A method and system for monitoring the total phenol content of wheat grains in saline-alkali soil by remote sensing

Abstract

The application belongs to the technical field of agricultural remote sensing monitoring, and discloses a kind of saline-alkali soil wheat grain total phenol content remote sensing monitoring method and system, based on multi-source remote sensing data, extract weather and vegetation characteristics by phenology matching, construct protein content estimation model containing salt stress nonlinear inhibition term, and accordingly establish the quantitative relationship between protein and total phenol content, finally, high-precision total phenol content is obtained by using optimization algorithm inversion, and effective monitoring of large-scale saline-alkali soil wheat grain quality is realized.The application initiates the "protein content-total phenol content" secondary inversion architecture, indirectly realizes the total phenol content monitoring through the protein remote sensing inversion technology, effectively solves the technical problem of low direct spectral inversion precision;Innovatively establishes the specific stress response model of saline-alkali soil, the nonlinear inhibition effect of salt on crop physiological process is considered in the inversion process, which significantly improves the monitoring precision and model adaptability, and has the characteristics of high monitoring precision, strong practicability and wide application range.

Description

Technical Field

[0001] This invention belongs to the field of agricultural remote sensing monitoring technology, and in particular relates to a remote sensing monitoring method and system for the total phenol content of wheat grains in saline-alkali land, which realizes a method and system for monitoring the total phenol content of wheat grains in saline-alkali land on a large scale, with high efficiency and precision. Background Technology

[0002] Saline-alkali land is a general term for saline soil, alkaline soil, and other soils with varying degrees of salinization and alkalization. It is characterized by high salt content or alkalization, poor soil structure, easy compaction, low organic matter content, nutrient deficiency, and weak soil fertility retention capacity, severely impacting crop growth. Through the development, management, and improvement of saline-alkali land, it can be converted into arable land, which is of great significance for increasing grain yield, improving land resource utilization, and promoting agricultural development.

[0003] Traditional crop quality monitoring methods mainly rely on manual field surveys and laboratory chemical analysis, which suffer from low efficiency, high cost, high destructiveness, and difficulty in large-scale application. In recent years, with the development of remote sensing technology, especially the application of multispectral and hyperspectral remote sensing, researchers have begun to explore using remote sensing to monitor and evaluate crop quality. For example, some studies have proposed using remote sensing data acquired by the Landsat 8 OLI sensor, combined with vegetation indices (such as NDVI), to predict the protein content of winter wheat grains. Other studies have used hyperspectral remote sensing technology to monitor wheat grain protein content, enabling rapid and accurate acquisition of current wheat grain protein content information.

[0004] Furthermore, the rapid development of technologies such as agricultural IoT, wireless network transmission, and UAV remote sensing data monitoring in recent years has greatly promoted the transformation and development of agriculture towards large-scale and intelligent management. For example, existing patents have proposed methods for monitoring crop growth based on UAV point cloud data processing, acquiring crop growth data through lidar and multispectral cameras. Other inventions have proposed remote sensing monitoring methods for crop protein content based on map-spectrum synergy, predicting crop protein content by acquiring texture, structural, growth, and pigment parameters. Although these technologies have promoted the development of agricultural remote sensing monitoring to some extent, they still have significant limitations when used for monitoring the total phenolic content of wheat grains in saline-alkali land.

[0005] 1. Lack of specificity for saline-alkali land: Existing remote sensing monitoring models are mostly based on ordinary cultivated land and do not consider the stress effects of soil salinity, pH, ionic composition, and other factors specific to saline-alkali land on the formation of secondary metabolites in wheat. High salinity in saline-alkali land can lead to abnormal physiological metabolism in wheat, affecting the synthesis and accumulation of antioxidants such as total phenols.

[0006] 2. Limited monitoring indicators: Most existing technologies only focus on protein content and fail to comprehensively capture key biochemical parameters for wheat quality formation in saline-alkali land, such as total phenol content and other antioxidants, resulting in insufficient applicability.

[0007] 3. Poor model universality: The spectral model based on ordinary cultivated land has not been optimized for the spectral response characteristics of wheat in saline-alkali land. Its inversion accuracy is severely affected by factors such as the degree of soil background salinization and variations in plant canopy structure, making it difficult to meet the needs of accurate quality monitoring of saline-alkali land.

[0008] These problems arise because saline-alkali land ecosystems are complex, and the formation of crop secondary metabolites is influenced by multiple factors interacting with soil, environment, and organisms. Previous technologies have failed to establish a cross-scale, multi-indicator, and causally clear remote sensing monitoring system. Challenges encountered include how to achieve simultaneous inversion of soil salinity parameters and crop biochemical parameters, how to establish a spectral diagnostic model suitable for saline-alkali stress environments, and how to integrate multi-source remote sensing data to form an operational monitoring system.

[0009] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0010] (1) Traditional crop quality monitoring methods mainly rely on manual field surveys and laboratory chemical analysis, which have problems such as low efficiency, high cost, strong destructiveness and difficulty in large-scale application.

[0011] (2) Existing methods for monitoring total phenol content in wheat grains in saline-alkali land lack specificity for saline-alkali land, have single monitoring indicators, low accuracy in monitoring total phenol content, and poor model universality, making it difficult to achieve large-scale application. Summary of the Invention

[0012] Currently, only total phenol content spectral detection technology is used in laboratory settings, which cannot meet the needs of acquiring large-scale wheat grain quality information. There are also no relevant technologies for estimating large-scale spatial ranges. To address the problems of low accuracy, lack of specificity for saline-alkali land, and difficulty in large-scale application of existing technologies, this invention discloses a remote sensing monitoring method and system for total phenol content in wheat grains in saline-alkali land. This method enables large-scale, multi-scale, high-precision, and operational monitoring of total phenol content in wheat grains in saline-alkali land, solving the problem of difficulty in acquiring total phenol content data for wheat grains in large-scale saline-alkali land. The technical solution is as follows:

[0013] This invention is implemented as follows: a remote sensing monitoring method for the total phenol content of wheat grains in saline-alkali land, comprising the following steps:

[0014] S1. Construct a remote sensing estimation model for soil salinity and estimate soil salinity based on multi-source remote sensing data;

[0015] S2, Construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of each meteorological feature parameter within the key period;

[0016] S3. Construct a protein content estimation model. This model is based on vegetation index, multi-dimensional meteorological characteristic parameters and soil salinity. The model includes a nonlinear term that is negatively correlated with soil salinity to characterize the inhibitory effect of salt stress on protein synthesis.

[0017] S4. Construct a total phenol content prediction model. Based on the output of the protein content estimation model, predict the total phenol content of wheat grains through a preset functional relationship.

[0018] S5, the parameters of the total phenol content monitoring model are optimized using an optimization algorithm, including parameter initialization, fitness calculation and parameter optimization, in order to minimize the error between the predicted value and the measured value; wherein, the parameters of the total phenol content monitoring model include the model parameters and independent variables in steps S1 to S4.

[0019] In step S1, the soil salinity is estimated based on multi-source remote sensing data as follows:

[0020] ;

[0021] In the formula, Soil salinity This is a functional relationship established based on multi-source remote sensing data. The vegetation index, Salinity index Clay content, These are terrain parameters.

[0022] In step S2, the pixel-by-pixel phenological matching method is as follows: starting from the wheat maturity date DOY, move forward three 30-day periods, calculate the cumulative value and / or maximum value of the feature parameters in each period, and the expression is as follows;

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] In the formula, This is the cumulative value of evaporation. This is the cumulative value of precipitation. This is the cumulative value of temperature. This represents the maximum value of the vegetation index. These are the observed values ​​for each time period. for Evapotranspiration value over a period of time for Rainfall values ​​for the specified time period. for Temperature values ​​for the time period To obtain the maximum value.

[0028] In step S3, the nonlinear term negatively correlated with soil salinity includes: a linear term of soil salinity SSC and an exponentially decaying term with respect to SSC.

[0029] In step S3, the expression for the protein content estimation model is:

[0030] ;

[0031] In the formula, Protein content, All of these are model parameters. The vegetation index, Soil salinity The attenuation coefficient is... This is the error term.

[0032] In step S4, the expression for the total phenol content prediction model is:

[0033] ;

[0034] In the formula, This refers to the total phenol content. All of these are model parameters. This refers to the protein content.

[0035] In step S5, the optimization of the parameters of the total phenol content monitoring model using an optimization algorithm includes:

[0036] The initial values ​​of the model parameters are determined based on the ground data, and the parameter optimization range is set.

[0037] The root mean square error (RMSE) between the measured and predicted values ​​is used as the fitness function.

[0038] The optimized parameter values ​​are obtained by iterative optimization using a genetic algorithm to minimize the RMSE.

[0039] Furthermore, the formula for calculating the fitness function is:

[0040] ;

[0041] In the formula, For individual fitness, This represents the measured total phenol content at the sample point. This represents the predicted total phenol content for the sample points. This represents the number of training samples.

[0042] The process after step S5 also includes: inputting the optimized parameter values ​​into the total phenol content monitoring model, and combining multi-dimensional feature parameters as input to achieve remote sensing monitoring of the total phenol content of wheat grains in saline-alkali land.

[0043] Another object of the present invention is to provide a remote sensing monitoring system for the total phenolic content of wheat grains in saline-alkali land, which is used to realize a remote sensing monitoring method for the total phenolic content of wheat grains in saline-alkali land. The system includes:

[0044] The soil salinity remote sensing estimation module is used to construct a soil salinity remote sensing estimation model and estimate soil salinity based on multi-source remote sensing data.

[0045] The multi-dimensional feature parameter extraction module is used to construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of multiple meteorological feature parameters related to crop growth during the key periods;

[0046] The protein content estimation module is used to construct a protein content estimation model. This model is based on vegetation index, multi-dimensional meteorological characteristic parameters and soil salinity. The model includes a nonlinear term that is negatively correlated with soil salinity, which is used to characterize the inhibitory effect of salt stress on protein synthesis.

[0047] The total phenol content prediction module is used to construct a total phenol content prediction model and predict the total phenol content of wheat grains based on the output of the protein content estimation model through a preset functional relationship.

[0048] The parameter optimization module is used to optimize the parameters of the total phenol content monitoring model using optimization algorithms, including parameter initialization, fitness calculation, and parameter optimization, to minimize the error between predicted and measured values. The parameters of the total phenol content monitoring model include parameters and independent variables from the soil salinity remote sensing estimation model, the multi-dimensional feature parameter extraction model, the protein content estimation model, and the total phenol content prediction model.

[0049] Combining all the above technical solutions, the beneficial effects of this invention are as follows:

[0050] First, this invention constructs a two-level inversion architecture of "protein content - total phenol content," indirectly monitoring total phenol content through mature remote sensing inversion technology for protein content, thus solving the technical problem of low accuracy in direct spectral inversion. It also establishes a saline-alkali land-specific monitoring model, innovatively considering the nonlinear inhibitory effect of salt on crop physiological processes, accurately reflecting the environmental characteristics of saline-alkali land. The total phenol content monitoring model of this invention is a semi-mechanistic model, possessing a certain degree of mechanistic validity, and can accurately monitor total phenol content during key stages of crop growth. The monitoring model comprehensively considers multi-source data such as meteorological, vegetation, and soil data, ensuring the timeliness of feature extraction through phenological matching. A salt stress term is added to the protein estimation model to better characterize the characteristics of saline-alkali land. Furthermore, this invention is applicable to wheat-growing areas with varying degrees of salinity and alkalinity. Monitoring results can provide timely guidance for optimized cultivation, ensuring the quality of agricultural products, and is of great significance for agricultural product quality monitoring and precision agriculture applications.

[0051] Secondly, this invention pioneers a two-level inversion architecture of "protein content - total phenol content," indirectly monitoring total phenol content through mature protein remote sensing inversion technology, effectively solving the technical problem of low accuracy in direct spectral inversion. It also innovatively establishes a stress response model specific to saline-alkali land, considering the nonlinear inhibitory effect of salt on crop physiological processes during the inversion process, significantly improving monitoring accuracy and model adaptability. This method represents a breakthrough from traditional single-indicator monitoring to multi-indicator synergistic inversion, providing a new technical approach for wheat quality monitoring in saline-alkali land, and features high monitoring accuracy, strong practicality, and wide applicability.

[0052] Third, this invention can be transformed into specialized software or an online service platform for monitoring the quality of wheat in saline-alkali land, providing precise crop quality monitoring services for agricultural management departments, agricultural cooperatives, and large farms. It is expected to reduce the costs of traditional manual sampling and laboratory analysis, improve monitoring efficiency, and generate significant economic benefits annually after widespread adoption in saline-alkali wheat-growing areas. Simultaneously, by guiding optimized cultivation practices through precise monitoring, it can improve the quality and market value of wheat in saline-alkali land, contributing to the development of agricultural product brands.

[0053] Fourth, there are currently no mature technical solutions for remote sensing monitoring of total phenol content in wheat grains in saline-alkali land, either domestically or internationally. This invention is the first to construct a two-level inversion architecture of "protein content-total phenol content" and establish a specific monitoring model that includes the nonlinear inhibitory effect of salt stress. It realizes multi-process coupled simulation from environmental stress to crop physiological response, filling the technical gap in the field of remote sensing monitoring of crop quality in saline-alkali land. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure;

[0055] Figure 1 This is a flowchart of the remote sensing monitoring method for total phenol content in wheat grains in saline-alkali land provided in this embodiment of the invention;

[0056] Figure 2 This is a scatter plot showing the relationship between total phenol content and protein content provided in an embodiment of the present invention. Detailed Implementation

[0057] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0058] There is currently no precedent for large-scale, wide-area remote sensing monitoring of the spatial distribution of total phenol content in wheat grains. The innovation of this invention lies in:

[0059] (1) This invention pioneered a two-level inversion architecture of “protein content-total phenol content”: the total phenol content is monitored indirectly through mature protein content remote sensing inversion technology, which solves the technical problem of low accuracy of direct spectral inversion of total phenol content.

[0060] (2) This invention establishes a specific stress response model for saline-alkali land: It innovatively uses the exponential decay form exp(-k×SSC) to characterize the nonlinear effect of salt stress, which more accurately describes the inhibitory effect of salt on crop physiological processes.

[0061] These innovative aspects collectively constitute the core of the technical solution of this invention, providing an effective technical means for monitoring the physiological state of wheat in saline-alkali land.

[0062] Example 1, as Figure 1 As shown, the remote sensing monitoring method for total phenol content in wheat grains in saline-alkali land provided in this embodiment of the invention includes the following steps:

[0063] S1. Construct a remote sensing estimation model for soil salinity and estimate soil salinity based on multi-source remote sensing data;

[0064] The remote sensing estimation model for soil salinity is based on multi-source data, and the specific formula is as follows:

[0065] ;

[0066] In the formula, This refers to soil salt content. This is a functional relationship established based on multi-source remote sensing data. The vegetation index, Salinity index Clay content, These are terrain parameters.

[0067] S2, Construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of each meteorological feature parameter within the key period;

[0068] Multi-dimensional feature parameter extraction employs a pixel-by-pixel phenological matching method, moving forward three 30-day periods from the maturity date (DOY) to calculate the cumulative values ​​of feature parameters within each period, including:

[0069] ;

[0070] ;

[0071] ;

[0072] ;

[0073] In the formula, This is the cumulative value of evaporation. This is the cumulative value of precipitation. This is the cumulative value of temperature. This represents the maximum value of the vegetation index. These are the observed values ​​for each time period. for Evapotranspiration value over a period of time for Rainfall values ​​for the specified time period. for Temperature values ​​for the time period To obtain the maximum value.

[0074] S3. Construct a protein content estimation model. This model is based on vegetation index, multi-dimensional meteorological characteristic parameters and soil salinity. The model includes a nonlinear term that is negatively correlated with soil salinity to characterize the inhibitory effect of salt stress on protein synthesis.

[0075] The expression for the protein content estimation model is as follows:

[0076] ;

[0077] In the formula, Protein content, All of these are model parameters. The vegetation index, Soil salinity The attenuation coefficient is... This is the error term.

[0078] S4. Construct a total phenol content prediction model. Based on the output of the protein content estimation model, predict the total phenol content of wheat grains through a preset functional relationship.

[0079] The scatter plot of the relationship between total phenol content and protein content is shown below. Figure 2 As shown. The prediction of total phenol content is based on the quantitative relationship between protein content and total phenol content, and the specific formula is as follows:

[0080] ;

[0081] In the formula, This refers to the total phenol content. All of these are model parameters. This refers to the protein content.

[0082] S5, the parameters of the total phenol content monitoring model are optimized using an optimization algorithm, including parameter initialization, fitness calculation and parameter optimization, in order to minimize the error between the predicted value and the measured value; wherein, the parameters of the total phenol content monitoring model include the model parameters and independent variables in steps S1 to S4.

[0083] The initial values ​​of the model parameters are determined based on the ground data, and the parameter optimization range is set.

[0084] The root mean square error (RMSE) between the measured and predicted values ​​is used as the fitness function.

[0085] The optimized parameter values ​​are obtained by iterative optimization using a genetic algorithm to minimize the RMSE.

[0086] The total phenol content monitoring model parameters obtained in step S4 are optimized using a genetic optimization algorithm. and independent variable Optimizations include:

[0087] S5.1, Initialize parameters and determine the parameter optimization range, using the model parameters established using ground data as initial values;

[0088] S5.2, Fitness Calculation: When using a genetic algorithm to optimize model parameters, the root mean square error (RMSE) between the measured and predicted values ​​is selected as the fitness function, and the minimum RMSE after using the monitoring model to predict the training data is taken as the optimization objective.

[0089] Individual fitness is calculated as follows:

[0090] ;

[0091] In the formula, For individual fitness, This represents the measured total phenol content at the sample point. This represents the predicted total phenol content for the sample points. This represents the number of training samples.

[0092] S5.3 Total phenol content monitoring: The optimized parameter values ​​are input into the monitoring model, and multi-dimensional feature parameters are used as inputs to obtain the predicted total phenol content through the monitoring model.

[0093] The optimized parameter values ​​are input into the total phenol content monitoring model, and combined with multi-dimensional feature parameters as input, to achieve remote sensing monitoring of the total phenol content of wheat grains in saline-alkali land.

[0094] Example 2: This embodiment of the invention provides a remote sensing monitoring system for the total phenolic content of wheat grains in saline-alkali land, used to implement the aforementioned remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land, comprising:

[0095] The soil salinity remote sensing estimation module is used to construct a soil salinity remote sensing estimation model and estimate soil salinity based on multi-source remote sensing data.

[0096] The multi-dimensional feature parameter extraction module is used to construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of multiple meteorological feature parameters related to crop growth during the key periods;

[0097] The protein content estimation module is used to construct a protein content estimation model. This model is based on vegetation index, multi-dimensional meteorological characteristic parameters and soil salinity. The model includes a nonlinear term that is negatively correlated with soil salinity, which is used to characterize the inhibitory effect of salt stress on protein synthesis.

[0098] The total phenol content prediction module is used to construct a total phenol content prediction model and predict the total phenol content of wheat grains based on the output of the protein content estimation model through a preset functional relationship.

[0099] The parameter optimization module is used to optimize the parameters of the total phenol content monitoring model using optimization algorithms, including parameter initialization, fitness calculation and parameter optimization, in order to minimize the error between the predicted value and the measured value; wherein, the parameters of the total phenol content monitoring model include the parameters and independent variables of the soil salinity remote sensing estimation model, the multi-dimensional feature parameter extraction model, the protein content estimation model and the total phenol content prediction model.

[0100] To further demonstrate the positive effects of the above embodiments, soil salinity, wheat protein, and total phenol content data from spring 2024 and spring 2025 were used to conduct the following simulation experiments based on the technical solution of this invention. The existing data include: Sentinel-2 multispectral remote sensing imagery (including red-edge and near-infrared bands), daily meteorological data for the entire wheat growth period in the study area in 2024 (including evapotranspiration ET, precipitation Pre, and temperature T), and laboratory measurements of soil salinity, wheat protein, and total phenol content from 120 sample points obtained in 2025. These data cover multi-dimensional information from environmental factors to crop biochemical parameters, providing sufficient data support for model construction and validation.

[0101] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention and within the spirit and principles of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A remote sensing method for monitoring the total phenolic content of wheat grains in saline-alkali land, characterized in that, The method includes the following steps: S1. Construct a remote sensing estimation model for soil salinity and estimate soil salinity based on multi-source remote sensing data; S2, Construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of each meteorological feature parameter within the key period; S3. Construct a protein content estimation model. This protein content estimation model is based on vegetation index, multi-dimensional meteorological characteristic parameters and soil salinity. The protein content estimation model includes a nonlinear term that is negatively correlated with soil salinity, which is used to characterize the inhibitory effect of salt stress on protein synthesis. S4. Construct a total phenol content prediction model. Based on the output of the protein content estimation model, predict the total phenol content of wheat grains through a preset functional relationship. S5, the parameters of the total phenol content monitoring model are optimized using an optimization algorithm, including parameter initialization, fitness calculation and parameter optimization, in order to minimize the error between the predicted value and the measured value; wherein, the parameters of the total phenol content monitoring model include the parameters and independent variables of the soil salinity remote sensing estimation model, the multi-dimensional feature parameter extraction model, the protein content estimation model and the total phenol content prediction model. The expression for the protein content estimation model is as follows: ； In the formula, Protein content, All of these are model parameters. The vegetation index, Soil salinity The attenuation coefficient is... This is the error term; This is the cumulative value of evaporation. This is the cumulative value of precipitation. This is the cumulative value of temperature.

2. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 1, characterized in that, In step S1, the soil salinity is estimated based on multi-source remote sensing data as follows: ； In the formula, Soil salinity This is a functional relationship established based on multi-source remote sensing data. The vegetation index, Salinity index Clay content, These are terrain parameters.

3. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 1, characterized in that, In step S2, the pixel-by-pixel phenological matching method is as follows: starting from the wheat maturity date DOY, move forward three 30-day periods, calculate the cumulative value and / or maximum value of the feature parameters in each period, and the expression is as follows; ； ； ； ； In the formula, This is the cumulative value of evaporation. This is the cumulative value of precipitation. This is the cumulative value of temperature. This represents the maximum value of the vegetation index. These are the observed values ​​for each time period. for Evapotranspiration value over a period of time for Rainfall values ​​for the specified time period. for Temperature values ​​for the time period To obtain the maximum value, for Vegetation index for a given time period.

4. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 1, characterized in that, In step S4, the expression for the total phenol content prediction model is: ； In the formula, This refers to the total phenol content. All of these are model parameters. This refers to the protein content.

5. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 1, characterized in that, In step S5, the optimization of the parameters of the total phenol content monitoring model using an optimization algorithm includes: The initial values ​​of the model parameters are determined based on the ground data, and the parameter optimization range is set. The root mean square error (RMSE) between the measured and predicted values ​​is used as the fitness function. The optimized parameter values ​​are obtained by iterative optimization using a genetic algorithm to minimize the RMSE.

6. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 5, characterized in that, The formula for calculating the fitness function is: ； In the formula, For individual fitness, This represents the measured total phenol content at the sample point. This represents the predicted total phenol content for the sample points. This represents the number of training samples.

7. The remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land according to claim 1, characterized in that, The process after step S5 also includes: inputting the optimized parameter values ​​into the total phenol content monitoring model, and combining multi-dimensional feature parameters as input to achieve remote sensing monitoring of the total phenol content of wheat grains in saline-alkali land.

8. A remote sensing monitoring system for total phenolic content in wheat grains in saline-alkali land, used to implement the remote sensing monitoring method for total phenolic content in wheat grains in saline-alkali land as described in any one of claims 1-7, characterized in that, include: The soil salinity remote sensing estimation module is used to construct a soil salinity remote sensing estimation model and estimate soil salinity based on multi-source remote sensing data. The multi-dimensional feature parameter extraction module is used to construct a multi-dimensional feature parameter extraction model, use the pixel-by-pixel phenological matching method to determine the key periods of wheat growth, and calculate the cumulative value and / or maximum value of multiple meteorological feature parameters related to crop growth during the key periods; The protein content estimation module is used to construct a protein content estimation model. This model estimates protein content based on vegetation index, multi-dimensional meteorological characteristic parameters, and soil salinity. The protein content estimation model includes a nonlinear term that is negatively correlated with soil salinity, which is used to characterize the inhibitory effect of salt stress on protein synthesis. The total phenol content prediction module is used to construct a total phenol content prediction model and predict the total phenol content of wheat grains based on the output of the protein content estimation model through a preset functional relationship. The parameter optimization module is used to optimize the parameters of the total phenol content monitoring model using optimization algorithms, including parameter initialization, fitness calculation, and parameter optimization, in order to minimize the error between the predicted and measured values. The parameters of the total phenol content monitoring model include parameters and independent variables of the soil salinity remote sensing estimation model, the multi-dimensional feature parameter extraction model, the protein content estimation model, and the total phenol content prediction model.

Images