Method and system for retrieving grass nitrogen content

Abstract

The application discloses a kind of grassland nitrogen content inversion method and system, it is related to the technical field of grassland vegetation quantitative remote sensing.The method comprises the following steps: obtaining the grassland measured hyperspectral data and environmental parameter data of target area, and the grassland measured hyperspectral data is preprocessed;According to the grassland measured hyperspectral data after preprocessing, the mapping relationship feature of high spectral-leaf area index in vegetation is extracted;According to environmental parameter data, leaf area index-nitrogen content is extracted;With leaf area index as connection, the extracted leaf area index-nitrogen content and high spectral-leaf area index are coupled, and the grassland nitrogen content is inverted.By the method disclosed in the application, the dynamic change of nitrogen in the process of plant growth can be better captured, and has higher stability and portability, to quickly and effectively invert the grassland nitrogen content, so as to stably and accurately extract plant nitrogen, improve the generalization ability of nitrogen extraction method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of quantitative remote sensing of grassland vegetation, and in particular to a method and system for retrieving grassland nitrogen content. BACKGROUND

[0002] Grassland is an important ecological resource worldwide, and its nitrogen content is an important indicator of grassland health and productivity, affecting plant growth, carbon and nitrogen cycles, and the overall function of the grassland ecosystem. Accurate monitoring of grassland nitrogen content is crucial for ecological environment management, agricultural production optimization, and grassland ecological restoration.

[0003] PROSAIL is a physical model that combines the leaf reflectance model PROSPECT and the vegetation canopy radiation transfer model SAIL to provide a systematic description of the vegetation reflectance process. The PROSPECT model simulates leaf-level reflectance and transmittance hyperspectra through leaf structure parameters and leaf biochemical characteristic parameters, and then simulates vegetation canopy reflectance hyperspectra in the SAIL model by combining vegetation canopy structure parameters and soil background parameters. The Decision Support System for Agrotechnology Transfer (DSSAT) integrates more than 40 crop growth models, simulates crop growth parameters according to the crop growth cycle, and combines crop genetic characteristics, management measures, climate, soil, and other parameters. It is widely used in yield prediction, risk assessment, climate change research, and other fields.

[0004] The method of nitrogen retrieval obtains remote sensing images and ground measured data, calculates various spectral indices, and analyzes the correlation between these indices and nitrogen content, selecting spectral indices highly correlated with nitrogen as model inputs. Then, a regression method is used to build a nitrogen estimation model, which is optimized and verified to ensure its accuracy. Finally, the optimal model is used to retrieve nitrogen in space and time based on remote sensing data, generating a nitrogen distribution map to support agricultural nitrogen management and decision-making.

[0005] However, this empirical relationship between multispectral features and plant nitrogen content established through regression analysis can quickly estimate plant nitrogen content, but its spatial and temporal portability is poor. Traditional vegetation radiation transfer physical models have higher spatial and temporal portability than empirical models, but do not include plant nitrogen as a model parameter, so they cannot extract plant nitrogen stably and accurately through inversion of traditional vegetation radiation physical models, resulting in the problem of high difficulty in plant nitrogen extraction. SUMMARY

[0006] In order to overcome the problem that the existing grassland nitrogen content detection technology cannot stably and accurately extract plant nitrogen, thereby causing great difficulty in plant nitrogen extraction, the present application provides a grassland nitrogen content inversion method and system, which can quickly and effectively invert the grassland nitrogen content, thereby stably and accurately extracting plant nitrogen and improving the generalization ability of the nitrogen extraction method.

[0007] To achieve the purpose of the present application, the present application adopts the following technical solutions:

[0008] A grassland nitrogen content inversion method, the method comprising the following steps:

[0009] Obtain grassland measured data and environmental parameter data of a target area, and preprocess the grassland measured data;

[0010] According to the preprocessed grassland measured hyperspectral data, extract the mapping relationship characteristics of hyperspectral-leaf area index in vegetation;

[0011] According to the environmental parameter data, extract the leaf area index-nitrogen content;

[0012] Couple the extracted leaf area index-nitrogen content and leaf area index with the leaf area index as the connection, and invert the grassland nitrogen content.

[0013] In the above technical solution, by preprocessing the obtained grassland measured hyperspectral data, the spectral changes caused by scattering in the grassland nitrogen content measurement process can be reduced, especially the spectral differences caused by different scattering levels can be eliminated, thereby standardizing the correlation between the spectrum and the data; and by coupling the extracted leaf area index-nitrogen content and hyperspectral-leaf area index with the extracted leaf area index as the connection, the grassland nitrogen content is inverted, which can better capture the dynamic changes of nitrogen in the plant growth process and has higher stability and portability, so as to quickly and effectively invert the grassland nitrogen content, thereby stably and accurately extracting plant nitrogen and improving the generalization ability of the nitrogen extraction method.

[0014] Further, the preprocessing process of the grassland measured data and the environmental parameter data comprises:

[0015] According to the obtained grassland measured data, divide the parameter range, and construct a vegetation canopy radiation transfer model;

[0016] According to the grassland measured data after parameter range division, adjust the basic parameters of the constructed vegetation canopy radiation transfer model, and generate a spectral reflectance database under different parameters by using the vegetation canopy radiation transfer model;

[0017] According to the spectral reflectance database under different parameters, the high spectral reflectance under different parameters is extracted by using the vegetation canopy radiation transfer model;

[0018] According to the high spectral reflectance extracted by the vegetation canopy radiation transfer model, the spectral correction processing is performed on the canopy spectral curve data in the obtained grassland measured data.

[0019] The grassland measured data includes leaf area index and vegetation canopy high spectral data, and the environmental parameter data includes meteorological data, soil data, field management data and field observation data.

[0020] Further, the spectral correction processing is performed on the canopy spectral curve data in the obtained grassland measured data, and the expression is:

[0021]

[0022] wherein, R sc (i,j) represents the scattering corrected reflectivity value of the i th sample at wavelength j, R(i,j) represents the measured original reflectivity value of the i th sample at wavelength j, mean(R(j)) represents the average value of the measured original reflectivity value at wavelength j, sd(R(i)) represents the standard deviation of the measured original reflectivity value at wavelength j, sd(S(j)) represents the standard deviation of the vegetation canopy radiation transfer model simulated spectral reflectivity value at j, and mean(S(j)) represents the average value of the vegetation canopy radiation transfer model simulated spectral reflectivity value at j.

[0023] Further, the expression of the constructed vegetation canopy radiation transfer model is:

[0024]

[0025] wherein, ρ c represents the canopy reflectivity, N represents the leaf structure parameter, Cab represents the chlorophyll content, Cw represents the water content, Cm represents the dry matter content, LAI represents the leaf area index, ALA represents the average leaf inclination angle, Hotspot represents the hotspot parameter, Psoil represents the soil brightness, rsoli represents the soil humidity, θ v represents the observation zenith angle, θ x represents the solar zenith angle, represents the relative azimuth angle between the sun and the observation.

[0026] In the above technical scheme, the spectral correction processing is performed on the canopy spectral curve data in the obtained grassland measured data, which can reduce the spectral change caused by scattering in the measurement process, especially eliminate the spectral difference caused by different scattering levels, so as to standardize the correlation between the spectrum and the data.

[0027] Further, the process of extracting the mapping relationship characteristics of the hyperspectral-leaf area index in the vegetation includes:

[0028] According to the spectral reflectance database under different parameters, the leaf area index under different parameters is extracted by using the vegetation canopy radiation transfer model;

[0029] The hyperspectral reflectance and leaf area index extracted by the vegetation canopy radiation transfer model are used as the training set, and the measured data of the grassland after spectral correction are used as the verification set;

[0030] The vegetation canopy radiation transfer model is optimized and trained for leaf area index inversion by using the training set, and the trained vegetation canopy radiation transfer model is verified by using the verification set, and the trained vegetation canopy radiation transfer model is obtained after verification;

[0031] The hyperspectral data in the input measured data of the grassland are inverted by using the trained vegetation canopy radiation transfer model, and the leaf area index of the vegetation is extracted.

[0032] Further, the process of optimizing and training the vegetation canopy radiation transfer model for leaf area index inversion by using the training set includes:

[0033] Based on the hyperspectral reflectance and leaf area index under different parameters in the training set, a plurality of rounds of optimization training are set to traverse different parameter combinations to optimize the parameters of the vegetation canopy radiation transfer model;

[0034] The vegetation canopy radiation transfer model inverses the leaf area index of the vegetation according to the hyperspectral reflectance and leaf area index under different parameters by using the partial least squares regression method;

[0035] The loss function is set to compare the inversion results of the leaf area index under different parameter combinations with the measured results, and the vegetation canopy radiation transfer model corresponding to the parameter combination with the minimum loss function value is used as the trained vegetation canopy radiation transfer model.

[0036] Further, the process of inversing the leaf area index of the vegetation by using the partial least squares regression method includes:

[0037] The partial least squares regression method is used to establish the relationship between the hyperspectral reflectance and the leaf area index, and the main features of the maximum co-variance in the hyperspectral reflectance and the leaf area index are extracted, and the expression is:

[0038] X=TP T +E

[0039] y=Tq+f

[0040] The leaf area index of the vegetation is inversely calculated according to the main features of the maximum co-variance in the hyperspectral reflectance and the leaf area index.

[0041] According to the measured leaf area index and the corrected hyperspectral data, the best parameter combination is selected, and the vegetation canopy radiation transfer model is parameter adjusted according to the parameter combination;

[0042] Wherein, X represents a hyperspectral reflectivity matrix, y represents a leaf area index vector, T represents a principal component matrix, P and q both represent regression coefficient matrices, E and f respectively represent residual matrices and error terms.

[0043] Further, the loss function expression set in the parameter optimization process of the vegetation canopy radiation transfer model is:

[0044]

[0045] Wherein, n represents the number of data points; y i represents the i-th real value; Y i represents the I-th predicted value; represents the sum of the square errors between the real value and the predicted value.

[0046] Further, the process of extracting the leaf area index-nitrogen content includes:

[0047] The agricultural technology transfer decision support system model is constructed, and the agricultural technology transfer decision support system model is adjusted according to different parameter range combinations of environmental parameter data;

[0048] Using the agricultural technology transfer decision support system model, the correlation characteristics between the leaf area index and the nitrogen content under different parameter range combinations are extracted, and the loss of each parameter range combination is calculated using the loss function RMSE. The parameter range combination with the minimum output value of the loss function RMSE is taken as the best parameter range combination;

[0049] The best parameter range combination is set as the new parameter of the agricultural technology transfer decision support system model for extracting the leaf area index-nitrogen content;

[0050] The vegetation canopy radiation transfer model and the agricultural technology transfer decision support system model are coupled by connecting the leaf area index, the correlation characteristics between the canopy spectrum and the leaf nitrogen content are extracted, the grassland nitrogen content is inversed, and the inversion result is output.

[0051] In the technical solution, the vegetation canopy radiation transfer model (PROSAIL model) is used to input hyperspectral data to retrieve the leaf area index (LAI) of the vegetation; the partial least squares regression (PLSR) is a kind of multivariate regression method used to establish the relationship between the independent variables (spectral reflectance) and the dependent variables (LAI), and is particularly suitable for the case where there is a multiple collinearity between the high-dimensional data and the independent variables; the PLSR extracts the information in the spectral reflectance data (independent variables) and the LAI (dependent variables) at the same time, and finds the principal components that can explain the maximum covariance of the two; the environmental parameters are input into the agricultural technology transfer decision support system model (DSSAT, Decision Support System for Agrotechnology Transfer) to simulate the change of the leaf area index and the leaf nitrogen content; the DSSAT model integrates the environmental and management data to simulate the crop growth process of the pasture, and the simulation result of the model needs to be verified according to the actual observation result, the model is adjusted according to the verification result, the coupling of the PROSAIL and DSSAT models is realized by first retrieving the LAI from the PROSAIL model through the remote sensing hyperspectral data, then simulating the mapping relationship between the leaf area index (LAI) and the nitrogen content by the DSSAT, and then finding the relationship between the canopy spectrum and the leaf nitrogen content by taking the LAI as the connection; in this way, the method for retrieving the nitrogen content based on the spectral data of the physical process of the spectral reflection and the physiological process of the plant growth is realized, so that the plant nitrogen is stably and accurately extracted, and the generalization ability of the nitrogen extraction method is improved.

[0052] A system for retrieving the nitrogen content of grassland, the system comprising:

[0053] An acquisition module is configured to acquire the measured data and the environmental parameter data of the grassland in a target area.

[0054] A processing module is configured to pre-process the measured hyperspectral data of the grassland.

[0055] A feature extraction module is configured to extract the mapping relationship between the hyperspectrum and the leaf area index in the vegetation according to the pre-processed measured hyperspectral data of the grassland.

[0056] The feature extraction module is further configured to extract the leaf area index-nitrogen content according to the environmental parameter data.

[0057] A grassland nitrogen content retrieval module is configured to couple the extracted leaf area index-nitrogen content and the hyperspectrum-leaf area index by taking the leaf area index as the connection, and retrieve the nitrogen content of the grassland.

[0058] Compared with the prior art, the present application has the following advantages:

[0059] The present application provides a method and system for retrieving grassland nitrogen content, which can reduce the spectral changes caused by scattering in the process of measuring grassland nitrogen content, especially eliminate the spectral differences caused by different scattering levels, so as to standardize the correlation between the spectrum and the data, by preprocessing the obtained grassland measured data; and the extracted leaf area index is used as a connection to couple the extracted leaf area index-nitrogen content and hyperspectral-leaf area index, so as to retrieve the grassland nitrogen content, which can better capture the dynamic changes of nitrogen in the process of plant growth, and has higher stability and portability, so as to quickly and effectively retrieve the grassland nitrogen content, thereby stably and accurately extracting the plant nitrogen, and improving the generalization ability of the nitrogen extraction method. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 A step flow chart of a method for retrieving grassland nitrogen content provided by the embodiment of the present application;

[0061] Figure 2 A specific flow chart of retrieving grassland nitrogen content provided by the embodiment of the present application;

[0062] Figure 3 A schematic diagram of retrieving LAI based on PLSR canopy spectrum provided by the embodiment of the present application;

[0063] Figure 4 A schematic diagram of simulated spectrum generated under different LAI provided by the embodiment of the present application;

[0064] Figure 5 A schematic diagram of leaf area index retrieval based on PROSAIL model provided by the embodiment of the present application;

[0065] Figure 6 A schematic diagram of the corresponding relationship between LAI and LeafN% in non-grass and grass simulated by DSSAT provided by the embodiment of the present application;

[0066] Figure 7 A schematic diagram of N retrieval result of DSSAT and PROSAIL coupled model provided by the embodiment of the present application;

[0067] Figure 8 A structure schematic diagram of a system for retrieving grassland nitrogen content provided by the embodiment of the present application. DETAILED DESCRIPTION

[0068] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the accompanying drawings. The preferred embodiments of the present application are shown in the drawings. However, the present application can be realized in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present application more thorough and comprehensive.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0070] Example 1:

[0071] This example provides a method for inverting the nitrogen content of grassland. Figure 1 , the method comprises the following steps:

[0072] Step S1: obtaining grassland measured data and environmental parameter data of a target area, and preprocessing the grassland measured data;

[0073] Step S2: extracting the mapping relationship features between the hyperspectral data of the vegetation and the leaf area index based on the pre-processed grassland measured hyperspectral data;

[0074] Step S3: extracting leaf area index-nitrogen content based on environmental parameter data;

[0075] Step S4: Using the leaf area index as a link, the extracted leaf area index-nitrogen content is coupled with the leaf area index to invert the grassland nitrogen content.

[0076] In step S1, the process of preprocessing the grassland measured data and environmental parameter data includes:

[0077] The parameter ranges were divided based on the obtained grassland measured data, and a vegetation canopy radiation transfer model was constructed;

[0078] The basic parameters of the constructed vegetation canopy radiation transfer model are adjusted according to the measured grassland data after parameter range division, and the spectral reflectance database under different parameters is generated using the vegetation canopy radiation transfer model.

[0079] Based on the spectral reflectance database under different parameters, the vegetation canopy radiation transfer model is used to extract the hyperspectral reflectance under different parameters;

[0080] According to the hyperspectral reflectance extracted from the vegetation canopy radiation transfer model, the canopy spectral curve data in the grassland measured data were spectrally corrected.

[0081] The grassland measured data includes leaf area index and vegetation canopy hyperspectral data, and the environmental parameter data includes meteorological data, soil data, field management data and field observation data.

[0082] Specifically, the canopy spectral curve data in the acquired grassland measured data is spectrally corrected, and the expression is:

[0083]

[0084] Among them, R sc (i, j) represents the reflectance value of the i-th sample after scattering correction at wavelength j, R(i, j) represents the measured original reflectance value of the i-th sample at wavelength j, mean(R(j)) represents the average value of the measured original reflectance value at wavelength j, sd(R(i)) represents the standard deviation of the measured original reflectance value at wavelength j, sd(S(j)) represents the standard deviation of the spectral reflectance value simulated by the vegetation canopy radiation transfer model at j, and mean(S(j)) represents the average value of the spectral reflectance value simulated by the vegetation canopy radiation transfer model at j.

[0085] The constructed vegetation canopy radiation transfer model (PROSAIL model) couples the leaf optical property model PROSPECT and the canopy reflectance model SAIL. The PROSPECT model simulates the reflectance and transmittance of vegetation leaves in the spectral range of 400nm to 2500nm. The PROSPECT model has two types of input parameters: the first type is the structural parameters of the leaves; the second type is the biochemical content parameters of the leaves, including chlorophyll content, water content and dry matter content. The SAIL model solves the scattering and absorption of four upward and downward radiation quantities to simulate the reflectance of the vegetation canopy. The SAIL model has two types of input parameters: the first type is the leaf reflectance and transmittance simulated by the PROSPECT model; the second type includes parameters such as leaf area index, mean leaf inclination, hot spot parameters, solar zenith angle, observation zenith angle and relative azimuth. The constructed vegetation canopy radiation transfer model is expressed as follows:

[0086]

[0087] Among them, ρ c represents canopy reflectance, N represents leaf structure parameters, Cab represents chlorophyll content, Cw represents water content, Cm represents dry matter content, LAI represents leaf area index, ALA represents average leaf inclination, Hotspos represents hot spot parameters, Psoil represents soil brightness, rsoil represents soil moisture, θ v represents the observation zenith angle, θ x represents the solar zenith angle, Indicates the relative azimuth between the sun and the observation.

[0088] It can be understood that spectral correction processing of the canopy spectral curve data in the acquired grassland measured data can reduce the spectral changes caused by scattering during the measurement process, especially eliminate the spectral differences caused by different scattering levels, thereby standardizing the correlation between the spectrum and the data, and reducing the influence of environmental factors on the canopy spectral curve measured in the field.

[0089] In step S2, see Figure 2 ,The process of extracting the mapping relationship features of vegetation hyperspectral ,and leaf area index includes:

[0090] S21: Based on the spectral reflectance database under different parameters, the vegetation canopy radiation transfer model is used to extract the leaf area index under different parameters;

[0091] S22: The hyperspectral reflectance and leaf area index extracted by the vegetation canopy radiation transfer model are used as the training set, and the spectrally corrected grassland measured data are used as the validation set;

[0092] S23: Optimize the leaf area index inversion of the vegetation canopy radiation transfer model using the training set, and verify the trained vegetation canopy radiation transfer model using the validation set. After the validation passes, the trained vegetation canopy radiation transfer model is obtained.

[0093] S24: Use the trained vegetation canopy radiation transfer model to invert the hyperspectral data in the input grassland measured data and extract the leaf area index of the vegetation.

[0094] Specifically, the range of PROSAIL model parameters is determined by measuring measured data and references. After setting the parameters, the PROSAIL model is used to generate a spectral reflectance database under different parameters. The PROSAIL model has numerous parameters, and the generated spectral curve is hyperspectral data, which can express complex reflectance spectral characteristics. To simplify the LAI inversion process based on the PROSAIL model, the present invention applies partial least squares regression to the spectral database generated based on the PROSAIL model. In this way, the LAI inversion relies on the physical model while also having the simple and fast characteristics of the empirical model. For specific observational data, the key to ensuring the accuracy of this inversion model is to determine the appropriate PROSAIL parameter range based on the measured data, among which the determination of the blade structure parameter N is particularly critical.

[0095] In step S23, the process of optimizing the leaf area index inversion of the vegetation canopy radiation transfer model using the training set includes:

[0096] Based on the hyperspectral reflectance and leaf area index under different parameters in the training set, several rounds of optimization training were set up to traverse different parameter combinations to optimize the parameters of the vegetation canopy radiation transfer model;

[0097] The vegetation canopy radiation transfer model uses the partial least squares regression method to invert the leaf area index of vegetation based on the hyperspectral reflectance and leaf area index under different parameters;

[0098] A loss function was set to compare the leaf area index inversion results under different parameter combinations with the measured results, and the vegetation canopy radiation transfer model corresponding to the parameter combination with the smallest loss function value was taken as the trained vegetation canopy radiation transfer model.

[0099] Furthermore, the process of inverting the leaf area index of vegetation using the partial least squares regression method includes:

[0100] The partial least squares regression method was used to establish the relationship between hyperspectral reflectance and leaf area index, and the main features with the largest covariance in hyperspectral reflectance and leaf area index were extracted. The expression is:

[0101] X=TP T +E

[0102] y=Tq+f

[0103] The leaf area index of vegetation is inverted based on the main features of the maximum covariance between hyperspectral reflectance and leaf area index.

[0104] The optimal parameter combination is selected based on the measured leaf area index and the corrected hyperspectral data, and the parameters of the vegetation canopy radiation transfer model are adjusted based on the parameter combination;

[0105] Where X represents the hyperspectral reflectance matrix, y represents the leaf area index vector, T represents the principal component matrix, P and q both represent the regression coefficient matrix, E and f represent the residual matrix and error term, respectively.

[0106] Specifically, partial least squares regression (PLSR) is a multivariate regression method used to establish the relationship between the independent variable (spectral reflectance) and the dependent variable (LAI). It is particularly well-suited for high-dimensional data and situations where multicollinearity exists between the independent variables. PLSR simultaneously extracts information from both the spectral reflectance data (independent variable) and the LAI (dependent variable) to identify the principal components that explain the greatest covariance between the two. These principal components represent the primary information in the spectrum and are closely correlated with LAI variations.

[0107] The core formula of PLSR can be expressed as:

[0108] X=TP T +E (3)

[0109] y=Tq+f(4)

[0110] Where X is the spectral reflectance matrix, y is the LAI value vector, T is the principal component matrix, P and q are the regression coefficient matrices, and E and f are the residual matrix and error term.

[0111] After the training is completed, for a new spectral reflectance data X new , the trained PLSR model can be used to predict its corresponding LAI value:

[0112] LAI pred =T new q(5)

[0113] Among them, T new is the principal component extracted from the new spectral data.

[0114] Furthermore, the loss function expression set in the process of parameter optimization of the vegetation canopy radiation transfer model is:

[0115]

[0116] Where n represents the number of data points; y i represents the i-th true value; Y i represents the I-th predicted value; It represents the sum of squared errors between the true value and the predicted value.

[0117] Specifically, the model construction process uses spectral data simulated by PROSAIL and input LAI as training data, and uses partial least squares regression (PLSR) to establish the relationship between spectral features and LAI. To ensure the accuracy of the model, the measured spectral data is used as input to obtain the inverted LAI, and the root mean square error (RMSE) with the measured LAI is calculated. By traversing the blade structure N in the input parameters of the PROSAIL model, the most appropriate parameter range for N is found based on the RMSE of the inversion result. Finally, the LAI inversion process based on the PROSAIL model and PLSR is completed. By inputting hyperspectral data into the model, the inverted LAI value can be output.

[0118]

[0119] Where: n is the number of data points; y i is the i-th true value; Y i is the I-th predicted value; It is the sum of the squared errors between the true and predicted values. RMSE is calculated by taking the mean of the squared errors and taking the square root. It reflects the average difference between the model's predicted values ​​and the true values. A smaller value indicates a more accurate model prediction.

[0120] The main role of the model is to obtain a LAI model based on hyperspectral characteristics. By inputting hyperspectral data, the LAI of vegetation is inversed, and the inversed LAI is used as an input parameter of the DSSAT model to simulate the change of nitrogen content. In this way, the PROSAIL model and the DSSAT model are coupled, and the inversion from hyperspectral data to nitrogen content is realized.

[0121] In step S3, referring to Figure 2 , the process of extracting the leaf area index-nitrogen content includes:

[0122] The agricultural technology transfer decision support system model is constructed, and the agricultural technology transfer decision support system model is adjusted according to different parameter range combinations of environmental parameter data;

[0123] According to different parameter range combinations, the agricultural technology transfer decision support system model is used to extract the correlation characteristics between the leaf area index and the nitrogen content, and the loss function RMSE is used to calculate the loss of the correlation characteristics of each parameter range combination. The parameter range combination with the minimum output value of the loss function RMSE is used as the best parameter range combination;

[0124] The best parameter range combination is set as a new parameter of the agricultural technology transfer decision support system model for extracting the leaf area index-nitrogen content;

[0125] The vegetation canopy radiation transfer model and the agricultural technology transfer decision support system model are coupled by taking the leaf area index as a connection, the correlation characteristics between the canopy spectrum and the leaf nitrogen content are extracted, the grassland nitrogen content is inversed, and the inversion result is output.

[0126] Specifically, the main purpose of the DSSAT model is to summarize various types of crop models, standardize the input and output modes of the model, and realize the simulation requirements under different growth conditions after inputting the corresponding soil, climate, crop variety parameters and field management data. DSSAT is not a general model, different models are developed for different crops, in order to simulate the growth of pasture, the Forages-Alfalfa model and the Bermuda Grass model are used in DSSAT to simulate the growth process of non-grass and grass pastures.

[0127] The DSSAT model input data mainly includes meteorological data, soil data, management data and observation data. Specifically: (1) Meteorological data: The minimum data set must include precipitation, maximum temperature, minimum temperature, and solar radiation to drive crop growth simulation. First, the collected data needs to be organized into a format that DSSAT can recognize and input using the meteorological input module Weather Data. The operations that need to be performed when importing data include: establishing a weather station, naming, inputting longitude and latitude, and finally generating a "WTH" file format for model runtime call. (2) Soil data: Soil properties are imported and stored in the model internal file using the soil input module soildata. Basic soil property information includes Clay%, Silt%, Stones%, Organic carbon%, pH in water, Cation exchange capacity cmol / kg, Total nitrogen%, and finally generating a "SOL" file format for model runtime call. (3) Field management data: Management data includes multiple parameters such as planting density, irrigation, fertilization, and harvesting. They are input and edited through the Xbuild interface of the Crop Management Data module. (4) Field observation data: After the field experiment is created, File A (End of Season) and File T (Time Course) are prepared through the Experimental Data module based on the time series data of leaf area index (LAI) and leaf nitrogen content (LeafN%) observed during the experiment. These data are input into the model and the crop parameters in the DSSAT model are adjusted locally (crop parameter files include .CUL, .ECO and .SPE) until the simulated LAI and Leaf N% are close to the observed LAI and Leaf N%.

[0128] The DSSAT model integrates environmental and management data to simulate the growth process of forage crops. The model simulation results need to be verified based on actual observations, and the model parameters are adjusted based on the verification results. The parameter adjustment process is to set different crop parameter range combinations, traverse each combination, and calculate the RMSE (RMSE) between the simulated values ​​of LAI and Leaf N% and the observed values. LAI and RMSE Leaf N％ ), select RMSE LAI and RMSE Leaf N％ The crop parameters with the smallest sum are taken as the optimal parameters.

[0129] In step S4, see Figure 2The coupling of the PROSAIL model and the DSSAT model is achieved by first inversing the LAI from the PROSAIL model through remote sensing spectral data, then simulating the mapping relationship between the leaf area index (LAI) and the nitrogen content by the DSSAT, and then finding the relationship between the canopy spectrum and the nitrogen content of the leaf by taking the LAI as a connection. In this way, the method for inversing the nitrogen content based on the physiological process of plant growth and the physical process of spectral reflection is realized.

[0130] In the technical solution, the vegetation canopy radiation transfer model (PROSAIL model) inverses the leaf area index (LAI) of the vegetation by inputting the hyperspectral data, and the inverses LAI is taken as an input parameter of the agricultural technology transfer decision support system model (DSSAT, Decision Support System for Agrotechnology Transfer) model to simulate the change of the nitrogen content. The partial least squares regression (PLSR) is a kind of multivariate regression method for establishing the relationship between the independent variables (spectral reflectance) and the dependent variables (LAI), and is especially suitable for the case that there is a multiple collinearity between the high-dimensional data and the independent variables. The PLSR finds the principal components that can explain the maximum covariance of the spectral reflection data (independent variables) and the LAI (dependent variables) by extracting the information in the two at the same time. The DSSAT model integrates the environmental and management data to simulate the crop growth process of the pasture. The simulation result of the model needs to be verified according to the actual observation result, the model is adjusted according to the verification result, the coupling of the PROSAIL and the DSSAT model is achieved by first inversing the LAI from the PROSAIL model through remote sensing spectral data, then simulating the mapping relationship between the leaf area index (LAI) and the nitrogen content by the DSSAT, and then finding the relationship between the canopy spectrum and the nitrogen content of the leaf by taking the LAI as a connection. In this way, the method for inversing the nitrogen content based on the physiological process of plant growth and the physical process of spectral reflection is realized, so that the plant nitrogen is stably and accurately extracted, and the generalization ability of the nitrogen extraction method is improved.

[0131] In the embodiment, the spectral change caused by scattering in the process of measuring the nitrogen content of the grassland is reduced by pre-processing the acquired grassland measured data, especially the spectral difference caused by different scattering levels is eliminated, so that the correlation between the spectrum and the data is standardized. The extracted leaf area index is taken as a connection to couple the extracted leaf area index-nitrogen content and the leaf area index, and the nitrogen content of the grassland is inversely calculated, so that the dynamic change of the nitrogen in the plant growth process is better captured, and the stability and portability are higher, so that the nitrogen content of the grassland is quickly and effectively inversely calculated, the plant nitrogen is stably and accurately extracted, and the work difficulty of the plant nitrogen extraction is reduced.

[0132] The application adopts coupling of PROSAIL and DSSAT models instead of only relying on a spectral empirical model, and the mapping relationship between LAI and nitrogen content generated by the DSSAT model is beneficial to improve the poor spatiotemporal portability of spectral inversion of nitrogen. Although the coupling of the DSSAT and PROSAIL models has high complexity, compared with a traditional empirical model, the coupling can more accurately capture the dynamic process of vegetation growth and is suitable for remote sensing monitoring in a large range. The coupling method not only provides an effective remote sensing data assimilation framework, but also can supplement the vegetation state of grassland during a period without remote sensing data, reduce the limitation of the time resolution of remote sensing data, and provide complete spatiotemporal dynamic distribution of grassland vegetation characteristics in a large area. The coupling method has important significance for long-term monitoring and evolution analysis of grassland.

[0133] Embodiment two

[0134] Based on the grassland nitrogen content inversion method provided in embodiment one, this embodiment provides corresponding experimental data to prove the method, and the method is specifically as follows.

[0135] The leaf N content inversion effect based on the coupling model of DSSAT and PROSAIL is close to that of the spectral empirical model of the canopy. The research result is beneficial to capture the dynamic change of nitrogen in the growth process of plants, and has higher stability and portability.

[0136] Reference Figure 3 The Leaf N% is modeled by partial least squares regression at the canopy level, and the result is as follows: the ncomp of PLSR is 3, the modeling is n=10, r=0.870, and RMSE=0.489%; the verification is n=4, r=0.592, and RMSE=0.787%.

[0137] The DSSAT and PROSAIL models are coupled through LAI. According to the measured canopy spectrum, the LAI is obtained by inversion through the spectral database constructed by PROSAIL, and then the nitrogen value corresponding to the original canopy spectrum is obtained through the mapping relationship between the LAI and the leaf nitrogen simulated by the DSSAT model. The evaluation result of the nitrogen inversion of the coupled model is r=0.744 and RMSE=0.777%.

[0138] The first step of the model coupling is to obtain accurate observation data to provide accurate input for DSSAT and PROSAIL models. These data include: (1) Collecting leaf and canopy spectral data, data collection is carried out by handheld field spectrometer; (2) Pasture. Physiological and biochemical data of vegetation, including: such as leaf area index (LAI), chlorophyll content, carotene content, leaf nitrogen content, etc. These data are obtained by laboratory measurement to ensure that they can be matched with spectral data; (3) Environmental data: including meteorological data, including precipitation, maximum temperature, minimum temperature, solar radiation (such as temperature, rainfall, light intensity, etc.), these data can be provided by the European Centre for Medium-Range Weather Forecasts; (4) Soil data, including Clay%, Silt%, Stones%, Organic carbon%, pH in water, Cation exchange capacity cmol / kg, Total nitrogen%, these data can be provided by the World Soil Database (HWSD) (such as soil nitrogen content, humidity, etc.) and; (5) Management measures (such as irrigation, etc.), such as irrigation, harvesting data, etc. These data need to be recorded according to the actual situation during the experiment, and these data are used as input parameters of DSSAT model.

[0139] 1) Running of PROSAIL model

[0140] Referring to Figure 4 , the parameter range of the model is determined according to the measured values and the empirical values provided in the reference, as follows: ① Leaf structure parameter N: 1 ~ 3.5, increment 0.5; ② Chlorophyll Cab: 10 ~ 80 ug / cm 2 , increment 2 ug / cm 2 ; ③ Carotene Car: 2 ~ 24 ug / cm 2 , increment 2 ug / cm 2 ; ④ The value of water content Cw is set to 0.014 g / cm 2 ; ⑤ The dry matter content Cm is set to 0.0125 g / cm 2 . After testing, the optimal leaf structure parameter range for this grassland sample is 1.5 ~ 2.5. The range of LAI is adjusted to 0.25 ~ 9.75 according to the measured LAI data, with an increment of 0.5. According to the measured soil spectral curve, the soil spectral parameters are set to 55% wet soil + 45% dry soil, and the parameters are input to generate simulated spectra. For example, Figure 2 is the simulated spectral reflectance under different LAI.

[0141] According to formula (1), the measured spectrum is corrected, and then the PROSAIL model parameters are continuously adjusted to reduce the deviation between the predicted value and the actual observation value. The final modeling and verification results are as follows Figure 5 ), and the gray points are the LAI modeling and prediction results based on the PROSAIL spectral database. As the LAI increases, the data is concentrated in the larger value area, so the logarithm value is taken to reduce the deviation of the model. The model prediction ability of the first principal component is the best. Among them, the modeling part r = 0.969, RMSE = 1.29%; the verification part r = 0.877, RMSE = 0.83%, and the model stability and prediction ability are within the acceptable range.

[0142] 2) DSSAT model running

[0143] Referring to Figure 6 , the relationship between LAI and LeafN% of C3 non-grass (left) and C4 grass vegetation (right) is simulated by setting the weather, soil, and management module files in the DSSAT model. Among them, the weather data is provided by the European Center for Medium-Range Weather Forecasts, the soil data is provided by the World Soil Database (HWSD), and the management data mainly includes irrigation records. By assimilating the measured LAI and nitrogen and the DSSAT model simulation data, the plant seeding density, plant variety, and ecology parameters are adjusted to obtain the mapping relationship between LAI and LeafN%, as shown in Figure 6 .

[0144] 3) Model coupling

[0145] Referring to Figure 7 , DSSAT and PROSAIL models are coupled through LAI. According to the measured canopy spectrum, the LAI is obtained by inversion of the spectral database constructed by PROSAIL, and then the nitrogen value corresponding to the original canopy spectrum is obtained through the LAI and leaf nitrogen mapping relationship simulated by the DSSAT model. The evaluation results of the nitrogen inversion of the coupled model are r = 0.744 and RMSE = 0.777%.

[0146] Example three

[0147] The embodiment provides a grassland nitrogen content inversion system, referring to Figure 8 , the system comprises:

[0148] An acquisition module is configured to acquire grassland measured data and environmental parameter data of a target area.

[0149] A processing module is configured to preprocess the grassland measured data.

[0150] The feature extraction module is configured to extract a mapping relationship feature of hyperspectral-leaf area index in vegetation according to the preprocessed grassland measured data.

[0151] The feature extraction module is further configured to extract a leaf area index-nitrogen content according to the environmental parameter data.

[0152] The grassland nitrogen content inversion module is configured to couple the extracted leaf area index-nitrogen content with the leaf area index by taking the leaf area index as a connection, and to perform inversion on the grassland nitrogen content.

[0153] The above description is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which is made by using the content of the specification and drawings, is also included in the patent protection scope of the present application.

Claims

1. A method for inverting grassland nitrogen content, characterized in that: The method comprises the following steps: Acquiring grassland measured data and environmental parameter data of the target area, and preprocessing the grassland measured data; Based on the pre-processed grassland measured data, the mapping relationship characteristics of vegetation hyperspectral and leaf area index are extracted; Extract leaf area index-nitrogen content based on environmental parameter data; Using leaf area index as a link, the extracted leaf area index-nitrogen content is coupled with the hyperspectral-leaf area index to invert grassland nitrogen content; The process of preprocessing the grassland measured hyperspectral data includes: The parameter ranges were divided based on the obtained grassland measured data, and a vegetation canopy radiation transfer model was constructed; The basic parameters of the constructed vegetation canopy radiation transfer model are adjusted according to the measured grassland data after parameter range division, and the spectral reflectance database under different parameters is generated using the vegetation canopy radiation transfer model. Based on the spectral reflectance database under different parameters, the vegetation canopy radiation transfer model is used to extract the hyperspectral reflectance under different parameters; According to the hyperspectral reflectance extracted from the vegetation canopy radiation transfer model, the canopy spectral curve data in the grassland measured data were spectrally corrected. The grassland measured data includes leaf area index and vegetation canopy hyperspectral data, and the environmental parameter data includes meteorological data, soil data, field management data and field observation data; The spectrum correction processing is performed on the canopy spectrum curve data in the obtained grassland measured data. The expression is: ； in, represents the reflectance value of the i-th sample after scattering correction at wavelength j, represents the measured original reflectivity value of the i-th sample at wavelength j, represents the average value of the measured original reflectivity value at wavelength j, Represents the standard deviation of the measured original reflectivity value at wavelength j, represents the standard deviation of the spectral reflectance value simulated by the vegetation canopy radiation transfer model at point j, It represents the average value of the spectral reflectance value simulated by the vegetation canopy radiation transfer model at point j.

2. The inversion method for grassland nitrogen content according to claim 1, characterized in that: The constructed vegetation canopy radiation transfer model expression is: ； in, represents the canopy reflectance, represents the blade structural parameters, Indicates the chlorophyll content, Indicates moisture content, Indicates dry matter content, represents the leaf area index, represents the average leaf inclination angle, represents the hotspot parameter, Indicates soil brightness, Indicates soil moisture, represents the observation zenith angle, represents the solar zenith angle, Indicates the relative azimuth between the sun and the observation.

3. The inversion method for grassland nitrogen content according to claim 2, characterized in that: The process of extracting the mapping relationship characteristics between vegetation hyperspectral and leaf area index includes: Based on the spectral reflectance database under different parameters, the vegetation canopy radiation transfer model is used to extract the leaf area index under different parameters; The hyperspectral reflectance and leaf area index extracted by the vegetation canopy radiation transfer model were used as the training set, and the spectrally corrected grassland measured data were used as the validation set. The training set is used to optimize the leaf area index inversion of the vegetation canopy radiation transfer model, and the trained vegetation canopy radiation transfer model is verified using the validation set. After passing the validation, the trained vegetation canopy radiation transfer model is obtained. The trained vegetation canopy radiation transfer model is used to invert the hyperspectral data in the input grassland measured data to extract the leaf area index of the vegetation.

4. The inversion method for grassland nitrogen content according to claim 3, characterized in that: The process of optimizing the leaf area index inversion of the vegetation canopy radiation transfer model using the training set includes: Based on the hyperspectral reflectance and leaf area index under different parameters in the training set, several rounds of optimization training were set up to traverse different parameter combinations to optimize the parameters of the vegetation canopy radiation transfer model; The vegetation canopy radiation transfer model uses the partial least squares regression method to invert the leaf area index of vegetation based on the hyperspectral reflectance and leaf area index under different parameters; A loss function was set to compare the leaf area index inversion results under different parameter combinations with the measured results, and the vegetation canopy radiation transfer model corresponding to the parameter combination with the smallest loss function value was taken as the trained vegetation canopy radiation transfer model.

5. The inversion method for grassland nitrogen content according to claim 4, characterized in that: The process of inverting the leaf area index of vegetation using the partial least squares regression method includes: The partial least squares regression method was used to establish the relationship between hyperspectral reflectance and leaf area index, and the main features with the largest covariance in hyperspectral reflectance and leaf area index were extracted. The expression is: The leaf area index of vegetation is inverted based on the main features of the maximum covariance between hyperspectral reflectance and leaf area index. The optimal parameter combination is selected based on the measured leaf area index and the corrected hyperspectral data, and the parameters of the vegetation canopy radiation transfer model are adjusted based on the parameter combination; in, represents the hyperspectral reflectance matrix, represents the leaf area index vector, represents the principal component matrix, and are regression coefficient matrices, and denote the residual matrix and error term respectively.

6. The inversion method for grassland nitrogen content according to claim 4, characterized in that: The loss function expression set in the process of parameter optimization of the vegetation canopy radiation transfer model is: ； in, Indicates the number of data points; Indicates the True values; Indicates the predicted values; It represents the sum of squared errors between the true value and the predicted value.

7. The inversion method for grassland nitrogen content according to any one of claims 1 to 6, characterized in that: The process of extracting leaf area index-nitrogen content includes: Construct an agricultural technology transfer decision support system model, and adjust the agricultural technology transfer decision support system model with different parameter range combinations based on environmental parameter data; The agricultural technology transfer decision support system model was used to extract the correlation characteristics between leaf area index and nitrogen content under different parameter range combinations, and the loss function was used to calculate the correlation characteristics between leaf area index and nitrogen content. The loss is calculated for the associated features of each parameter range combination, and the loss function The parameter range combination with the smallest output value is taken as the optimal parameter range combination; The optimal parameter range combination was set as the new parameter of the agricultural technology transfer decision support system model for leaf area index-nitrogen content extraction; Using the leaf area index as a connection, the vegetation canopy radiation transfer model is coupled with the agricultural technology transfer decision support system model, the correlation characteristics between the canopy spectrum and the leaf nitrogen content are extracted, the grassland nitrogen content is inverted, and the inversion results are output.

8. A grassland nitrogen content inversion system, the system being used to implement the steps of the method according to any one of claims 1 to 7, characterized in that: The system comprises: Acquisition module, used to obtain grassland measured data and environmental parameter data of the target area; A processing module, configured to pre-process the grassland measured hyperspectral data; The feature extraction module is used to extract the mapping relationship features of the vegetation medium and high spectrum-leaf area index based on the pre-processed grassland measured data; The feature extraction module is also used to extract leaf area index-nitrogen content based on environmental parameter data; The grassland nitrogen content inversion module is used to couple the extracted leaf area index-nitrogen content with the hyperspectral-leaf area index using the leaf area index as the connection to invert the grassland nitrogen content.

Images