Soil main nutrient content estimation method, system and equipment

Abstract

The invention discloses a soil main nutrient content estimation method, system and equipment, and belongs to the technical field of soil nutrient detection. The method comprises the following steps: acquiring spectral image data of a target area, and preprocessing the spectral image data; collecting soil sample data of a target area, and preprocessing the spectral image data; determining the main nutrient content of the soil in the target area, performing correlation analysis on the preprocessed spectral image data and soil sample data and the main nutrient content of the soil, and screening a characteristic wave band of the nutrient content of the soil; based on the screened characteristic wave bands, constructing a multiple linear regression model of the soil nutrient content; and estimating the soil nutrient content of the to-be-detected area by using the multiple linear regression model, and making a soil nutrient content distribution map by using an estimation result. And a solution is provided for realizing rapid and accurate estimation of the content of main nutrients.

Description

Technical Field

[0001] The present invention relates to the technical field of soil nutrient detection, and more particularly to a method, system and equipment for estimating the main nutrient content of soil. Background Art

[0002] In agriculture and environmental science, the content of soil nutrients is crucial to plant growth and soil health. With the increasing attention to sustainable agriculture and environmental protection, the scientific community has paid more and more attention to the quantitative estimation of major soil nutrients (such as nitrogen, phosphorus, potassium, etc.), whose content directly affects the yield and quality of crops. However, traditional soil nutrient determination methods, such as chemical analysis, although highly accurate, are complex to operate, time-consuming and costly, and are difficult to meet the needs of large-scale, rapid and real-time soil nutrient monitoring.

[0003] In recent years, spectral technology, especially multispectral and hyperspectral technology, has been widely used in soil analysis due to its non-invasiveness and ability to quickly obtain data. However, existing multispectral research mainly focuses on the estimation models of soil salt content, water content and heavy metals, while research on the content of major soil nutrients is relatively lacking, making it difficult to meet the needs of large-scale, rapid and real-time soil nutrient monitoring.

[0004] Therefore, how to provide a method, system and equipment for estimating the main nutrient content in soil that can solve the above problems is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method, system and equipment for estimating the main nutrient content in soil, which uses multispectral technology to establish a quantitative estimation model for soil nutrient content, thereby achieving rapid and accurate estimation of the main nutrient content in soil, such as nitrogen, phosphorus and potassium.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for estimating the main nutrient content of soil comprises the following steps:

[0008] Collecting spectral image data of the target area and preprocessing the spectral image data;

[0009] Collect soil sample data of the target area and pre-process the spectral image data;

[0010] Determine the main nutrient content of the soil in the target area, and perform correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content;

[0011] Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content was constructed;

[0012] The soil nutrient content of the tested area was estimated using a multiple linear regression model, and the estimated results were used to produce a soil nutrient content distribution map.

[0013] Preferably, the collecting spectral image data of the target area and preprocessing the spectral image data includes:

[0014] Use drones to collect spectral image data of the target area;

[0015] The spectral image data is subjected to radiance conversion, reflectance conversion, geometric correction, atmospheric correction, mosaicking and clipping processing.

[0016] Preferably, the collecting of soil sample data of the target area and preprocessing of the spectral image data include:

[0017] Use a ground-based spectrometer to collect soil sample data in the target area;

[0018] The spectral image data is subjected to whiteboard correction and spectral reflectance curve optimization processing.

[0019] Preferably, the main nutrient content of the soil in the target area is determined, and the pre-processed spectral image data and soil sample data are subjected to correlation analysis with the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content, including:

[0020] Determine the main nutrient content of the soil in the target area, wherein the main nutrient content of the soil includes: nitrogen, phosphorus and potassium;

[0021] Extract each band data from the preprocessed spectral image data;

[0022] Integrate each band data with the main nutrient content of the soil to form a band data set;

[0023] The correlation between each band and soil nutrient content was calculated using the Pearson correlation coefficient;

[0024] The correlation is compared with a preset correlation coefficient threshold, and the band whose absolute value of the correlation coefficient threshold is greater than the correlation is taken as the characteristic band of soil nutrient content.

[0025] Preferably, ENVI software or ArcGIS software is used to extract each band data from the preprocessed spectral image data.

[0026] Preferably, the Pearson correlation coefficient is used to calculate the correlation between each band and the soil nutrient content, including

[0027] The Pearson correlation coefficient was calculated to obtain the correlation between each band and the soil nutrient content. The expression is:

[0028]

[0029] In the formula, x i Represents the i-th observation value of a certain band of data; y i Indicates the i-th observation value of the main nutrient content of soil corresponding to a certain band of data, and are the means of the observations, and n represents the number of observations.

[0030] Preferably, based on the screened characteristic bands, a multiple linear regression model of soil nutrient content is constructed, including:

[0031] The selected characteristic bands are used as independent variables, where each characteristic band corresponds to an independent variable;

[0032] Soil nutrient content was used as the dependent variable;

[0033] The multiple linear regression model of soil nutrient content was constructed as follows:

[0034] y=β 0 +β 1 x 1 +β 2 x 2 +…+β n x n +∈;

[0035] In the formula, y represents the soil nutrient content, x 1 ,x 2 ,…,x n represents the spectral reflectance of the selected characteristic band, β 0 ,β 1 ,…,β n represents the parameters of the multiple linear regression model, and ∈ represents the random error term.

[0036] Preferably, the method further comprises: verifying the multiple linear regression model, and optimizing and adjusting the model according to the verification result to obtain an optimized multiple linear regression model.

[0037] A soil main nutrient content estimation system, comprising:

[0038] Data acquisition module: collects spectral image data and soil sample data of the target area;

[0039] Data processing module: pre-processing spectral image data and soil sample data;

[0040] Data screening module: determine the main nutrient content of the soil in the target area, and perform correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content;

[0041] Construction module: Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content is constructed;

[0042] Estimation output module: Use the multivariate linear regression model to estimate the soil nutrient content in the test area, and use the estimation results to produce a soil nutrient content distribution map.

[0043] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for estimating the main nutrient content in soil when executing the computer program.

[0044] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method, system and device for estimating the main nutrient content of soil, and the established multiple linear regression model is that the soil nutrient content is a linear combination of the spectral reflectance of multiple characteristic bands plus a constant term and a random error term. By solving the parameters of the multiple linear regression model, a quantitative relationship between the characteristic bands and the soil nutrient content can be established, thereby realizing a rapid estimation of the soil nutrient content. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0046] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0047] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The purpose of the present invention is to provide a method, system and equipment for estimating the main nutrient content of soil, the method comprising: collecting spectral image data of the target area and preprocessing the spectral image data; collecting soil sample data of the target area and preprocessing the spectral image data; determining the main nutrient content of the soil in the target area, and performing correlation analysis between the preprocessed spectral image data and soil sample data and the main nutrient content of the soil, screening the characteristic bands of the soil nutrient content; constructing a multivariate linear regression model of the soil nutrient content based on the screened characteristic bands; estimating the soil nutrient content of the test area using the multivariate linear regression model, and using the estimation results to produce a soil nutrient content distribution map. A solution is provided for realizing the rapid and accurate estimation of the main nutrient content of soil nitrogen, phosphorus, potassium, etc.

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] See also Figure 1 As shown, the embodiment of the present invention discloses a method for estimating the main nutrient content of soil, comprising the following steps:

[0052] Collect spectral image data of the target area and pre-process the spectral image data;

[0053] Collect soil sample data in the target area and pre-process the spectral image data;

[0054] Determine the main nutrient content of the soil in the target area, and conduct correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil to screen the characteristic bands of soil nutrient content;

[0055] Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content was constructed;

[0056] The soil nutrient content of the tested area was estimated using a multiple linear regression model, and the estimated results were used to produce a soil nutrient content distribution map.

[0057] In a specific embodiment, collecting spectral image data of a target area and preprocessing the spectral image data include:

[0058] Use drones to collect spectral image data of the target area;

[0059] Perform radiosity conversion, reflectance conversion, geometric correction, atmospheric correction, mosaicking and clipping on spectral image data.

[0060] Through the above preprocessing, noise and interference factors can be eliminated, the image quality can be improved, and a high-quality data foundation can be provided for the follow-up.

[0061] In a specific embodiment, collecting soil sample data of a target area and preprocessing the spectral image data include:

[0062] Use a ground-based spectrometer to collect soil sample data in the target area;

[0063] The spectral image data is subjected to whiteboard correction and spectral reflectance curve optimization processing.

[0064] Through the above processing, the errors and background noise of the ground spectrometer are eliminated, the accuracy of the spectral data is ensured, and a high-quality data foundation is provided for the follow-up.

[0065] In a specific embodiment, the main nutrient content of the soil in the target area is determined, and the pre-processed spectral image data and soil sample data are subjected to correlation analysis with the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content, including:

[0066] Determine the main nutrient content of the soil in the target area, including nitrogen, phosphorus and potassium;

[0067] Use ENVI software or ArcGIS software to extract each band data from the preprocessed spectral image data;

[0068] Integrate each band data with the main nutrient content of the soil to form a band data set;

[0069] The correlation between each band and soil nutrient content was calculated using the Pearson correlation coefficient;

[0070] The correlation is compared with a preset correlation coefficient threshold, and the band whose absolute value of the correlation coefficient threshold is greater than the correlation is taken as the characteristic band of soil nutrient content.

[0071] In a specific embodiment, correlation analysis is a statistical method for evaluating the strength and direction of the relationship between two or more variables. In soil nutrient estimation, characteristic bands can be screened by calculating the correlation coefficient between spectral reflectance and soil nutrient content.

[0072] Specifically, the Pearson correlation coefficient is used to calculate the correlation between each band and the soil nutrient content, including: calculating the Pearson correlation coefficient to obtain the correlation degree between each band and the soil nutrient content, the expression is:

[0073]

[0074] In the formula, x i Represents the i-th observation value of a certain band of data; y i Indicates the i-th observation value of the main nutrient content of soil corresponding to a certain band of data, and are the means of the observations, and n represents the number of observations.

[0075] More specifically, x=[x 1 ,x 2 ,…,x n ] represents the reflectance value vector of a certain band in the spectral image data, where x i is the reflectance value of this band at the i-th observation point.

[0076] y=[y 1 ,y 2 ,…,y n ] represents the measured value vector of soil nutrient content, where y i is with x i Corresponding soil nutrient content measurements.

[0077] represents the sample mean of x, that is:

[0078] represents the sample mean of y, that is

[0079] r represents the Pearson correlation coefficient, which indicates the degree of linear correlation between x and y. Specifically, when r>0, it indicates a positive correlation; when r<0, it indicates a negative correlation; and when r=0, it indicates no linear correlation. The closer |r| is to 1, the stronger the linear relationship; the closer |r| is to 0, the weaker the linear relationship.

[0080] In a specific embodiment, constructing a multiple linear regression model of soil nutrient content includes:

[0081] The selected characteristic bands are used as independent variables, where each characteristic band corresponds to an independent variable;

[0082] Soil nutrient content was used as the dependent variable;

[0083] The multiple linear regression model of soil nutrient content was constructed as follows:

[0084] y=β 0 +β 1 x 1 +β 2 x 2 +…+β n x n +∈;

[0085] In the formula, y represents the soil nutrient content, x 1 ,x 2 ,…,x n represents the spectral reflectance of the selected characteristic band, β0 ,β 1 ,…,β n represents the parameters of the multiple linear regression model, and ∈ represents the random error term.

[0086] Specifically, choose appropriate statistical analysis software, such as SPSS, R language, Python, etc. These software support the construction of multiple linear regression models.

[0087] Specifically, the multiple linear regression model is verified, and according to the verification result, the model is optimized and adjusted to obtain an optimized multiple linear regression model.

[0088] Specifically, the least squares method is used to fit the multivariate linear regression model and find the optimal regression coefficient (β value) and constant term (β 0 );

[0089] During the fitting process, the software will automatically calculate the coefficient, standard error, t value, P value and other statistics of each independent variable to evaluate the degree and significance of the influence of the independent variable on the dependent variable.

[0090] Specifically, significance test:

[0091] Check the overall significance of the model, usually through the F test. If the P value of the F test is less than the significance level (such as 0.05), the model is considered to be significant as a whole.

[0092] Check the significance of each independent variable, usually through a t-test. If the t-test P value of an independent variable is greater than the significance level, you may want to consider removing it from the model.

[0093] Specifically, goodness of fit evaluation:

[0094] The goodness of fit of the model is evaluated using indicators such as R-square (coefficient of determination). The closer the R-square is to 1, the better the model fits.

[0095] You can also look at the adjusted R-squared to take into account the effect of the number of independent variables in the model on the goodness of fit.

[0096] Specifically, residual analysis:

[0097] Check whether the residuals meet the assumptions of normal distribution and homogeneity of variance. This can be done by plotting a histogram and a scatter plot of the residuals.

[0098] If the residuals are significantly non-normal or heteroskedastic, it may be necessary to consider modifying the model.

[0099] See also Figure 2 As shown, a system for estimating the main nutrient content in soil comprises:

[0100] Data acquisition module: collects spectral image data and soil sample data of the target area;

[0101] Data processing module: pre-processing spectral image data and soil sample data;

[0102] Data screening module: determine the main nutrient content of the soil in the target area, and conduct correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil, and screen the characteristic bands of the soil nutrient content;

[0103] Construction module: Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content is constructed;

[0104] Estimation output module: Use the multivariate linear regression model to estimate the soil nutrient content in the test area, and use the estimation results to produce a soil nutrient content distribution map.

[0105] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a method for estimating the main nutrient content in soil when executing the computer program.

[0106] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0107] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for estimating the main nutrient content of soil, characterized in that: The following steps are involved: Collecting spectral image data of the target area and preprocessing the spectral image data; Collect soil sample data of the target area and pre-process the spectral image data; Determine the main nutrient content of the soil in the target area, and perform correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content; Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content was constructed; The soil nutrient content of the tested area was estimated using a multiple linear regression model, and the estimated results were used to produce a soil nutrient content distribution map.

2. A method for estimating the main nutrient content of soil according to claim 1, characterized in that: The collecting of spectral image data of the target area and preprocessing of the spectral image data include: Use drones to collect spectral image data of the target area; The spectral image data is subjected to radiance conversion, reflectance conversion, geometric correction, atmospheric correction, mosaicking and clipping processing.

3. A method for estimating the main nutrient content of soil according to claim 1, characterized in that: The collecting of soil sample data of the target area and preprocessing of the spectral image data include: Use a ground-based spectrometer to collect soil sample data in the target area; The spectral image data is subjected to whiteboard correction and spectral reflectance curve optimization processing.

4. A method for estimating the main nutrient content of soil according to claim 1, characterized in that: Determine the main nutrient content of the soil in the target area, and perform correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil, and screen the characteristic bands of soil nutrient content, including: Determine the main nutrient content of the soil in the target area, wherein the main nutrient content of the soil includes: nitrogen, phosphorus and potassium; Extract each band data from the preprocessed spectral image data; Integrate each band data with the main nutrient content of the soil to form a band data set; The correlation between each band and soil nutrient content was calculated using the Pearson correlation coefficient; The correlation is compared with a preset correlation coefficient threshold, and the band whose absolute value of the correlation coefficient threshold is greater than the correlation is taken as the characteristic band of soil nutrient content.

5. A method for estimating the main nutrient content of soil according to claim 4, characterized in that: Use ENVI software or ArcGIS software to extract each band data from the preprocessed spectral image data.

6. A method for estimating the main nutrient content of soil according to claim 4, characterized in that: The Pearson correlation coefficient was used to calculate the correlation between each band and soil nutrient content, including The Pearson correlation coefficient was calculated to obtain the correlation between each band and the soil nutrient content. The expression is: In the formula, x i Represents the i-th observation value of a certain band of data; y i Indicates the i-th observation value of the main nutrient content of soil corresponding to a certain band of data, and are the means of the observations, and n represents the number of observations.

7. A method for estimating the main nutrient content of soil according to claim 1, characterized in that: Based on the screened characteristic bands, a multivariate linear regression model of soil nutrient content was constructed, including; The selected characteristic bands are used as independent variables, where each characteristic band corresponds to an independent variable; Soil nutrient content was used as the dependent variable; The multiple linear regression model of soil nutrient content was constructed as follows: y=β0+β1x1+β2x2+…+β n x n +∈; In the formula, y represents the soil nutrient content, x1, x2,…, x n represents the spectral reflectance of the selected characteristic bands, β0, β1,…, β n represents the parameters of the multiple linear regression model, and ∈ represents the random error term.

8. A method for estimating the main nutrient content of soil according to claim 1, characterized in that: Also includes: The multiple linear regression model is verified, and based on the verification results, the model is optimized and adjusted to obtain the optimized multiple linear regression model.

9. A soil main nutrient content estimation system using the soil main nutrient content estimation method according to any one of claims 1 to 8, characterized in that: include: Data acquisition module: collects spectral image data and soil sample data of the target area; Data processing module: pre-processing spectral image data and soil sample data; Data screening module: determine the main nutrient content of the soil in the target area, and perform correlation analysis between the pre-processed spectral image data and soil sample data and the main nutrient content of the soil to screen the characteristic bands of the soil nutrient content; Construction module: Based on the selected characteristic bands, a multivariate linear regression model of soil nutrient content is constructed; Estimation output module: Use the multivariate linear regression model to estimate the soil nutrient content in the test area, and use the estimation results to produce a soil nutrient content distribution map.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for estimating the main nutrient content in soil as described in any one of claims 1 to 8 is implemented.

Images