unit is used to perform hierarchical fusion on monitoring indicators marked with the same type of importance labels to obtain hierarchical fusion results; A stratification result output unit is used to perform spatial overlay analysis on all the stratification fusion results to obtain a stratification result of spatial heterogeneity of cultivated land quality in the measured area; The monitoring indicator classification unit marks the importance label of each monitoring indicator according to the comparison result of the indicator weight of each monitoring indicator and the indicator importance threshold, specifically including: Marking the importance label of the monitoring indicator whose indicator weight is greater than the indicator importance threshold as an important indicator label; Marking the importance label of the monitoring indicator whose indicator weight is not greater than the indicator importance threshold as a general indicator label; The indicator hierarchical fusion unit performs hierarchical fusion on monitoring indicators marked with the same type of importance labels to obtain hierarchical fusion results, specifically including: Performing indicator stratification on each monitoring indicator whose importance label is marked as the important indicator label, and obtaining an indicator stratification result corresponding to each monitoring indicator; performing spatial overlay analysis on the indicator stratification results corresponding to all the monitoring indicators, and obtaining spatial combinations of various indicator stratification results to form a plurality of different first-level combinations; performing fusion processing on all the first-level combinations according to a preset stratification fusion rule to determine an important indicator stratification result; According to the preset number of clusters, cluster analysis is performed on each monitoring indicator whose importance label is marked as the general indicator label, and each monitoring indicator marked as the general indicator label is divided into multiple cluster categories; the average value of the comprehensive index of cultivated land quality of each cluster category is determined; according to the size of the average value of the comprehensive index of cultivated land quality, according to the preset level division rules, the cluster categories are divided into multiple preset levels to obtain the general indicator stratification results.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for stratifying spatial heterogeneity of cultivated land quality as described in any one of claims 1 to 8 is implemented.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for stratifying spatial heterogeneity of cultivated land quality as claimed in any one of claims 1 to 8 is implemented.

12. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for stratifying spatial heterogeneity of cultivated land quality as claimed in any one of claims 1 to 8 is implemented.

Images