Dynamic diagnosis method of cultivated land soil quality based on multi-source remote sensing collaborative inversion

Abstract

The present invention discloses a dynamic diagnosis method for cultivated land soil quality based on multi-source remote sensing collaborative inversion, which relates to the field of agricultural information technology. The method includes real-time collection of physical and chemical parameters such as soil temperature, humidity, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations through a multi-source sensor network, application of a data preprocessing algorithm to eliminate outliers and perform standardization processing, and generation of a standardized soil parameter set for subsequent climate and management factor correlation analysis. Risk warning signals are automatically generated and pushed to an agricultural management decision-making platform, completing the full-process analysis from data collection to decision support. The dynamic diagnosis method for cultivated land soil quality based on multi-source remote sensing collaborative inversion can provide a scientific basis for agricultural management decisions, effectively guide farmland management practices, and improve soil quality and agricultural production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of agricultural information technology, and in particular to a method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion. Background Art

[0002] Current soil quality assessments rely primarily on traditional field monitoring and empirical assessments, but these methods are insufficient in the face of complex and changing climatic conditions. Existing research often focuses on single-factor analysis and lacks a deep understanding of the interactions between multiple environmental factors, making it difficult to provide a scientifically reliable basis for agricultural management decisions.

[0003] Traditional assessment methods are time-consuming and costly, and cannot meet the urgent need for real-time dynamic monitoring in modern agriculture. Soil quality change is a complex dynamic process affected by multiple factors, and there is an intricate interaction between different climatic conditions and agricultural management practices. Climate factors such as changes in precipitation patterns, temperature fluctuations, and extreme weather events can affect the physical and chemical properties of the soil, thereby changing the fertility level and structural stability of the soil. This multi-factor coupling effect makes it extremely difficult to accurately simulate the evolution of soil quality. To further complicate matters, different tillage methods such as no-till, deep plowing, and crop rotation have a cumulative impact on the soil over long periods of time. This long-term effect often has hysteresis and nonlinear characteristics. When extreme climate events are superimposed on specific agricultural management practices, the extent and duration of their impact on soil quality are even more difficult to predict, and traditional methods cannot effectively capture this complex spatiotemporal variation. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic diagnosis method for cultivated land soil quality based on multi-source remote sensing collaborative inversion, to construct a dynamic soil quality simulation system that can comprehensively consider climate change, agricultural management measures and their interactions, to accurately predict the potential impact of extreme climate events on cultivated land soil, and to scientifically evaluate the long-term cumulative effects of different tillage methods.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solution: a method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion, the method comprising:

[0006] Soil temperature, humidity, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations are collected in real time through a multi-source sensor network. Data preprocessing algorithms are used to remove outliers and perform standardization to generate a standardized soil parameter set for subsequent analysis of the relationship between climate and management factors.

[0007] Based on a standardized set of soil parameters, dynamic simulation data for soil quality is generated. Uncertainty analysis is performed on this data. One thousand parameter combinations are generated through random sampling, and the probability distribution interval of soil quality changes is calculated. If the prediction accuracy, assessed based on the ratio of the mean square error to the actual observed value, falls below a preset threshold of 85 percent, the system returns to the previous step to readjust the weight parameters of each factor in the neural network framework, generating soil quality change trend prediction data.

[0008] Based on the predicted data on soil quality change trends, an early warning mechanism is established. If the predicted value of the soil fertility level based on the comprehensive nitrogen, phosphorus and potassium index is lower than the preset safety threshold, that is, 80% of the historical average value, for three consecutive months, or the structural stability index drops by more than 20% of the preset threshold, a risk warning signal will be automatically generated and pushed to the agricultural management decision-making platform, completing the full process analysis from data collection to decision support.

[0009] Preferably, generating soil quality dynamic simulation data according to the standardized soil parameter set includes:

[0010] For the standardized soil parameter set, combined with the historical records of meteorological stations and satellite remote sensing data, we obtain climate change indicators such as precipitation, temperature fluctuations, humidity changes and wind speed. If the precipitation for seven consecutive days exceeds twice the historical average value for the same period calculated based on the local ten-year average, or the daily average temperature fluctuation exceeds the preset threshold of ten degrees Celsius, it is judged as an extreme weather event and the time node is marked to generate a climate change and extreme weather marker set.

[0011] Preferably, generating soil quality dynamic simulation data according to the standardized soil parameter set further comprises:

[0012] Agricultural management records of tillage methods, crop rotation cycles, fertilization frequency, and irrigation patterns are obtained from the farmland management database to construct a numerical coding system for tillage methods. Based on historical record analysis, the quantitative weight coefficients of different management measures are determined to generate a set of agricultural management quantitative parameters for subsequent soil quality impact assessment.

[0013] Preferably, generating soil quality dynamic simulation data according to the standardized soil parameter set further comprises:

[0014] Using a set of standardized soil parameters and a set of climate change and extreme weather markers, a multivariate regression analysis method was used to construct a correlation model between climate change factors and soil physical and chemical properties. If the monthly precipitation change rate exceeded the preset threshold of 30%, the soil moisture influence coefficient was adjusted. If the daily average temperature fluctuation range exceeded 10 degrees Celsius, the organic matter decomposition rate parameter was corrected. If an extreme weather marker was detected, a short-term impact factor was added to generate a climate-soil coupling correlation matrix.

[0015] Preferably, generating soil quality dynamic simulation data according to the standardized soil parameter set further comprises:

[0016] Based on a set of quantitative parameters of agricultural management, the time series analysis method is used to construct a model of the cumulative effects of tillage methods on soil quality. According to historical management records, the cumulative impact weights of different tillage measures on monthly, quarterly and annual time scales are calculated to generate a long-term agricultural management effect prediction model for multi-factor comprehensive simulation.

[0017] Preferably, generating soil quality dynamic simulation data according to the standardized soil parameter set further comprises:

[0018] By integrating the climate-soil coupling correlation matrix and the long-term effect prediction model of agricultural management, a neural network algorithm is used to construct a dynamic simulation framework of soil quality with multi-factor interaction. If extreme weather markers are detected, the emergency response mechanism is activated, the weights of soil moisture and organic matter decomposition rate parameters are adjusted, and soil quality dynamic simulation data are generated.

[0019] Preferably, the specific formula for performing uncertainty analysis is:

[0020] ;

[0021] in, represents the predicted soil quality evolution function, Indicates the number of sampling times, Indicates the The entropy weight factor corresponding to the sample is represents the nonlinear model function used for soil state assessment, Indicates the Set input dataset, represents the model structure parameters, Indicates the corresponding The perturbation sensitivity coefficient of samples, Indicates the The variance of the set of sample input variables, Indicates the sample number.

[0022] Preferably, the The calculation formula is:

[0023] ;

[0024] in, Indicates the The entropy weight factor corresponding to the sample is represents the weighted normalization factor of all samples, represents the input variable dimension, Indicates the The sample in The fuzzy membership of the indicators, Indicates the The sample in The normalized value of the indicator, Indicates the The average value of the indicators in the whole sample, Indicates the sample number, Represents the dimension number index.

[0025] Preferably, the The calculation formula is:

[0026] ;

[0027] in, Indicates the corresponding The perturbation sensitivity coefficient of samples, represents the total number of soil input features, Indicates the dimension number of the input variable currently traversed, Represents the model function B on the input variables The partial derivative of represents the input variable, Indicates the The sample in The degree of fluctuation on the input variables, Indicates that all samples in The maximum fluctuation value of a variable.

[0028] Preferably, the early warning mechanism uses color classification to display the risk level, which is divided into three levels: normal, mild warning, and severe warning, and synchronizes the early warning results to the plot view of the agricultural management platform in the form of a graphical interface.

[0029] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0030] This method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion collects soil physicochemical parameters, climate change indicators, and agricultural management data to construct a climate-soil coupling relationship matrix and a prediction function for the long-term effects of agricultural management. A neural network algorithm is used to integrate the above models to establish a dynamic simulation model of soil quality with multi-factor interactions. Combined with the extreme weather event response mechanism, this method enables accurate prediction of soil quality change trends. The present invention also uses the Monte Carlo method for uncertainty analysis to improve prediction accuracy and establishes an early warning mechanism to automatically generate a risk warning signal when the prediction result falls below a safety threshold. This method can provide a scientific basis for agricultural management decisions, effectively guide farmland management practices, and improve soil quality and agricultural production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] like Figure 1 As shown, the present invention provides a technical solution: a dynamic diagnosis method for cultivated land soil quality based on multi-source remote sensing collaborative inversion, the method comprising:

[0034] Soil temperature, humidity, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations are collected in real time through a multi-source sensor network. Data preprocessing algorithms are used to remove outliers and perform standardization to generate a standardized soil parameter set for subsequent analysis of the relationship between climate and management factors.

[0035] Based on a standardized set of soil parameters, dynamic simulation data for soil quality is generated. Uncertainty analysis is performed on this data. One thousand parameter combinations are generated through random sampling, and the probability distribution interval of soil quality changes is calculated. If the prediction accuracy, assessed based on the ratio of the mean square error to the actual observed value, falls below a preset threshold of 85 percent, the system returns to the previous step to readjust the weight parameters of each factor in the neural network framework, generating soil quality change trend prediction data.

[0036] Based on the predicted data on soil quality change trends, an early warning mechanism is established. If the predicted value of the soil fertility level based on the comprehensive nitrogen, phosphorus and potassium index is lower than the preset safety threshold, that is, 80% of the historical average value, for three consecutive months, or the structural stability index drops by more than 20% of the preset threshold, a risk warning signal will be automatically generated and pushed to the agricultural management decision-making platform, completing the full process analysis from data collection to decision support.

[0037] This implementation addresses the challenges of high spatial heterogeneity, strong temporal variability, and discontinuous data collection in cultivated land by constructing a data collection system based on the fusion of multi-source remote sensing and ground sensors. The system architecture combines drone remote sensing, satellite imagery, and fixed-point ground sensors to form a multi-source information input, enabling real-time, high-density collection of key physical and chemical parameters such as soil temperature, moisture, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations. The collected data is initially processed by an edge computing unit. Outliers are removed using the IQR, 3σ, or time series difference detection techniques. Normalization is performed using normalization or the Z-score method to ensure parameter scale consistency and data comparability. During the data modeling phase, a deep learning model based on an LSTM (Long Short-Term Memory) or GRU (Gated Recurrent Unit) architecture is introduced to develop a dynamic prediction model for the standardized set of soil parameters. This model effectively captures the temporal variations of soil properties, constructs a multidimensional input matrix using time windows, and learns soil quality trends through model training. Based on this, Monte Carlo simulations were used to randomly sample the input parameter space, generating one thousand parameter combinations. The corresponding set of predicted soil quality change values ​​was calculated through forward propagation of the model. Probability distribution intervals were then constructed, and statistical indicators such as confidence intervals and quantiles were extracted to assess uncertainty. If the error between the predicted output and the actual observed data (such as mean squared error (MSE), root mean squared error (RMSE), and coefficient of determination (R²)) fell below a set accuracy threshold (e.g., 85%), the model's reverse optimization process was triggered. By adjusting the weights of various factors in the neural network (such as climate variables, fertilization methods, and tillage practices), the model was retrained and the prediction results were iteratively optimized. The resulting soil quality trend data were then fed into an early warning module, which employed a variety of triggering indicators, primarily focusing on the nitrogen, phosphorus, and potassium (NPK) index and structural stability as early warning benchmarks. If the comprehensive fertility index is lower than 80% of the historical average for three consecutive months, or the structural stability drops by more than 20% compared with the benchmark, the system will automatically generate a risk signal and push it to the manager's terminal through the agricultural Internet of Things platform, including graphical interface prompts, alarm record archiving, and auxiliary decision-making recommendation generation, thus completing the full-chain intelligent analysis system from data collection, processing, prediction to intelligent decision-making.

[0038] This implementation builds a dynamic diagnostic system for cultivated land soil quality based on a data acquisition architecture that integrates multi-source remote sensing data with ground sensor networks. This system addresses current soil monitoring challenges, including strong spatial heterogeneity, insufficient time-series data, and delayed analysis. Its core technical approach encompasses five steps: real-time data acquisition, standardized processing, deep learning prediction, uncertainty assessment, and a risk early warning mechanism.

[0039] First, the system uses sensor nodes distributed across different areas of the farmland to collect key parameters such as temperature, humidity, pH, organic matter content, and nitrogen, phosphorus, and potassium. This data comes from ground-based fixed sensors, drone hyperspectral imagery, and satellite remote sensing imagery. To improve the accuracy of the analysis, the system standardizes the collected raw data. Specifically, for each parameter category, the historical mean and standard deviation are calculated. The mean is then subtracted from the current value and divided by the standard deviation, bringing all parameters onto a comparable standard scale.

[0040] Next, the system uses a long-short-term memory neural network to train a dynamic prediction model based on the standardized parameter sequence. This model learns how soil properties change over time and predicts soil quality trends over future time periods. During training, the system calculates the error between the predicted results and the actual observed values, using this error as a measure of model accuracy. If the error is large, the system automatically adjusts the model's internal parameters, such as re-weighting the input factors, to improve prediction accuracy.

[0041] In addition to model prediction, the system also incorporates an uncertainty analysis mechanism. Using the Monte Carlo method, the input parameter set is randomly sampled one thousand times, generating one thousand sets of simulated data. Each set of simulated data is processed by the prediction model to produce a corresponding set of predicted soil quality values. The system then statistically processes these results, extracting representative values ​​such as the 5th and 95th percentiles, thereby constructing a confidence interval that describes the range of variation in the prediction results. This process helps determine the stability and reliability of the prediction results.

[0042] To implement risk management and early warning functions, the system has designed a judgment standard centered around the soil fertility index and structural stability indicators. The soil fertility index is calculated by weighting the concentrations of nitrogen, phosphorus, and potassium, with nitrogen weighted at 40%, and phosphorus and potassium each at 30%. If the soil fertility index of a particular farmland falls below 80% of the historical average for three consecutive months, the system will deem it a risk of fertility decline. Furthermore, if the structural stability index drops by more than 20% compared to the average level of the previous year, it will be considered a sign of potential structural damage. In either case, the system will automatically generate a risk warning signal.

[0043] Warning results are distributed as data reports via the Agricultural IoT platform to relevant agricultural management terminals. These reports include not only the warning type, level, and triggering cause, but also charts of soil quality trends, time series statistics of key parameters, and recommended interventions. This helps managers develop scientific response strategies to ensure the sustainable and healthy use of cultivated land resources.

[0044] The prediction accuracy threshold is used to determine whether the model prediction results are accurate enough. The "prediction accuracy percentage" is calculated by the ratio of the mean square error to the variance of the observed value. Referring to the relevant literature on agricultural soil quality prediction, an accuracy of 85% can meet most management decision requirements. If it is lower than this threshold, it means that the prediction error is large and the system needs to readjust the neural network model.

[0045] The fertility warning threshold (80% of the historical average) is used to determine whether the current soil fertility is at a low level and trigger an early warning. The comprehensive index calculated by weighting the nitrogen, phosphorus and potassium concentrations is used as the fertility level indicator. The historical fertility index of the same cultivated land in the past 5-10 years is statistically analyzed and its average value is calculated. If the current value is lower than 80% of the average value for three consecutive months, it is considered to have a risk of decline. 80% is a commonly used "low value alarm line" in meteorological and soil environmental monitoring and provides a practical reference basis.

[0046] The soil structure stability threshold (a decrease of more than 20%) is used to assess the degree of deterioration of soil aggregate structure. The 12-month moving average of the structural stability of a certain area (such as the content of water-stable aggregates) is monitored. If the current indicator decreases by more than 20% compared with the average, it is judged to be structural deterioration. Soil structure changes usually lag behind chemical indicators, so a 20% decrease is regarded as a significant fluctuation. The empirical data of the long-term positioning monitoring project of farmland ecosystems is used as a reference.

[0047] This method significantly improves the timeliness and spatial coverage of soil quality monitoring by integrating multi-source remote sensing data with ground-based sensor data. Standardization and anomaly rejection algorithms enhance data reliability, while deep learning models combined with uncertainty analysis improve the accuracy and stability of soil quality predictions. An early warning mechanism, based on continuous data trend analysis, ensures timely and accurate risk response, helping agricultural managers intervene early and develop scientific farming strategies. The entire system incorporates data-driven, adaptive parameter adjustment, and automatic early warning capabilities, achieving a closed-loop management system from data perception to intelligent decision-making.

[0048] Based on the standardized soil parameter set, dynamic simulation data of soil quality is generated, including obtaining precipitation, temperature fluctuations, humidity changes and wind speed climate change indicators for the standardized soil parameter set in combination with meteorological station historical records and satellite remote sensing data. If the precipitation for seven consecutive days exceeds twice the historical average value for the same period calculated based on the local ten-year average, or the average daily temperature fluctuation exceeds the preset threshold of ten degrees Celsius, it is judged as an extreme weather event and the time node is marked to generate a climate change and extreme weather mark set.

[0049] Based on dynamic soil quality diagnosis, this implementation method further integrates climate change factors and extreme weather event identification mechanisms, thereby enhancing the model's sensitivity and predictive ability to respond to external disturbances and improving the integrity and timeliness of dynamic soil quality simulation.

[0050] After constructing a standardized set of soil parameters, the team then accessed an external meteorological database and a high-resolution remote sensing platform to obtain historical and current meteorological observation data corresponding to the geographic coordinates of the target area. Key meteorological factors included daily precipitation, average daily temperature, relative humidity, and wind speed. By developing a meteorological indicator fluctuation model, these parameters were identified as key external drivers of soil quality evolution.

[0051] Judgment rules for extreme precipitation events

[0052] First, the system calculates the seven-day average cumulative precipitation for the region on the same date (e.g., May 10th) over the ten years preceding the target date. This is used as the historical average precipitation baseline. If the daily precipitation total for seven consecutive days during the target period exceeds twice this historical baseline, the period is defined as an extreme precipitation event.

[0053] In practice, the system calculates the average of ten years of historical data. This means the daily precipitation in the region from May 4th to May 10th is summed and averaged, and this average is referred to as the "historical seven-day precipitation average." This average is then compared with the actual cumulative precipitation from May 4th to May 10th of the current year. If the current value is greater than twice the historical average, the system marks the period as an "extreme precipitation event."

[0054] Abnormal temperature fluctuation identification method

[0055] To identify abnormal temperature fluctuations, the system calculates the difference between the highest and lowest temperatures daily as the daily temperature range. If the average daily temperature range for any seven consecutive days exceeds 10 degrees Celsius, which is the system's preset extreme fluctuation threshold, the current climate is considered unstable and highly volatile, potentially significantly affecting soil water evaporation, organic matter decomposition, and microbial activity.

[0056] The default threshold is 10°C, based on the standards of the Agricultural Climate Data Center, and can be adjusted between 8°C and 15°C, with fine-tuning based on the climate resilience of different regions. Temperature data comes from national meteorological stations or ground-based automatic weather stations.

[0057] Climate change and extreme event marker collection construction

[0058] In the above detection process, each time period generated by the system that meets the definition of an "extreme precipitation event" or "high temperature fluctuation event" is recorded as an "extreme climate event node." Each node contains the following fields: event type, start and end dates, the multiple or magnitude of the excess over the baseline value, and the corresponding affected soil parameter category (such as moisture content, organic matter content, etc.), forming a "climate change and extreme weather marker set."

[0059] This set serves as an additional input variable for the soil quality dynamic simulation model, assisting the model training phase in identifying abnormal disturbance responses in soil quality changes and improving the model's generalization ability under climate change.

[0060] Data fusion mechanism and parameter optimization path

[0061] Ultimately, this set of climate factors, along with a set of standardized soil parameters, is fed into a neural network model. A climate input channel is then expanded at the network's input layer, forming a time-synchronized input matrix. The model learns the interactions between soil responses and climate events, optimizing weight distribution through backpropagation to enable it to adapt to the nonlinear effects of extreme weather disturbances.

[0062] Through this integrated approach, not only can the scenario integrity of soil quality simulation be improved, but climate risk factors that may lead to soil degradation can also be identified in advance, providing more accurate and dynamic forecasting support for agricultural management.

[0063] The precipitation anomaly threshold (twice the historical average for the same period) is used to judge extreme precipitation events. The cumulative daily precipitation data for the target area in the same period of the previous ten years (for example, the seven days including May 10) is averaged. If the total precipitation in the same period of the current year exceeds twice this average, it is marked as "extreme precipitation". "Twice the historical average" can reflect the significant impact of sudden heavy precipitation on soil permeability and fertilizer loss.

[0064] The temperature fluctuation threshold (daily average temperature difference exceeding 10°C) is an indicator used to identify strong temperature fluctuations. The difference between the highest and lowest temperatures of each day is calculated to obtain the "daily temperature difference". If the average daily temperature difference for seven consecutive days exceeds 10 degrees Celsius, it is marked as a "high temperature fluctuation event". A large amount of agricultural meteorological data shows that a continuous temperature difference of 10°C will significantly interfere with the surface water evaporation rate, microbial activity and fertility absorption, which is the standard limit value for abnormal climate fluctuations in the agricultural field.

[0065] This implementation effectively identifies and labels extreme climate events (such as heavy rain and dramatic temperature fluctuations) by integrating historical meteorological data and remote sensing observations with a standardized set of soil parameters. This enhances the stability and predictive accuracy of the soil quality dynamic diagnostic model in response to sudden environmental disturbances. By setting scientific climate thresholds, potential risks can be identified in advance, effectively improving the timeliness of early warnings and the relevance of agricultural management. The system also possesses strong adaptability and regional scalability.

[0066] Generating dynamic simulation data for soil quality based on a standardized set of soil parameters also involves obtaining agricultural management records of tillage methods, crop rotation cycles, fertilization frequency, and irrigation patterns from the farmland management database, constructing a numerical coding system for tillage methods, and determining the quantitative weight coefficients of different management measures based on historical record analysis to generate a set of quantitative agricultural management parameters for subsequent soil quality impact assessments.

[0067] This implementation further incorporates agricultural management factors, building on a soil quality diagnostic model established using remote sensing and sensor networks. By structured encoding and modeling historical farming practices, the model's ability to simulate the impact of human agricultural intervention is enhanced. First, the system extracts multiple agricultural management parameters from a farmland management database, including tillage methods, crop rotation cycles, fertilization frequency, and irrigation patterns. To facilitate modeling, this management information is uniformly encoded as numerical values. For example, within the tillage method, no-till is coded as 1, rotary tillage as 2, and deep plowing as 3. Crop rotation cycles are categorized by years, such as one year as 1, two to three years as 2, and four or more years as 3. Fertilization frequency is categorized by the number of applications per year, with codes 1, 2, and 3 corresponding to 1, 2, and 4 or more times. Irrigation patterns are also coded based on whether they are manual, automatic, or non-irrigation. The system then leverages a large amount of existing historical farmland sample data to analyze the actual impact of these various management behaviors on soil quality improvements. This analysis utilizes regression modeling techniques. By comparing management behavior codes with changes in soil quality, the system calculates the degree of positive or negative impact of each management practice on soil improvement, thereby deriving corresponding impact weights. For example, if historical data shows that no-tillage management significantly improves soil organic matter content compared to deep tillage, the system will assign a higher positive impact weight to no-tillage. The system multiplies the current management behavior code for each plot of land by the weights derived from the regression model, and the sum total is calculated to generate a composite score, known as the agricultural management score. This score reflects the potential impact of the current management strategy on soil quality. To ensure consistency with other input features, the system also normalizes the score by subtracting the historical mean and then dividing it by the standard deviation, converting it into a dimensionless, standardized score. Ultimately, the agricultural management score, along with a standardized set of previously processed soil physical and chemical indicators and a set of climate change markers, constitutes the complete input feature set that is fed into the deep prediction model. The model comprehensively assesses future soil quality trends based on current soil conditions, external climate conditions, and management practices, providing forecasts that better align with real-world farming conditions. The definition of management behavior categories is based on agricultural technical standards and local agricultural bureau guidelines. The weight coefficients are obtained through training with historical plot monitoring data for more than 5 years. The model accuracy is comprehensively evaluated through cross-validation, mean square error, and determination coefficient to ensure the interpretability and prediction effectiveness of management scoring parameters.

[0068] This implementation enhances the agronomic relevance of the soil quality prediction model by incorporating agricultural management records to structure and quantify management factors such as tillage practices, fertilization, irrigation, and crop rotation. Regression analysis extracts the impact weights of different management measures, enabling the model to accurately reflect the long-term impact of human farming practices on soil evolution. This improves the scientific nature, relevance, and management guidance of the simulation. It also integrates agricultural data with remote sensing soil information, expanding the system's application in precision agriculture.

[0069] Generating dynamic simulation data of soil quality based on a standardized soil parameter set also includes using a standardized soil parameter set and a set of climate change and extreme weather markers, and employing a multivariate regression analysis method to construct a correlation model between climate change factors and soil physical and chemical properties. If the monthly precipitation change rate exceeds a preset threshold of 30 percent, the soil moisture influence coefficient is adjusted; if the daily average temperature fluctuation exceeds 10 degrees Celsius, the organic matter decomposition rate parameter is corrected; and if an extreme weather marker is detected, a short-term impact factor is added to generate a climate-soil coupling correlation matrix.

[0070] To further improve the accuracy and flexibility of soil quality prediction, this system incorporates climate change factors, combines a standardized set of soil parameters with climate change data, and constructs a dynamic climate-soil coupling matrix. This model can adjust key soil quality parameters based on real-time climate change and extreme weather conditions, achieving more accurate dynamic simulations of soil quality.

[0071] Data preprocessing and input feature construction

[0072] First, the system uses a set of standardized soil parameters and a set of climate change markers as input features. These parameters include key physical and chemical indicators such as soil temperature, moisture, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations; climate change factors, including the monthly rate of change in precipitation and the amplitude of daily average temperature fluctuations; and a set of extreme weather markers, which mark the occurrence of events such as extreme precipitation and temperature fluctuations.

[0073] All these data were standardized before being input into the regression analysis model to ensure that they had the same dimension and scale.

[0074] Multiple regression analysis and association model construction

[0075] The system uses multiple regression analysis to construct a correlation model between climate change factors and soil physical and chemical properties. Specifically, the goal of the regression model is to establish a mathematical relationship between climate factors (such as precipitation and temperature fluctuations) and soil properties (such as moisture and organic matter content). The form of the regression model is:

[0076] ;

[0077] in, represents the predicted value of soil quality (such as moisture, organic matter, etc.), represents climate-related variables (such as precipitation, temperature fluctuations, etc.), is the regression coefficient, which represents the weight of each climate factor on soil characteristics. is the error term in the regression model.

[0078] The regression coefficients were trained using historical data and used to deduce the specific effects of climate change on soil parameters.

[0079] Correction and adjustment mechanism

[0080] In the process of generating the climate-soil coupling correlation matrix, the system sets multiple thresholds to correct the parameters in the soil quality model to cope with the immediate impact of climate change. The specific correction mechanism is as follows:

[0081] Correction of monthly precipitation change rate:

[0082] If the monthly rate of change in precipitation exceeds a preset threshold (30%), meaning there's a significant change in precipitation compared to the previous month, the system automatically adjusts the soil moisture impact coefficient. Specifically, if precipitation increases, soil moisture will also increase, so the moisture impact coefficient is adjusted positively based on the change in precipitation, and negatively based on the change in precipitation.

[0083] Correction for daily average temperature fluctuation:

[0084] If the daily average temperature fluctuates beyond a set threshold (10°C), indicating a significant temperature change, the system will adjust the soil organic matter decomposition rate parameters. Large temperature fluctuations typically lead to changes in soil organic matter decomposition rates, and the system will adjust the predicted organic matter decomposition rate to reflect the impact of this temperature fluctuation on soil microbial activity.

[0085] Extreme Weather Marker Fixes:

[0086] When the system detects extreme weather markers, such as extreme precipitation or temperature fluctuations, it applies a short-term impact factor to immediately correct soil quality trends. For example, extreme precipitation can cause soil erosion or nutrient loss, increasing the system's sensitivity to soil quality changes in the short term and adjusting its predictions of soil texture and nutrient status.

[0087] Generate climate-soil coupling correlation matrix

[0088] Ultimately, all climate factors, soil parameters, and corrected influencing coefficients are combined into a climate-soil coupling matrix. This matrix describes the coupling relationship between climate factors and soil physical and chemical properties. It can dynamically adjust soil quality predictions under different climate scenarios, providing more accurate decision support for agricultural managers.

[0089] This implementation method closely integrates the dynamic relationship between climate factors and soil quality by introducing climate change factors and extreme weather event markers, thereby improving the accuracy and real-time performance of soil quality predictions. The impact of climate change on soil quality, especially the short-term impact under extreme weather conditions, is corrected by key parameters such as soil moisture and organic matter decomposition rate, enabling the model to respond to environmental changes in a timely manner, ensuring that the prediction results are more accurate and reliable. This climate-soil coupled correlation model not only improves the model's adaptability to climate disturbances, but also provides agricultural management with a more accurate climate impact assessment, helping to improve the risk resistance of agricultural production and optimize farming management and decision-making processes.

[0090] Generating dynamic simulation data of soil quality based on a set of standardized soil parameters also includes constructing a cumulative effect model of tillage methods on soil quality based on a set of quantitative agricultural management parameters using time series analysis methods, calculating the cumulative impact weights of different tillage measures on monthly, quarterly, and annual time scales based on historical management records, and generating a long-term agricultural management effect prediction model for multi-factor comprehensive simulation.

[0091] To improve the accuracy and adaptability of dynamic soil quality simulations, this implementation introduces a cumulative effect model of tillage practices based on a set of quantitative agricultural management parameters. Time series analysis is applied to capture and predict the long-term cumulative impacts of different tillage practices on monthly, quarterly, and annual timescales. This approach helps the system better understand the long-term impacts of different tillage practices on soil quality and further optimizes soil quality predictions through multi-factor integrated simulations.

[0092] First, the system extracts agricultural management parameters from the farmland management database, including tillage methods, fertilization frequency, crop rotation cycle, and irrigation patterns. Based on these parameters, it constructs a set of quantitative agricultural management parameters. Each management behavior (such as no-till, rotary tillage, deep plowing, etc.) is quantified into a numerical code. These codes are recorded in the system over time, forming a time series dataset that reflects the ongoing impact of agricultural management measures at different time scales (such as monthly, quarterly, and annual).

[0093] To model the cumulative effects of tillage practices, the system uses time series analysis methods, such as autoregressive or exponential smoothing models, to examine the long-term impact of each tillage practice on soil quality. These tillage practices, through specific weighting coefficients, exert varying degrees of influence on soil quality changes. Time series analysis methods help the system analyze and capture the cumulative effects of tillage practices at different time scales. For example, on a monthly time scale, the impact of tillage practices on soil quality is relatively short-term, while on a quarterly or annual time scale, the impact is more significant and lasts longer.

[0094] Based on historical data, the system calculates the cumulative impact of each tillage practice at different time scales. For example, the system evaluates how management measures taken each month, quarter, and year affect key soil parameters such as soil moisture, organic matter content, and soil pH. By reviewing historical data, the system analyzes the long-term effects of these tillage practices on soil quality and generates impact weights. These weight coefficients help the system reflect the long-term impact of different tillage practices on soil quality at different time scales. After calculating the monthly, quarterly, and annual impacts of each tillage practice, the system integrates these data into a long-term agricultural management effect prediction model. This model can predict how future changes in tillage practices will affect soil quality, especially the cumulative effects of these management practices at different time scales.

[0095] The system also combines quantitative agricultural management parameters with other factors influencing soil quality (such as climate change and soil physical and chemical properties) to conduct multi-factor comprehensive simulations, further enhancing the comprehensiveness and accuracy of soil quality predictions. This multi-factor comprehensive simulation can help agricultural managers better understand the combined impacts of different agricultural management practices, climate change, and soil characteristics on soil quality, thereby providing more accurate decision-making support for agricultural management.

[0096] This implementation combines quantitative agricultural management parameters with time series analysis to construct a model for the cumulative effects of tillage practices on soil quality. This model comprehensively captures the impact of agricultural management practices across different timescales. By analyzing monthly, quarterly, and annual timescales, this method enables more refined and long-term soil quality predictions and accurately simulates the cumulative impacts of different tillage practices on soil quality, thereby improving the practicality and scientific nature of soil quality management. By integrating multiple factors, such as agricultural management and climate change, the system can provide agricultural managers with more reliable and long-term decision support, helping to improve the sustainability and efficiency of agricultural production.

[0097] Generating dynamic simulation data of soil quality based on a set of standardized soil parameters also includes integrating the climate-soil coupling correlation matrix and the long-term effect prediction model of agricultural management, and using a neural network algorithm to construct a dynamic simulation framework of soil quality with multiple factor interactions. If extreme weather markers are detected, the emergency response mechanism is activated, and the weights of soil moisture and organic matter decomposition rate parameters are adjusted to generate dynamic simulation data of soil quality.

[0098] In this implementation, the system integrates a climate-soil coupling correlation matrix with a long-term agricultural management effect prediction model, employing a neural network algorithm to construct a dynamic soil quality simulation framework that integrates multiple factors. This framework comprehensively considers the interactions among multiple factors, including climate, soil properties, and agricultural management practices, to accurately predict soil quality trends in real time. Furthermore, when extreme weather events are detected, the system activates an emergency response mechanism to further optimize simulation results by adjusting the weights of key soil quality parameters (such as soil moisture and organic matter decomposition rate) to address the impact of climate anomalies on soil quality.

[0099] Integrating climate-soil coupling correlation matrix with agricultural management long-term effect prediction model

[0100] First, the system combines a climate-soil coupling matrix with a long-term agricultural management effect prediction model. The climate-soil coupling matrix is ​​derived by modeling the relationship between climate change factors (such as precipitation and temperature fluctuations) and soil physical and chemical properties (such as moisture and organic matter content). The long-term agricultural management effect prediction model analyzes the long-term impact of agricultural management measures such as tillage practices, fertilization frequency, and crop rotation on soil quality.

[0101] The two models are combined to form a multi-dimensional input feature set that includes dynamic information on climate, soil, agricultural management, and other factors. This feature set will provide a multi-level, comprehensive foundation for the neural network to predict soil quality.

[0102] A framework for dynamic simulation of soil quality with multi-factor interactions constructed using neural network algorithms

[0103] By integrating the climate-soil coupling correlation matrix with a long-term agricultural management effect prediction model, the system uses a neural network algorithm to construct a dynamic soil quality simulation framework that accounts for multiple interactive factors. The neural network model receives input from multiple sources, including soil parameters, climate change data, and agricultural management practices. It learns the complex interactions between these factors through architectures such as multilayer perceptrons or recurrent neural networks.

[0104] The training process of the neural network includes the following steps: input layer: receiving standardized soil parameters, climate factors, agricultural management data, etc.; hidden layer: simulating the interaction and coupling effects between different factors through calculations of multiple layers of neurons; output layer: generating dynamic simulation data of soil quality, including predicted values ​​such as soil moisture, organic matter content, nitrogen, phosphorus and potassium concentrations.

[0105] Through the back-propagation algorithm, the neural network continuously optimizes the weight coefficients to minimize the error between the actual observation value and the predicted value, thereby achieving efficient and accurate dynamic simulation of soil quality.

[0106] Extreme Weather Emergency Response Mechanism

[0107] When the system detects extreme weather events (such as heavy rain or extreme temperature fluctuations), the model automatically activates emergency response mechanisms. Extreme weather events often have sudden impacts on soil quality, particularly changes in soil moisture and organic matter decomposition rates. Therefore, the system adjusts the weights of these parameters to promptly respond to the impact of climate anomalies.

[0108] Specifically, if precipitation increases abnormally or daily temperature fluctuations exceed the preset threshold, the system will make adjustments in the following ways: Soil moisture adjustment: In extreme precipitation conditions, the system will increase the impact weight of soil moisture to reflect the short-term impact of increased soil moisture on soil quality; Organic matter decomposition rate adjustment: If the temperature fluctuates greatly, the system will increase the impact of temperature changes on microbial activity and organic matter decomposition by adjusting the weight of the organic matter decomposition rate parameters.

[0109] These adjustments will take effect in real time in the model, ensuring that the soil quality simulation data can reflect the real changes in the current environment.

[0110] Generate soil quality dynamic simulation data

[0111] By integrating multi-factor interactions and emergency response mechanisms, the system ultimately generates dynamic soil quality simulation data that comprehensively reflects the impact of various factors on soil quality and provides accurate forecasts of soil quality trends. This data not only helps agricultural managers understand soil conditions in real time but also provides a scientific basis for future soil management strategies.

[0112] This implementation method integrates the climate-soil coupling correlation matrix with the long-term agricultural management effect prediction model to construct a multi-factor interactive soil quality dynamic simulation framework, which can more comprehensively and accurately predict changes in soil quality. The neural network algorithm enables the system to identify the complex relationships between climate, soil, and agricultural management factors, thereby improving the accuracy of soil quality predictions. The introduction of an emergency response mechanism enables the system to promptly adjust key parameters (such as soil moisture and organic matter decomposition rate) when extreme weather events occur, thereby avoiding the impact of climate anomalies on soil quality simulation results. Through this method, agricultural managers can obtain more accurate and real-time soil quality prediction data, providing more scientific and reliable support for decision-making, and enhancing the sustainability and risk resistance of agricultural production.

[0113] The specific formula for uncertainty analysis is:

[0114] ;

[0115] in, represents the predicted soil quality evolution function, Indicates the number of sampling times, Indicates the The entropy weight factor corresponding to the sample is represents the nonlinear model function used for soil state assessment, Indicates the Set input dataset, represents the model structure parameters, Indicates the corresponding The perturbation sensitivity coefficient of samples, Indicates the The variance of the set of sample input variables, Indicates the sample number.

[0116] By fusing multi-source remote sensing data with sample data and employing uncertainty analysis techniques, the dynamic inversion and prediction of cultivated land soil quality is performed. First, the method combines data from multiple sensors to assess soil status using a nonlinear model, adjusting it based on the weight and impact of each sample. Through the attenuation function, the system accounts for the temporal decay of soil quality, while adjusting the model based on the sensitivity of different input changes to more accurately predict changes in soil quality. This process combines multi-source data with sample information, utilizing uncertainty analysis to improve the accuracy of soil quality predictions and enhance the adaptability and reliability of the model. Ultimately, the generated dynamic simulation data for soil quality can provide more accurate decision-making support for agricultural management, helping to effectively address spatiotemporal changes in soil quality and the impact of external factors.

[0117] By utilizing the synergy of multi-source data and uncertainty analysis, the accuracy and reliability of soil quality predictions have been effectively improved, providing strong technical support for precision agriculture, especially in the face of complex soil quality changes and extreme climate changes, with better adaptability and operability.

[0118] The calculation formula is:

[0119] ;

[0120] in, Indicates the The entropy weight factor corresponding to the sample is represents the weighted normalization factor of all samples, represents the input variable dimension, Indicates the The sample in The fuzzy membership of the indicators, Indicates the The sample in The normalized value of the indicator, Indicates the The average value of the indicators in the whole sample, Indicates the sample number, Represents the dimension number index.

[0121] By calculating the weight coefficient of each sample, its contribution to the soil quality prediction model is dynamically adjusted. First, the system evaluates the performance of each sample on different soil quality-related indicators, and calculates the weight coefficient by combining the standardized value of the sample on each indicator with the average value of all samples. Specifically, by weighting the fuzzy coefficient of each sample in multiple dimensions, its relative importance in each indicator is taken into account. During the calculation process, the system uses a normalization factor to standardize the samples to ensure that all samples have a consistent scale when weighted. Ultimately, the weight coefficient of the sample reflects its influence on the overall soil quality prediction results, and can dynamically adjust the weight according to the performance of the sample on different indicators, thereby improving the accuracy and adaptability of the model prediction. In this way, the system can more accurately evaluate the contribution of different samples to soil quality and achieve a more reasonable weight distribution in soil quality prediction.

[0122] The calculation formula is:

[0123] ;

[0124] in, Indicates the corresponding The perturbation sensitivity coefficient of samples, represents the total number of soil input features, Indicates the dimension number of the input variable currently traversed, Represents the model function B on the input variables The partial derivative of represents the input variable, Indicates the The sample in The degree of fluctuation on the input variables, Indicates that all samples in The maximum fluctuation value of a variable.

[0125] The dynamic changes in soil quality are evaluated by calculating the sensitivity of each sample to disturbances. Specifically, the system evaluates the sensitivity of each sample to changes in different characteristics by analyzing the response of the soil model to different input characteristics. First, the system calculates the partial derivatives of each sample on different input characteristics, reflecting the degree of influence of the input characteristics on soil quality prediction. Then, the contribution of the sample to the overall prediction is adjusted based on the relative change amplitude and maximum fluctuation range of these characteristics. In this way, the system can identify which input characteristics have a greater impact on soil quality changes, and adjust the weights of the model according to the sensitivity of these characteristics, thereby improving the accuracy and adaptability of soil quality prediction. Finally, the system takes into account the sensitivity of each sample to dynamically adjust the input of the soil quality simulation model to ensure that the model can accurately reflect the impact of different input changes on soil quality.

[0126] The early warning mechanism uses color classification to display the risk level, which is divided into three levels: normal, mild warning, and severe warning. The early warning results are synchronized to the plot view of the agricultural management platform in the form of a graphical interface.

[0127] First, a multi-source remote sensing and ground sensor network monitors farmland soil quality in real time, collecting key soil parameters (such as moisture, organic matter content, pH, and nitrogen, phosphorus, and potassium concentrations) as well as climate change data (such as temperature and precipitation). The system then analyzes and processes this data. Based on this data, the system calculates soil quality trends and assesses soil health.

[0128] During the analysis process, the system evaluates changes in soil quality based on set thresholds. Based on the magnitude of the change in soil quality, the system classifies the risk level into three categories:

[0129] Normal: When soil quality is stable or the variation range is within a safe range, the system determines it as normal and displays green, indicating that the soil quality is healthy and no special measures are required.

[0130] Mild warning: When there are slight changes in soil quality and these changes may pose potential risks to agricultural production, the system will issue a mild warning and display yellow, prompting agricultural managers to pay attention to the soil conditions of the land and may need to take appropriate management measures, such as timely fertilization or adjustment of irrigation methods.

[0131] Serious warning: When soil quality deteriorates significantly or there is a major risk (for example, soil fertility is significantly reduced or acidification occurs), the system will issue a serious warning in red, reminding agricultural managers to take immediate and effective measures, such as adjusting farming methods, strengthening fertilization management, or conducting soil remediation, to avoid long-term damage to crop yields and soil ecology.

[0132] All warning results are synchronized in real time to the plot view on the agricultural management platform via a graphical interface. This view displays the risk status of each plot, allowing agricultural managers to clearly and intuitively review the soil quality of each plot and formulate appropriate response measures based on different risk levels. The system may also provide specific action recommendations for different risk levels, such as soil improvement suggestions for minor warnings and emergency remediation plans for severe warnings.

[0133] This early warning mechanism simplifies the complexity of soil quality monitoring through color-grade classification, enabling agricultural managers to quickly identify potential risks and make timely decisions, thereby improving soil management efficiency and ensuring the sustainability of agricultural production.

[0134] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic diagnosis method for cultivated land soil quality based on multi-source remote sensing collaborative inversion, characterized by: The method comprises: Soil temperature, humidity, pH, organic matter content, and nitrogen, phosphorus, and potassium concentrations are collected in real time through a multi-source sensor network. Data preprocessing algorithms are used to remove outliers and perform standardization to generate a standardized soil parameter set for subsequent analysis of the relationship between climate and management factors. Based on a standardized set of soil parameters, dynamic simulation data for soil quality is generated. Uncertainty analysis is performed on this data. One thousand parameter combinations are generated through random sampling, and the probability distribution interval of soil quality changes is calculated. If the prediction accuracy, assessed based on the ratio of the mean square error to the actual observed value, falls below a preset threshold of 85 percent, the system returns to the previous step to readjust the weight parameters of each factor in the neural network framework, generating soil quality change trend prediction data. Based on the predicted data on soil quality trends, an early warning mechanism is established. If the predicted soil fertility level, based on the nitrogen, phosphorus, and potassium comprehensive index, falls below the preset safety threshold (80% of the historical average) for three consecutive months, or if the structural stability index drops by more than 20% of the preset threshold, a risk warning signal will be automatically generated and pushed to the agricultural management decision-making platform, completing the full process analysis from data collection to decision support. Generating soil quality dynamic simulation data according to the standardized soil parameter set includes: Based on the standardized soil parameter set, combined with historical records of meteorological stations and satellite remote sensing data, we obtain climate change indicators such as precipitation, temperature fluctuations, humidity changes, and wind speed. If the precipitation for seven consecutive days exceeds twice the historical average value for the same period calculated based on the local ten-year average, or if the daily average temperature fluctuation exceeds a preset threshold of ten degrees Celsius, it is judged as an extreme weather event and the time node is marked, generating a set of climate change and extreme weather markers. Obtain agricultural management records of tillage methods, crop rotation cycles, fertilization frequency, and irrigation patterns from the farmland management database, build a numerical coding system for tillage methods, and determine the quantitative weight coefficients of different management measures based on historical records analysis to generate a set of agricultural management quantitative parameters for subsequent soil quality impact assessments; Using a standardized set of soil parameters and a set of climate change and extreme weather markers, a multivariate regression analysis method was used to construct a correlation model between climate change factors and soil physical and chemical properties. If the monthly precipitation change rate exceeded a preset threshold of 30 percent, the soil moisture influence coefficient was adjusted. If the daily average temperature fluctuation exceeded 10 degrees Celsius, the organic matter decomposition rate parameter was modified. If extreme weather markers were detected, a short-term impact factor was added to generate a climate-soil coupling correlation matrix. Based on a set of quantitative agricultural management parameters, a time series analysis method was used to construct a model for the cumulative effects of tillage practices on soil quality. The cumulative impact weights of different tillage practices on monthly, quarterly, and annual time scales were calculated based on historical management records. This generated a long-term agricultural management effect prediction model for multi-factor comprehensive simulation. By integrating the climate-soil coupling correlation matrix and the long-term effect prediction model of agricultural management, a neural network algorithm is used to construct a dynamic simulation framework of soil quality with multi-factor interaction. If extreme weather markers are detected, the emergency response mechanism is activated, the weights of soil moisture and organic matter decomposition rate parameters are adjusted, and soil quality dynamic simulation data are generated.

2. The method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion according to claim 1 is characterized by: The specific formula for uncertainty analysis is: ； in, represents the predicted soil quality evolution function, Indicates the number of sampling times, Indicates the The entropy weight factor corresponding to the sample is represents the nonlinear model function used for soil state assessment, Indicates the Set input dataset, represents the model structure parameters, Indicates the corresponding The perturbation sensitivity coefficient of samples, Indicates the The variance of the set of sample input variables, Indicates the sample number.

3. The method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion according to claim 2 is characterized by: described The calculation formula is: ； in, Indicates the The entropy weight factor corresponding to the sample is represents the weighted normalization factor of all samples, represents the input variable dimension, Indicates the The sample in The fuzzy membership of the indicators, Indicates the The sample in The normalized value of the indicator, Indicates the The average value of the indicators in the whole sample, Indicates the sample number, Represents the dimension number index.

4. The method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion according to claim 2 is characterized by: described The calculation formula is: ； in, Indicates the corresponding The perturbation sensitivity coefficient of samples, represents the total number of soil input features, Indicates the dimension number of the input variable currently traversed, Represents the model function B on the input variables The partial derivative of represents the input variable, Indicates the The sample in The degree of fluctuation on the input variables, Indicates that all samples in The maximum fluctuation value of a variable.

5. The method for dynamic diagnosis of cultivated land soil quality based on multi-source remote sensing collaborative inversion according to claim 1 is characterized by: The early warning mechanism uses color classification to display the risk level, which is divided into three levels: normal, mild warning, and severe warning, and synchronizes the early warning results to the plot view of the agricultural management platform in the form of a graphical interface.

Images