A farmland quality monitoring system based on black soil protection

Abstract

The present invention belongs to the technical field of land quality monitoring, and provides a arable land quality monitoring system based on black soil protection, which adopts a layered architecture, including a data acquisition layer, a data transmission layer, a data processing layer and an application service layer, wherein the data acquisition layer is responsible for collecting various physical and chemical indicators of the soil; the data transmission layer transmits the collected data to a data center; the data processing layer analyzes and processes the data; the application service layer provides users with monitoring results and decision support; the present invention collects arable land quality data in real time, and realizes multi-dimensional monitoring of arable land through multiple means such as satellite remote sensing, unmanned aerial vehicles and ground sensors, and performs multi-source data fusion, and the multi-level monitoring system ensures the comprehensiveness and accuracy of the data, so that scientific researchers can timely understand the changing trend of arable land quality, and the subsequent implementation is conducive to protecting the arable land resources of black soil, maintaining ecological balance, and promoting the sustainable development of agriculture.

Description

Technical Field

[0001] The invention belongs to the technical field of land quality monitoring, and in particular is a farmland quality monitoring system based on black soil protection. Background Art

[0002] With the growth of global population and the rise of food demand, agricultural production is facing tremendous pressure. As the basic resource for food production, the quality of arable land directly affects the yield and quality of crops. In particular, black soil, as one of the most important agricultural soils in the world, is highly fertile and rich in organic matter, making it a key area for food production. Black soil usually has a deep humus layer, good water retention capacity and rich nutrients. Its physical structure is stable, which is conducive to the growth of roots and the activity of microorganisms. However, these characteristics also make black soil more susceptible to degradation when it is disturbed by external factors; however, due to long-term over-cultivation, unreasonable land use and environmental pollution, the quality of black soil is gradually declining.

[0003] At present, the degradation of black soil is mainly manifested in soil erosion, organic matter decline, nutrient loss, soil acidification and salinization. These problems not only reduce the fertility of the soil, but also affect the growth and yield of crops; the causes of black soil degradation are diverse, including natural factors such as climate change and natural disasters, as well as human factors such as over-cultivation, excessive use of fertilizers and pesticides, and industrial pollution.

[0004] Therefore, establishing an effective arable land quality monitoring system is crucial for the protection and sustainable utilization of black soil. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a farmland quality monitoring system based on black soil protection to solve the problems raised in the background technology.

[0006] A farmland quality monitoring system based on black soil protection adopts a layered architecture, including a data acquisition layer, a data transmission layer, a data processing layer and an application service layer. The data acquisition layer is responsible for collecting various physical and chemical indicators of the soil; the data transmission layer transmits the collected data to the data center; the data processing layer analyzes and processes the data; and the application service layer provides monitoring results for users.

[0007] The data collection layer also includes a sensor module, a remote sensing monitoring module, a ground penetrating radar module and a mobile laboratory module to collect various physical and chemical indicators of the soil;

[0008] The data transmission layer also includes a wireless transmission module and a wired transmission module, which transmit the collected data through two transmission modes;

[0009] The data processing layer also includes a data preprocessing module, a spatiotemporal data analysis module, a prediction analysis module, and a soil quality assessment module to clean, analyze and store the collected data;

[0010] The application service layer also includes an early warning mechanism module and an interactive module, which provide soil degradation early warning based on real-time monitoring data and the results of analysis by the data processing layer, and provide an interactive interface for users.

[0011] Preferably, the sensor module is responsible for collecting various physical and chemical index data of the soil in real time, including using a spectral sensor to measure the organic matter content of the soil, and using an electrode sensor to measure the pH value and conductivity of the soil;

[0012] The remote sensing monitoring module uses drones to obtain a wide range of soil surface information, including soil moisture and vegetation coverage. By using multispectral or hyperspectral cameras, the drones conduct repeated observations of the same area at different time points to monitor the dynamic changes in soil quality.

[0013] Preferably, the ground penetrating radar module uses ground penetrating radar (GPR) technology to detect the deep soil structure and water content information and assess the degree of soil erosion and groundwater level changes;

[0014] The degree of soil erosion is assessed by using GPR equipment to scan the target area, obtain high-resolution images of the underground soil layer, pre-process the collected GPR data, including denoising and signal enhancement, and then use image processing technology to extract the boundaries and characteristics of the soil layer; by comparing GPR data at different time points, analyze the changes in the soil layer;

[0015] Monitoring of groundwater level changes: Multiple groups of GPR measuring points are set up in the monitoring area, and GPR equipment is used to scan regularly to detect the reflection signal of the groundwater level, obtain the groundwater level change data, and then use time-space series analysis to compare the GPR data at different time points to determine the trend of groundwater level changes;

[0016] The mobile laboratory module is a vehicle integrated with a variety of sensors and analytical instruments, and is used to collect and conduct preliminary analysis of soil samples directly in the field.

[0017] Preferably, the data preprocessing module first performs standardization processing on the data collected by different sensors so that the data can be compared and analyzed on the same scale, which is expressed as:

[0018]

[0019] in, Represents the original data, represents the mean of the data, Indicates the standard deviation of the data;

[0020] Then, different weights are assigned to each data source according to its reliability and accuracy. Remote sensing data with high spatial resolution has a higher weight, while data with low spatial resolution has a lower weight. The weight calculation is expressed as:

[0021]

[0022] in, Indicates The weight of the data source, Indicates The reliability score of each data source, Indicates the total number of data sources;

[0023] Then the weighted average method is used to fuse the multi-source data, which is expressed as: ,in represents the fused data, Indicates Data from the data source, Indicates The weight of each data source;

[0024] Finally, the fused data is input into the Geographic Information System (GIS) to generate a cultivated land quality monitoring map.

[0025] Preferably, the spatiotemporal data analysis module analyzes the time series data of each monitoring point to identify long-term trends and seasonal changes, which are expressed as:

[0026]

[0027] in, Represents a time series at a point in time The value of represents a constant term, represents the autoregressive coefficient, represents the moving average coefficient, represents the white noise term;

[0028] Then spatial interpolation is used to generate the spatial distribution map of soil quality, which is expressed as: ,in Indicates that at unsampled points The estimated value of Indicates that at a known sampling point The observed value of represents the weight coefficient;

[0029] Combined with time and space information, the space-time Kriging method is used for data fusion, which is expressed as:

[0030]

[0031] in, Indicates at location and time The estimated value of Indicates at location and time The observed value of represents the weight coefficient;

[0032] The Gi statistic is then used to identify hotspots of soil quality, which is expressed as:

[0033]

[0034] in, Indicates location of The statistics of Represents distance-based The spatial weight of Indicates location The observed value of represents the mean, S represents the standard deviation, and n represents the number of samples;

[0035] Trend surface analysis is then used to identify the spatial trend of soil quality, expressed as:

[0036]

[0037] Among them, Z represents the response variable, x and y represent the independent variables, represents the regression coefficient, represents the error term; finally, the analysis results are input into GIS to generate a spatiotemporal distribution map of soil quality, showing the changing trend and hot spots of soil quality.

[0038] Preferably, the prediction analysis module calculates the correlation coefficient between each feature and soil degradation, retaining the features with high correlation, which is expressed as: ,in represents the correlation coefficient, X and Y represent the values ​​of two features respectively. , are their means respectively;

[0039] The classification algorithm is then used to predict the risk level of soil degradation, expressed as:

[0040]

[0041] in, Indicates the sample The prediction results, represents the indicator function, Indicates The prediction results of decision trees, n represents the number of decision trees;

[0042] And use deep neural network to process high-dimensional data, its forward propagation is expressed as:

[0043]

[0044] in, Indicates The input of the layer, , denote weight and bias respectively, Represents the activation function; for classification problems, the logistic regression model outputs the probability of each category, which is used to determine the risk level of soil degradation, expressed as: ,in represents the Sigmaid function, represents the weight vector, represents the feature vector, b represents the bias; by setting the threshold, the probability is converted into the risk level, and the prediction results with a probability higher than 0.7 are marked as high-risk areas.

[0045] Preferably, the soil quality assessment module is trained based on historical data and verified by actual observation data to construct a soil quality assessment model, which is expressed as: , where y represents the response variable, represents the weight vector, Represents the feature vector, b represents the bias, the historical data is divided into training set and test set, K-fold cross validation is used to evaluate the stability and generalization ability of the model, and the trained model is verified using the latest actual observation data.

[0046] Preferably, the early warning mechanism module is based on real-time monitoring data and historical trend analysis, that is, the time series data of soil quality analyzed by the spatiotemporal data analysis module and the risk level of soil degradation predicted in the prediction analysis module. According to the degree of soil degradation, the early warning level is divided into 0-0.3 for low risk, 0.4-0.6 for medium risk, and 0.7-1 for high risk. The real-time data is continuously monitored and substituted into the model for prediction, and the corresponding level of early warning is triggered according to the set threshold.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. Through multi-layer architecture and multi-source data fusion, the system of the present invention can comprehensively monitor the physical, chemical, biological activity and environmental impact indicators of the soil, provide comprehensive arable land quality information, collect soil quality data in real time, and promptly discover signs of soil degradation. Based on historical trend analysis and machine learning models, the system can predict future changes in soil quality and issue early warnings.

[0049] 2. The implementation of this system helps to protect the precious arable land resource of black soil, maintain ecological balance, and promote the sustainable development of agriculture. By improving soil quality and optimizing agricultural management measures, the system can improve the productivity of arable land and increase crop yields.

[0050] 3. The present invention can "consult" for specific plots through this system and provide accurate soil improvement and fertilization plans. By obtaining high-precision data such as erosion barriers, crop growth, elevation slope, soil fertility, etc. in the plot, combined with the laws of water and fertilizer migration, personalized management measures can be formulated for the plot. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a block diagram of the farmland quality monitoring system of the present invention. DETAILED DESCRIPTION

[0052] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0053] As attached Figure 1 As shown:

[0054] Embodiment 1: The present invention provides a arable land quality monitoring system based on black soil protection, which adopts a layered architecture, including a data acquisition layer, a data transmission layer, a data processing layer and an application service layer. The data acquisition layer is responsible for collecting various physical and chemical indicators of the soil; the data transmission layer transmits the collected data to the data center; the data processing layer analyzes and processes the data; the application service layer provides monitoring results to users; at the same time, a detection index system is constructed, including soil physical indicators, soil chemical indicators, biological activity indicators and environmental impact indicators. Its soil physical indicators include but are not limited to soil texture, structure, density, porosity and permeability, etc., which reflect the water holding capacity and ventilation conditions of the soil, which are crucial to crop growth; soil chemical indicators include soil pH value, electrical conductivity, organic matter content, nutrient (such as nitrogen, phosphorus, potassium) content, etc., which are directly related to soil fertility and crop nutrient absorption; biological activity indicators involve the number and diversity of microorganisms in the soil, enzyme activity and root health status, etc., reflecting the health of the soil ecosystem and biodiversity; environmental influencing factors include climate change, land use methods, agricultural management measures, etc.

[0055] The data collection layer also includes sensor modules, remote sensing monitoring modules, ground penetrating radar modules and mobile laboratory modules to collect various physical and chemical indicators of the soil; it is responsible for real-time collection of indicators such as soil temperature, humidity, pH value, and organic matter content.

[0056] The data transmission layer also includes a wireless transmission module and a wired transmission module, which transmit the collected data through two transmission methods; a wireless sensor network is used to transmit the collected data to the data center in real time, and a wired method is combined to ensure the stability and reliability of data transmission.

[0057] The data processing layer also includes a data preprocessing module, a spatiotemporal data analysis module, a prediction analysis module, and a soil quality assessment module to clean, analyze, and store the collected data;

[0058] The application service layer also includes an early warning mechanism module and an interactive module, which provide soil degradation early warning based on real-time monitoring data and the results of analysis by the data processing layer, as well as an interactive interface for users.

[0059] The sensor module is responsible for collecting various physical and chemical index data of the soil in real time, including using spectral sensors to measure the organic matter content of the soil and electrode sensors to measure the pH value and conductivity of the soil;

[0060] The remote sensing monitoring module uses drones to obtain a wide range of soil surface information, including soil moisture and vegetation coverage. By using multispectral or hyperspectral cameras, the drones can repeatedly observe the same area at different time points to monitor the dynamic changes in soil quality.

[0061] The ground penetrating radar module uses ground penetrating radar (GPR) technology to detect deep soil structure and moisture content information, and assess soil erosion and groundwater level changes;

[0062] The degree of soil erosion is assessed by using GPR equipment to scan the target area, obtain high-resolution images of the underground soil layer, pre-process the collected GPR data, including denoising and signal enhancement, and then use image processing technology to extract the boundaries and characteristics of the soil layer; by comparing GPR data at different time points, the changes in the soil layer are analyzed. If the thickness of the soil layer decreases or the structure changes significantly, it may be a sign of soil erosion.

[0063] To monitor the change of groundwater level, multiple groups of GPR measuring points are deployed in the monitoring area, and GPR equipment is used to scan regularly to detect the reflection signal of the groundwater level and obtain the change data of the groundwater level. Then, time-space series analysis is used to compare the GPR data at different time points to determine the trend of groundwater level changes.

[0064] The mobile laboratory module is a vehicle that integrates a variety of sensors and analytical instruments. It is used to collect and conduct preliminary analysis of soil samples directly in the field, which improves the efficiency and flexibility of data collection and is particularly suitable for rapid monitoring of large areas of cultivated land.

[0065] The data preprocessing module uses multi-source data fusion technology to integrate the data collected by the four modules in the data acquisition unit to generate a high-precision farmland quality monitoring map; for the data collected by different sensors, standardization is first performed to make the data comparable and analyzed on the same scale, which is expressed as:

[0066]

[0067] in, Represents the original data, represents the mean of the data, Indicates the standard deviation of the data;

[0068] Then, different weights are assigned to each data source according to its reliability and accuracy. Remote sensing data with high spatial resolution has a higher weight, while data with low spatial resolution has a lower weight. The weight calculation is expressed as:

[0069]

[0070] in, Indicates The weight of the data source, Indicates The reliability score of each data source, Indicates the total number of data sources;

[0071] Then the weighted average method is used to fuse the multi-source data, which is expressed as: ,in represents the fused data, Indicates Data from the data source, Indicates The weight of each data source;

[0072] Finally, the fused data is input into the geographic information system (GIS), and the spatial distribution of cultivated land quality is displayed using visualization tools to generate a cultivated land quality monitoring map. The above can effectively integrate data from different sensors and platforms to improve the accuracy and reliability of cultivated land quality monitoring.

[0073] The spatiotemporal data analysis module analyzes the time series data of each monitoring point to identify long-term trends and seasonal changes, which are expressed as:

[0074]

[0075] in, Represents a time series at a point in time The value of represents a constant term, represents the autoregressive coefficient, represents the moving average coefficient, represents the white noise term;

[0076] Then spatial interpolation is used to generate the spatial distribution map of soil quality, which is expressed as: ,in Indicates that at unsampled points The estimated value of Indicates that at a known sampling point The observed value of represents the weight coefficient;

[0077] Combined with time and space information, the space-time Kriging method is used for data fusion, which is expressed as:

[0078]

[0079] in, Indicates at location and time The estimated value of Indicates at location and time The observed value of represents the weight coefficient;

[0080] The Gi statistic is then used to identify hotspots of soil quality, which is expressed as:

[0081]

[0082] in, Indicates location of The statistics of Represents distance-based The spatial weight of Indicates location The observed value of represents the mean, S represents the standard deviation, and n represents the number of samples;

[0083] Trend surface analysis is then used to identify the spatial trend of soil quality, expressed as:

[0084]

[0085] Among them, Z represents the response variable, x and y represent the independent variables, represents the regression coefficient, represents the error term; finally, the analysis results are input into GIS to generate a spatiotemporal distribution map of soil quality, showing the changing trend and hot spots of soil quality. The spatiotemporal data analysis module can effectively reveal the changing patterns of soil quality over time and space, identify degradation hot spots and trends, and provide a scientific basis for farmland protection and management.

[0086] The prediction analysis module performs feature selection through correlation analysis, that is, by calculating the correlation coefficient between each feature and soil degradation, the features with high correlation are retained, which is expressed as: ,in represents the correlation coefficient, X and Y represent the values ​​of two features respectively. , are their means respectively;

[0087] The classification algorithm is then used to predict the risk level of soil degradation, expressed as:

[0088]

[0089] in, Indicates the sample The prediction results, represents the indicator function, Indicates The prediction results of decision trees, n represents the number of decision trees;

[0090] And use deep neural network to process high-dimensional data, its forward propagation is expressed as:

[0091]

[0092] in, Indicates The input of the layer, , denote weight and bias respectively, Represents the activation function; for classification problems, the logistic regression model outputs the probability of each category, which is used to determine the risk level of soil degradation, expressed as: ,in represents the Sigmaid function, represents the weight vector, represents the feature vector, and b represents the bias; by setting the threshold, the probability is converted into the risk level, and the prediction results with a probability higher than 0.7 are marked as high-risk areas. The prediction analysis module can effectively extract valuable information from a large amount of complex data and accurately predict the risk level of soil degradation. The soil quality assessment module is trained based on historical data and verified by actual observation data to construct a soil quality assessment model, which is expressed as: , where y represents the response variable, represents the weight vector, Represents the feature vector, b represents the bias, and the historical data is divided into a training set and a test set. K-fold cross validation is used to evaluate the stability and generalization ability of the model, and the trained model is verified using the latest actual observation data. By analyzing the prediction results of the model, we can understand which features have the greatest impact on land quality, and use the latest actual observation data to verify the trained model. The hyperparameters of the model are adjusted through grid search, random search and other methods to optimize the model performance. The verified model is then applied to the actual soil quality assessment, so that the soil quality assessment module can be effectively trained based on historical data, and strictly verified through actual observation data to build an accurate soil quality assessment model.

[0093] The early warning mechanism module is based on real-time monitoring data and historical trend analysis, that is, the time series data of soil quality analyzed by the spatiotemporal data analysis module and the risk level of soil degradation predicted in the prediction analysis module. According to the degree of soil degradation, the early warning level is divided into 0-0.3 for low risk, 0.4-0.6 for medium risk, and 0.7-1 for high risk. The real-time data is continuously monitored and substituted into the model for prediction, and the corresponding level of early warning is triggered according to the set threshold. And the response measures for the corresponding early warning are set. For low-risk responses, preventive suggestions are provided, including improving farming methods and increasing the use of organic fertilizers; for medium-risk responses, irrigation strategies are adjusted, soil and water conservation measures are taken, etc.; for high-risk responses, emergency measures are taken, including restricting land use and implementing ecological restoration projects. The early warning mechanism module can issue an early warning of soil degradation based on real-time monitoring data and historical trend analysis, and take corresponding response measures according to the severity of soil degradation, effectively protecting cultivated land resources and promoting sustainable agricultural development.

[0094] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those features that are not relevant to implementing the invention).

[0095] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, without undue experimentation, the development effort will be a routine task of design, fabrication, and production.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A farmland quality monitoring system based on black soil protection, characterized in that: A layered architecture is adopted, including data collection layer, data transmission layer, data processing layer and application service layer. The data collection layer is responsible for collecting various physical and chemical indicators of soil; the data transmission layer transmits the collected data to the data center; the data processing layer analyzes and processes the data; and the application service layer provides monitoring results to users; The data collection layer also includes a sensor module, a remote sensing monitoring module, a ground penetrating radar module and a mobile laboratory module to collect various physical and chemical indicators of the soil; The data transmission layer also includes a wireless transmission module and a wired transmission module, which transmit the collected data through two transmission modes; The data processing layer also includes a data preprocessing module, a spatiotemporal data analysis module, a prediction analysis module, and a soil quality assessment module to clean, analyze and store the collected data; The spatiotemporal data analysis module analyzes the time series data of each monitoring point to identify long-term trends and seasonal changes, which are expressed as: ； in, represents the value of the time series at time point t, represents a constant term, represents the autoregressive coefficient, represents the moving average coefficient, represents the white noise term; Then spatial interpolation is used to generate the spatial distribution map of soil quality, which is expressed as: ,in Indicates that at unsampled points The estimated value of Indicates that at a known sampling point The observed value of represents the weight coefficient; Combined with time and space information, the space-time Kriging method is used for data fusion, which is expressed as: ； in, represents the estimated value at position s and time t, Indicates at location and time The observed value of represents the weight coefficient, Represents the index of the time dimension, and m represents the number of known time sampling points; The Gi statistic is then used to identify hotspots of soil quality, which is expressed as: ； in, Indicates the position ib The statistics of represents the spatial weight based on the distance d, Indicates location The observed value of represents the mean, S represents the standard deviation, represents the number of samples; Trend surface analysis is then used to identify the spatial trend of soil quality, expressed as: ； Among them, Z represents the response variable, x and y represent the independent variables, represents the autoregressive coefficient, represents the error term; finally, the analysis results are input into GIS to generate a spatiotemporal distribution map of soil quality, showing the changing trend and hot spots of soil quality; The prediction analysis module calculates the correlation coefficient between each feature and soil degradation, retaining the features with high correlation, which is expressed as: ,in represents the correlation coefficient, X and Y represent the values ​​of two features respectively. are their means respectively; The classification algorithm is then used to predict the risk level of soil degradation, expressed as: ； in, Indicates the sample The prediction results, represents the indicator function, Indicates The prediction results of a decision tree are represents the number of decision trees; And use deep neural network to process high-dimensional data, its forward propagation is expressed as: ； in, Indicates The input of the layer, denote weight and bias respectively, Represents the activation function; for classification problems, the logistic regression model outputs the probability of each category, which is used to determine the risk level of soil degradation, expressed as: ,in represents the Sigmaid function, represents the weight vector, represents the feature vector, and b represents the bias; by setting the threshold, the probability is converted into the risk level, and the prediction results with a probability higher than 0.7 are marked as high-risk areas; The application service layer also includes an early warning mechanism module and an interactive module, which provide soil degradation early warning based on real-time monitoring data and the results of analysis by the data processing layer, and provide an interactive interface for users.

2. A farmland quality monitoring system based on black soil protection as claimed in claim 1, characterized in that: The sensor module is responsible for collecting various physical and chemical index data of the soil in real time, including using a spectral sensor to measure the organic matter content of the soil and using an electrode sensor to measure the pH value and conductivity of the soil; The remote sensing monitoring module uses drones to obtain a wide range of soil surface information, including soil moisture and vegetation coverage. By using multispectral or hyperspectral cameras, the drones conduct repeated observations of the same area at different time points to monitor the dynamic changes in soil quality.

3. A farmland quality monitoring system based on black soil protection as claimed in claim 1, characterized in that: The ground penetrating radar module uses ground penetrating radar (GPR) technology to detect deep soil structure and moisture content information and assess soil erosion and groundwater level changes; The degree of soil erosion is assessed by using GPR equipment to scan the target area, obtain high-resolution images of the underground soil layer, pre-process the collected GPR data, including denoising and signal enhancement, and then use image processing technology to extract the boundaries and characteristics of the soil layer; By comparing GPR data at different time points, the changes in soil layers can be analyzed; Monitoring of groundwater level changes: Multiple groups of GPR measuring points are set up in the monitoring area, and GPR equipment is used to scan regularly to detect the reflection signal of the groundwater level, obtain the groundwater level change data, and then use time-space series analysis to compare the GPR data at different time points to determine the trend of groundwater level changes; The mobile laboratory module is a vehicle integrated with a variety of sensors and analytical instruments, and is used to collect and conduct preliminary analysis of soil samples directly in the field.

4. The cultivated land quality monitoring system based on black soil protection as claimed in claim 1, characterized in that: The data preprocessing module first performs standardization processing on the data collected by different sensors so that the data can be compared and analyzed on the same scale, which is expressed as: ； in, Represents the original data, represents the mean of the data, Indicates the standard deviation of the data; Then, different weights are assigned to each data source according to its reliability and accuracy. Remote sensing data with high spatial resolution has a higher weight, while data with low spatial resolution has a lower weight. The weight calculation is expressed as: ； in, Indicates The weight of the data source, Indicates The reliability score of each data source, Indicates the total number of data sources. Indicates the serial number or identifier of the data source, used to reference different data sources. Indicates reliability scores of the data sources; Then the weighted average method is used to fuse the multi-source data, which is expressed as: , where Y represents the fused data, Indicates Data from the data source, Indicates The weight of each data source; Finally, the fused data is input into the Geographic Information System (GIS) to generate a cultivated land quality monitoring map.

5. The cultivated land quality monitoring system based on black soil protection as claimed in claim 1, characterized in that: The soil quality assessment module is trained based on historical data and verified by actual observation data to construct a soil quality assessment model, which is expressed as: , where y represents the response variable, represents the weight vector, Represents the feature vector, b represents the bias, the historical data is divided into training set and test set, K-fold cross validation is used to evaluate the stability and generalization ability of the model, and the trained model is verified using the latest actual observation data.

6. A farmland quality monitoring system based on black soil protection as claimed in claim 1, characterized in that: The early warning mechanism module is based on real-time monitoring data and historical trend analysis, that is, the time series data of soil quality analyzed by the spatiotemporal data analysis module and the risk level of soil degradation predicted in the prediction analysis module. According to the degree of soil degradation, the early warning level is divided into 0-0.3 for low risk, 0.4-0.6 for medium risk, and 0.7-1 for high risk. The real-time data is continuously monitored and substituted into the model for prediction, and the corresponding level of early warning is triggered according to the set threshold.

Images