Regional soil moisture content detection method and device and electronic equipment

Abstract

The embodiment of the invention relates to the technical field of soil moisture content detection, and provides a regional soil moisture content detection method and device and electronic equipment, and the method comprises the steps: obtaining a current remote sensing satellite image and soil moisture content sensor data in a target region; inputting the current remote sensing satellite image into the latest target area soil moisture content inversion model to invert soil moisture content data of each pixel point in the target area; and correcting the soil moisture content data of each pixel point based on the data of the soil moisture content sensor to obtain a soil moisture content detection result of the target area. Therefore, the ground soil moisture content sensor, the remote sensing image receiver and various algorithms are carried, the advantages of the two measurement methods are combined, the measurement result is more accurate, the influence of various factors such as landforms, soil properties, vegetation coverage and vegetation types in different areas on the result is avoided, and the representativeness and precision of the detection result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of soil moisture detection, and in particular to a method, device and electronic equipment for detecting regional moisture content. Background Art

[0002] Soil moisture refers to the amount of water in the crop soil. Soil moisture refers to the humidity of the soil. Soil moisture refers to the moisture of the soil. Soil moisture refers to the amount of water in the soil. Soil moisture is the degree of dryness or wetness of the soil, that is, the actual water content of the soil, which can be expressed as the percentage of soil moisture to dry soil weight: soil moisture = water weight / dry soil weight × 100%. It can also be expressed as a relative moisture content such as the percentage of soil moisture equivalent to field water holding capacity, or the percentage relative to saturated water content. Soil moisture affects field climate, soil aeration and nutrient decomposition, and is one of the important conditions for soil microbial activity and crop growth and development. Soil moisture is affected by atmospheric, soil quality, vegetation and other conditions. Based on the relative humidity of the soil, we can know the degree of soil moisture and how much water can be retained, which has reference value for irrigation.

[0003] At present, the main methods for measuring soil moisture include soil sample analysis, soil moisture sensor measurement, and satellite remote sensing technology. Soil sample analysis is a traditional method for measuring soil moisture. It mainly collects soil samples, performs drying and weighing in the laboratory, and calculates the actual water content of the soil. The specific operation process includes: using a soil drill to drill a certain amount of soil samples at a selected location and depth. To prevent evaporation, the soil samples should be immediately put into an aluminum box and marked. Take them back to the laboratory for weighing, drying, re-weighing, analysis and calculation to obtain the soil moisture content. However, the disadvantages of soil sample analysis are as follows: the workload is large and time-consuming. Soil sample analysis requires collecting soil samples and performing a series of operations such as drying and weighing in the laboratory. The process is cumbersome and time-consuming. This method is particularly inconvenient for situations where frequent monitoring or large-area monitoring is required. It is difficult to obtain on-site data. Since soil samples need to be taken back to the laboratory for analysis, data cannot be obtained immediately on site. This may cause certain delays for agricultural production that needs to grasp the changes in soil moisture in real time. Limited representativeness: soil sample analysis can usually only be performed on specific sampling points, which makes it difficult to fully reflect the soil moisture conditions of the entire field. In addition, the selection of sampling points and sampling methods may also affect the accuracy of the results. Human errors: human factors may cause errors during sampling, transportation, processing and analysis. For example, unclean sampling tools, uneven sample mixing, inaccurate drying temperature control, etc. may affect the accuracy of the final results.

[0004] Soil moisture sensor measurement uses electronic equipment, such as soil moisture thermometers and wireless moisture monitoring systems, to monitor soil moisture in real time. Sensors can measure parameters such as the dielectric constant, temperature, and conductivity of the soil, thereby indirectly calculating the moisture content of the soil. The disadvantages and problems of soil moisture sensor measurement include: high equipment cost, high-quality soil moisture sensors are usually expensive, and multiple soil moisture sensors need to be installed in a monitoring area to achieve full coverage, which is economically unaffordable for small-scale farmers or areas with limited economic conditions. Maintenance and management are complex, and sensors need to be regularly maintained and calibrated to ensure the accuracy of measurement results. However, this requires users to have certain professional knowledge and technical capabilities, which may be difficult for non-professionals. Affected by environmental factors, soil moisture sensors may not work properly or produce errors under harsh environmental conditions (such as extreme temperature, humidity, electromagnetic interference, etc.). In addition, factors such as soil texture and salt content may also affect the measurement accuracy of the sensor. Limited representativeness, sensor measurements can only be carried out at the equipment installation point, and it is difficult to fully reflect the soil moisture conditions of the entire field. In addition, the selection and method of deployment points may also affect the accuracy of the results. Data interpretation requires professional knowledge. The data provided by the sensor usually needs to be comprehensively analyzed in combination with multiple factors such as weather and crop growth conditions to obtain effective irrigation recommendations. This may be difficult for ordinary farmers.

[0005] Satellite remote sensing technology uses sensors carried by satellites, drones and other high-altitude platforms to conduct long-distance, non-contact observations of farmland. By capturing characteristic information such as the spectrum and texture of the farmland surface, it can monitor and evaluate soil moisture. This method has the advantages of a wide monitoring range, fast data acquisition, and strong analysis and processing capabilities. Remote sensing inversion of soil moisture is mainly achieved from spectral characteristic analysis, vegetation index analysis, and microwave remote sensing. Spectral reflectance characteristics: The moisture content in the soil will affect its spectral reflectance characteristics, especially the near-infrared band and short-wave infrared band are more sensitive to changes in soil moisture. By measuring the reflectivity of the soil in different bands, the moisture content in the soil can be indirectly inferred. Vegetation index: The growth state of vegetation is closely related to the soil moisture status. When the soil moisture is sufficient, the vegetation grows lushly and the vegetation index is high; otherwise, it is low. By analyzing the changing trend of the vegetation index, the soil moisture can be indirectly evaluated. Microwave remote sensing: The change in the soil dielectric constant is used to monitor soil moisture. The moisture in the soil has a strong absorption and scattering effect on microwave signals, so the soil moisture content can be inferred by measuring the propagation characteristics of microwave signals in the soil. However, remote sensing technology has disadvantages and problems such as limited data resolution and accuracy. Although remote sensing technology can cover a large area, its data resolution and accuracy may be limited by multiple factors such as satellite orbits and sensor performance. For small-area or refined soil moisture monitoring, it may not meet the needs. Affected by weather conditions, weather conditions such as cloud cover and rainfall may cause remote sensing data to be unable to be obtained or the quality to decline, thereby affecting the accuracy of monitoring results. Complex data processing: Remote sensing data is large in volume and diverse in type, requiring highly specialized skills and knowledge to effectively process and analyze. This may be a certain threshold for ordinary users. Summary of the invention

[0006] The present invention provides a regional soil moisture detection method, device and electronic equipment to solve the defects of the prior art, such as high cost, low detection accuracy, great environmental influence and low representativeness of detection results, and realize low-cost, high-precision, low-impact and highly representative soil moisture measurement.

[0007] The present invention provides a method for detecting regional soil moisture, comprising: Obtain current remote sensing satellite images and soil moisture sensor data in the target area; Inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area; Correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain a soil moisture detection result of the target area; The target area soil moisture inversion model is trained based on the following steps: Acquire soil moisture measurement data in the sample soil and the first remote sensing satellite image corresponding to the measurement time period; Performing band calculation on the first remote sensing satellite image to obtain a plurality of band calculation results reflected in the first remote sensing satellite image, wherein each band calculation result includes a plurality of parameter indexes; Randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the calculation results of the multiple bands, respectively, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; Inputting the inversion data set and the plurality of band operation results into a full subset screening algorithm to screen out an optimal variable combination; Inputting the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; The test data set is input into the inversion model established by each algorithm respectively, and the inversion models established by the multiple algorithms are tested by calculating the determination coefficient and the root mean square error to determine the best soil moisture inversion model for the target area.

[0008] In a possible implementation, the method further includes: The ratio of the soil moisture data of each pixel point to the corresponding soil moisture sensor data is used as a correction coefficient; The ratio of the soil moisture data of each pixel point to the correction coefficient is used as the soil moisture detection result of the target area.

[0009] In a possible implementation, the method further includes: Determine the number and location of soil sampling plots within the target area based on preset rules to obtain multiple soil sampling plots; The soil moisture measurement data are respectively obtained in the plurality of soil sampling plots through soil moisture sensors.

[0010] In a possible implementation, the method further includes: Obtaining the blue light band reflectivity, green light band reflectivity, red light band reflectivity and near infrared band reflectivity displayed in the first remote sensing satellite image; Based on the blue light band reflectivity, green light band reflectivity, red light band reflectivity and near infrared band reflectivity and the soil line coefficient of the sample soil, multiple band operation results reflected in the first remote sensing satellite image are calculated.

[0011] In a possible implementation, the method further includes: Based on a preset ratio and a preset number of extractions, an inversion data set and a test data set are randomly extracted from the soil moisture measurement data and the calculation results of the multiple bands to obtain an inversion data set and a test data set extracted each time.

[0012] In a possible implementation, the method further includes: The model is trained based on the inversion dataset and test dataset extracted each time; The goodness of fit of the model trained based on the inversion data set and the test data set extracted each time is evaluated, and the best variable combination with the best goodness of fit is screened out.

[0013] In a possible implementation, the method further includes: The optimal variable combination and the inversion data set extracted each time are input into a plurality of preset algorithm models to obtain a plurality of inversion models established by each algorithm.

[0014] In a possible implementation, the method further includes: Inputting the test data set into the multiple inversion models established by each algorithm respectively, and verifying the inversion models based on goodness of fit and root mean square error; The inversion model with the highest goodness of fit and the smallest root mean square error is selected as the best target area soil moisture inversion model; Alternatively, the relative error coefficient of each inversion model is calculated based on the goodness of fit and the root mean square error, and the inversion model with the highest relative error coefficient is selected as the optimal target area soil moisture inversion model.

[0015] The present invention also provides a regional soil moisture detection device, comprising the following modules: The acquisition module is used to obtain the current remote sensing satellite image and soil moisture sensor data in the target area; A detection module, used for inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area; A correction module, used for correcting the soil moisture data of each pixel point based on the soil moisture sensor data to obtain a soil moisture detection result of the target area; A model training module is used to obtain soil moisture measurement data in a sample soil and a first remote sensing satellite image within a corresponding measurement time period; perform band calculation on the first remote sensing satellite image to obtain multiple band calculation results reflected in the first remote sensing satellite image, wherein each band calculation result includes multiple parameter indexes; randomly extract an inversion data set and a test data set from the soil moisture measurement data and the multiple band calculation results, respectively, wherein the inversion data set is used to train a soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; input the inversion data set and the multiple band calculation results into a full subset screening algorithm to screen out an optimal variable combination; input the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; input the test data set into the inversion models established by each algorithm, respectively, and perform model testing on the inversion models established by the plurality of algorithms by calculating the determination coefficient and the root mean square error to determine the optimal target area soil moisture inversion model.

[0016] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the regional soil moisture detection method as described in any one of the above is implemented.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for detecting regional soil moisture as described in any one of the above is implemented.

[0018] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for detecting regional moisture content as described above is implemented.

[0019] The regional soil moisture detection method, device and electronic equipment provided by the present invention obtain the current remote sensing satellite image and soil moisture sensor data in the target area; input the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel point in the target area; corrects the soil moisture data of each pixel point based on the soil moisture sensor data to obtain the soil moisture detection result of the target area; wherein the soil moisture inversion model of the target area is trained based on the following steps: obtaining soil moisture measurement data in sample soil and a first remote sensing satellite image in a corresponding measurement time period; performing band calculation on the first remote sensing satellite image to obtain multiple band calculation results reflected in the first remote sensing satellite image, wherein , each band calculation result includes multiple parameter indexes; randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the multiple band calculation results, respectively, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; inputting the inversion data set and the multiple band operation results into the full subset screening algorithm to screen out the best variable combination; inputting the best variable combination and the inversion data set into a plurality of preset algorithm models to obtain the inversion model established by each algorithm; inputting the test data set into the inversion model established by each algorithm respectively, and performing model testing on the inversion models established by the plurality of algorithms by calculating the determination coefficient and the root mean square error to determine the best target area soil moisture inversion model. Compared with the existing technology with its high cost, low detection accuracy, great environmental influence and low representativeness of detection results, this solution is equipped with ground moisture sensors, remote sensing image receivers and multiple algorithms, so as to combine the advantages of the two measurement methods and make up for their disadvantages. Each measurement area can be conveniently re-modeled to make the measurement results more accurate and avoid the influence of various factors such as topography, soil properties, vegetation coverage and vegetation types in different regions on the results. Remote sensing satellites are used to quantitatively invert the moisture content of the entire area, which improves the representativeness and accuracy of the results. Moisture content sensors are used to make up for the problem of dynamic monitoring of soil moisture content during the time interval between the acquisition of two satellite images, thus realizing low-cost, high-precision, low-impact and highly representative soil moisture measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1It is a schematic flow chart of the regional soil moisture detection method provided by the present invention.

[0022] Figure 2 It is a flow chart of the target area soil moisture inversion model training method provided by the present invention.

[0023] Figure 3 It is a schematic diagram of the instrument layout principle and use flow provided by the present invention.

[0024] Figure 4 It is a schematic diagram of the regional soil moisture detection output result provided by the present invention.

[0025] Figure 5 It is a structural schematic diagram of the regional soil moisture detection device provided by the present invention.

[0026] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] To facilitate understanding of the embodiments of the present invention, specific embodiments will be further explained below in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the embodiments of the present invention.

[0029] Figure 1 is a flow chart of the regional soil moisture detection method provided by the present invention, such as Figure 1 As shown, the method specifically includes: S11. Acquire current remote sensing satellite images and soil moisture sensor data in the target area.

[0030] In the embodiment of the present invention, firstly, a target area soil moisture inversion model is trained for the target area. The specific training method is as follows: Figure 2 The corresponding embodiments are described in detail. After the soil moisture inversion model training of the target area is completed, the current remote sensing satellite image and soil moisture sensor data in the target area to be detected are obtained.

[0031] S12, inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area.

[0032] During the model training process, the system automatically records the best inversion model at each sampling time (the latest soil moisture inversion model of the target area). After recording the best inversion model, the next measured data is input into the latest soil moisture inversion model of the target area.

[0033] The current remote sensing satellite image is input into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area can invert the soil moisture data of each pixel in the target area.

[0034] S13. Correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain a soil moisture detection result of the target area.

[0035] The ratio of the soil moisture data of each pixel point to the corresponding soil moisture sensor data is used as a correction coefficient; the ratio of the soil moisture data of each pixel point to the correction coefficient is used as the soil moisture detection result of the target area.

[0036] The calibration steps are as follows: 1) The instrument records the soil moisture values ​​at different depths of the deployed points every 30 minutes.

[0037] 2) Use the latest remote sensing images combined with the best inversion model to invert the soil moisture results at the time when the image was acquired.

[0038] 3) Obtain the latest remote sensing image time, and use the soil moisture value at the closest recording time as the correction value. If the time is the same, take the average value of the data at two times as the correction value.

[0039] 4) Obtain the inversion result of the soil moisture at the location of the instrument and obtain the correction factor, which is = the inversion result of the soil moisture at the location of the instrument / correction value.

[0040] 5) Divide the data of each pixel point of the inversion results by the correction coefficient to obtain the corrected true value of all pixels, thereby obtaining the corrected soil moisture sensor data to correct the soil moisture result value of the measurement area. The final regional soil moisture detection output result is as follows: Figure 4 shown.

[0041] Optionally, if the current remote sensing satellite image in the target area is not acquired within a preset time period (eg, 30 minutes), the soil moisture data obtained last time is corrected based on the soil moisture sensor data.

[0042] Specifically, in the time interval between the acquisition of two images, the instrument can record the soil moisture values ​​at different depths of the deployed points every 30 minutes to correct the soil moisture result values ​​of the entire area at that time, thereby obtaining regional soil moisture data every 30 minutes.

[0043] The calibration steps are as follows: 1) The instrument records the soil moisture values ​​at different depths of the deployed points every 30 minutes.

[0044] 2) The soil moisture value obtained each time is used as the correction value.

[0045] 3) Obtain the soil moisture inversion result obtained from the latest remote sensing image at the location of the instrument, and obtain the correction coefficient, which is the soil moisture inversion result obtained from the latest remote sensing image at the location of the instrument / correction value.

[0046] 4) Dividing the data of each pixel point of the soil moisture inversion results obtained from all the latest remote sensing images by the correction coefficient can obtain the corrected true value of all pixels, thereby obtaining the soil moisture result value of the measurement area during that period.

[0047] The embodiment of the present invention combines ground soil moisture sensors with satellite remote sensing technology to determine the optimal regional soil moisture inversion model. The characteristics of ground sensors, such as strong real-time performance, good adaptability, and good data continuity, are used to effectively make up for the data gap between the two images provided by satellite remote sensing inversion, and the resolution and accuracy of the result data are effectively adjusted and improved. Satellite remote sensing technology is used to effectively improve the representativeness of the results, reduce the impact of the equipment deployment environment on the results, reduce the amount of equipment deployment, thereby reducing equipment costs, and reduce the impact of deploying multiple devices on cultivated land.

[0048] The regional soil moisture detection method provided by the present invention obtains the current remote sensing satellite image and soil moisture sensor data in the target area; inputs the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel point in the target area; corrects the soil moisture data of each pixel point based on the soil moisture sensor data to obtain the soil moisture detection result of the target area. Compared with the existing technologies with the defects of high cost, low detection accuracy, great environmental influence and low representativeness of detection results, this method is equipped with ground moisture sensors, remote sensing image receivers and multiple algorithms, so as to combine the advantages of the two measurement methods and make up for the disadvantages. Each measurement area can be conveniently re-modeled to make the measurement results more accurate and avoid the influence of various factors such as topography, soil properties, vegetation coverage and vegetation types in different regions on the results; remote sensing satellites are used to quantitatively invert the moisture content of the entire area, thereby improving the representativeness and accuracy of the results; moisture sensors are used to make up for the problem of dynamic monitoring of soil moisture content during the time interval between the acquisition of two satellite images, thereby realizing low-cost, high-precision, low-impact and highly representative soil moisture measurement.

[0049] Figure 2is a flow chart of the target area soil moisture inversion model training method provided by the present invention, such as Figure 2 As shown, the method specifically includes: S21. Obtain soil moisture measurement data in the sample soil and a first remote sensing satellite image corresponding to a measurement time period.

[0050] Embodiments of the present invention are combined Figure 3 The schematic diagram of the instrument layout principle and the use process shown in the figure is used to illustrate. The system of the embodiment of the present invention is installed in a soil moisture measurement instrument (electronic device), and the hardware of the instrument also includes a soil moisture sensor. First, the number and location of soil sampling plots are determined in the target area based on preset rules to obtain multiple soil sampling plots; soil moisture measurement data is obtained through soil moisture sensors in the multiple soil sampling plots.

[0051] Specifically, determine the number of sample plots in the sample measurement area. The number of sample plots is required to be at least one for every 60 mu. For example, if the area to be measured is 300 mu, 300 mu / 60 mu = 5 sample plots need to be selected in this area. If the area to be measured is 320 mu, 6 sample plots need to be selected in this area.

[0052] Furthermore, the location of the sample plot is determined in the measurement area. The sample plot is a 3m*3m square. The sample plot location is required to be more than 30m away from the boundary of the measurement area. The sample plot should be evenly and randomly distributed in the measurement area. For special terrain or land properties in the measurement area, additional sample plots can be added as needed. If there are areas with significantly higher or lower terrain in the measurement area, or areas with greater salinity and alkalinity in the measurement area, additional sample plots can be added in this special area.

[0053] Furthermore, 5 sampling points can be randomly selected in each sample plot, and the soil moisture sensor of the instrument can be used to measure the soil moisture in the sample plot. Before the measurement, create sample plots 1-sample plot N in the instrument, open the corresponding sample plot number when measuring the sample plot, and the measurement method is to insert the ground moisture sensor part of the instrument into the soil at the sampling point, and the insertion depth is the standard scale marked on the instrument, and 80cm, and the insertion time is more than 30s, so that the soil moisture data of the insertion point 20cm, 40cm, 60cm, and 80cm, as well as the sampling point coordinates, can be obtained, and the instrument will automatically record them as sample plot N point 1-N. For example, when measuring the first point of sample plot 1, create numbered sample plot 1, and after the first measurement is completed, the instrument automatically records the data as sample plot 1 point 1, and when measuring the second point of sample plot 1, after the second measurement is completed, the instrument automatically records the data as sample plot 1 point 2, and so on.

[0054] Furthermore, the measurement of samples in the same measurement area should be completed within 2 days as much as possible. After all sample plots are measured, the instrument will screen all data using an algorithm to filter out abnormal data. Users can confirm or delete the abnormal data according to actual conditions.

[0055] S22: Perform band calculation on the first remote sensing satellite image to obtain multiple band calculation results reflected in the first remote sensing satellite image.

[0056] Obtain the blue light band reflectance, green light band reflectance, red light band reflectance and near infrared band reflectance displayed in the first remote sensing satellite image; calculate the multiple band operation results reflected in the first remote sensing satellite image based on the blue light band reflectance, green light band reflectance, red light band reflectance and near infrared band reflectance and the soil line coefficient of the sample soil. Each band calculation result includes multiple parameter indexes.

[0057] Specifically, after the user confirms the data, the instrument will obtain remote sensing images within the measurement period (remote sensing images with at least four bands: red, green, blue, and near-red, with a spatial resolution of ≤10m). Perform the following band operations on the acquired remote sensing images: In the following formula and They correspond to the reflectance of the blue, green, red and near-infrared bands of the remote sensing image. These parameters are inherent in the remote sensing image and can be read directly. a and b represent the soil line coefficients.

[0058] Normalized Difference Vegetation Index (NDVI)

[0059] Enhanced Vegetation Index (EVI)

[0060] Ratio Vegetation Index (RVI)

[0061] Difference Vegetation Index (DVI)

[0062] Soil Adjusted Vegetation Index (SAVI)

[0063] Modified Soil Adjusted Vegetation Index (MSAVI)

[0064] Optimized Soil Adjusted Vegetation Index (OSAVI)

[0065] Improved Simple Ratio Vegetation Index (MSR)

[0066] Structure-Insensitive Pigment Index (SIPI)

[0067] Modified Nonlinear Vegetation Index (MNVI)

[0068] Triangular Vegetation Index (TVI)

[0069] Green Leaf Index (GLI)

[0070] Modified Chlorophyll Absorption Reflectance Index 2 (MCARI2)

[0071] Salinity Index (SI-T)

[0072] Salinity Index (S3)

[0073] Salinity Index (S5)

[0074] Salinity Index (S6)

[0075] Soil Adjustment Index (TSAVI)

[0076] Building Index (BI)

[0077] Bare Soil Index (SI2)

[0078] Vertical Drought Index (PDI)

[0079] Soil Adjusted Vegetation Index (SAVI2)

[0080] S23. Randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the calculation results of the multiple bands.

[0081] S24, inputting the inversion data set and the plurality of band operation results into a full subset screening algorithm to screen and obtain an optimal variable combination.

[0082] Based on a preset ratio and a preset number of extractions, an inversion data set and a test data set are randomly extracted from the soil moisture measurement data and the calculation results of multiple bands, respectively, to obtain an inversion data set and a test data set extracted each time. The inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model. A model is trained based on the inversion data set and the test data set extracted each time; a goodness of fit evaluation is performed on the model trained based on the inversion data set and the test data set extracted each time, and the best variable combination with the best goodness of fit is screened out.

[0083] Specifically, after the operation of the acquired image bands is completed, 80% of the sampled data can be randomly extracted as inversion data, and the remaining 20% ​​of the data is test data. Random extraction is performed 100 times. The method of extracting data once here is: randomly select 80% of all sampled data as inversion data, and the remaining 20% ​​after this extraction is test data. Use the inversion data to input the above-mentioned multiple band operation formulas to calculate the corresponding results, and input the results into the full subset screening model. After the data is extracted once and input into the full subset screening model, many model combinations will appear. Use test data to test the determination coefficient of all model combinations. The combination with the highest determination coefficient is the best model selection after the data is extracted this time. The combination and the determination coefficient need to be recorded.

[0084] The above is the result of one optimal variable selection. It is necessary to conduct 100 optimal variable selection results, that is, randomly draw 100 times to obtain 100 optimal variables and their corresponding determination coefficients. The optimal variable selection result with the highest determination coefficient is the best variable combination screened out.

[0085] The core of the All-Subsets Selection algorithm is to traverse all possible combinations of feature variables, evaluate the model built for each combination, and finally select the optimal feature subset. This method aims to find the best performing model under the current data set by comprehensively considering all possible feature combinations. This comprehensiveness makes the All-Subsets Selection algorithm highly flexible and accurate in the feature selection process. The steps are as follows: 1) Generate all subsets: Assuming there are p feature variables, the full subset screening algorithm will generate all possible feature subsets. This includes univariate subsets containing only one feature, bivariate subsets containing two features, and even the full set containing all p features. There are a total of 2^p-1 (including the full set, excluding the empty set).

[0086] 2) Model fitting: For each feature subset generated, the same modeling method (such as linear regression, logistic regression, decision tree, etc.) is used to fit the model. The methods used in the embodiment of the present invention are linear regression and random forest. This step needs to be performed separately for each subset, so the amount of calculation will increase rapidly as the number of features increases.

[0087] 3) Performance evaluation: The coefficient of determination R² is used to evaluate the performance of each model. These criteria are used to measure the goodness of fit, complexity, and predictive ability of the model. Based on the evaluation results, different feature subsets are compared and ranked to identify the best performing feature subset. The coefficient of determination (R²), also known as goodness of fit, is an indicator of the model's ability to predict the target variable. It indicates the degree to which the independent variables in the model explain the dependent variable, that is, the percentage of the dependent variable variation that the independent variables in the model can explain. The calculation formula for R² is: R² = 1 - (SSE / SST).

[0088] Among them, SSE is the sum of squared errors, which represents the sum of squares of the differences between the model prediction value and the actual value; SST is the total sum of squares, which represents the sum of squares of the differences between the actual value and the mean. The value of R² is between 0 and 1. The closer the value is to 1, the better the goodness of fit of the model is, that is, the independent variables in the model can better explain the variation of the dependent variable.

[0089] 4) Select the optimal subset: The instrument selects the optimal feature subset based on the results of the performance evaluation. This subset is usually the best performing subset under the current evaluation criteria, that is, it can best balance the goodness of fit and complexity of the model.

[0090] S25, inputting the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm.

[0091] The optimal variable combination and the inversion data set extracted each time are input into the preset multiple algorithm models to obtain multiple inversion models established by each algorithm.

[0092] Specifically, the best combination of variables selected and 100 randomly selected inversion data are input into the instrument's partial least squares regression algorithm (common algorithm), back propagation neural network model (common algorithm), genetic algorithm optimized back propagation neural network model (common algorithm), and random forest algorithm (common algorithm) to obtain 100 models established by each algorithm. Here, 100 data can be re-extracted from all the data according to the preset ratio, for example, 80% is inversion data and 20% is test data.

[0093] Partial Least Squares Regression (PLSR) is a multivariate statistical data analysis method that can establish a linear relationship model between independent variables and dependent variables. PLSR extracts the most explanatory comprehensive variables (i.e. components) for the dependent variable by projecting the original data into a new space, and then performs regression modeling. This method is particularly suitable for processing data sets with many variables and multicollinearity. The advantage is that it can effectively deal with multicollinearity problems, extract important information, and simplify the model. The disadvantage is that the explanatory power of the model may be weak because the components may be difficult to directly interpret.

[0094] The Backpropagation Neural Network (BPNN) model is a multi-layer feedforward neural network that is trained using the backpropagation algorithm. The algorithm uses the chain rule to calculate the gradient of the loss function with respect to each weight and updates the weights to minimize the loss function. BPNN is able to learn complex nonlinear relationships between input and output. The advantage is that it has powerful nonlinear modeling capabilities and can handle complex problems. The disadvantage is that the training process may be slow and prone to overfitting.

[0095] Genetic Algorithm for Optimizing BPNN is an optimization algorithm that simulates the biological evolution process. It searches for the optimal solution through operations such as selection, crossover and mutation. Applying genetic algorithm to the optimization of BPNN can improve the weight initialization of BPNN, improve training efficiency and model performance. The advantage is that it can combine the global search ability of genetic algorithm and the local search ability of BPNN to improve model performance. The disadvantage is that the computational complexity is high and appropriate genetic algorithm parameter settings are required.

[0096] The Random Forest algorithm is an ensemble learning method that builds multiple decision trees and averages or majority votes their predictions. When building each tree, the random forest increases the diversity of the tree by introducing randomness in samples and features. The advantages are high prediction accuracy, the ability to handle high-dimensional data, and good robustness to outliers and noise. The disadvantages are that the model is less interpretable and the training process may be relatively slow.

[0097] S26, inputting the test data set into the inversion model established by each algorithm respectively, and performing model testing on the inversion models established by the multiple algorithms by calculating the determination coefficient and the root mean square error, so as to determine the best soil moisture inversion model for the target area.

[0098] The test data sets were input into multiple inversion models established by each algorithm, and the inversion models were verified based on goodness of fit and root mean square error; the inversion model with the highest goodness of fit and the smallest root mean square error was selected as the optimal soil moisture inversion model for the target area.

[0099] Specifically, the remaining 20% ​​of the test data from 100 draws were input into the 100 models established by each algorithm, and the inversion model was tested by calculating the coefficient of determination (R²) and the root mean square error (MSE). If the best inversion model selected by the two parameters is the same, then the model is the best inversion model for the measurement area. If the best inversion model selected by the two parameters is different, the difference between the two coefficients is calculated and the model with the larger difference is the best inversion model.

[0100] The relative error coefficient (REC) of each model was calculated using the coefficient of determination (R²) and root mean square error (MSE) of each model, and the model with the highest relative error coefficient (REC) was the best inversion model.

[0101] The coefficient of determination, also known as goodness of fit, is an indicator of the model's ability to predict the target variable. It indicates the degree to which the independent variables in the model explain the dependent variable, that is, the percentage of the dependent variable variation that the independent variables in the model can explain. The calculation formula for R² is: R² = 1 - (SSE / SST).

[0102] Among them, SSE is the sum of squared errors, which represents the sum of squares of the differences between the model prediction value and the actual value; SST is the total sum of squares, which represents the sum of squares of the differences between the actual value and the mean. The value of R² is between 0 and 1. The closer the value is to 1, the better the goodness of fit of the model is, that is, the independent variables in the model can better explain the variation of the dependent variable.

[0103] The root mean square error (MSE) is an indicator to measure the prediction accuracy of the model. It represents the square root of the average of the sum of the squares of the differences between the model prediction value and the actual value. The calculation formula of MSE is: .

[0104] Among them, SSE is the residual sum of squares, and n is the number of samples. The smaller the value of MSE, the higher the prediction accuracy of the model.

[0105] Alternatively, the relative error coefficient of each inversion model is calculated based on the goodness of fit and the root mean square error, and the inversion model with the highest relative error coefficient is selected as the optimal soil moisture inversion model for the target area.

[0106] The relative error coefficient (REC) is the coefficient obtained according to its reference significance after normalizing the determination coefficient (R²) and the root mean square error (RMSE), which is used to judge the model fitting accuracy.

[0107] The calculation formula of REC is: .

[0108] in, For a model result, For all models The minimum value among the results, For all models The maximum value among the results, For a model result, For all models The minimum value among the results, For all models The maximum value among the results.

[0109] Furthermore, the best inversion model at each sampling time is automatically recorded. After recording the best inversion model, the data of the next measurement is input into the latest best inversion model. Multiple tasks can be performed in the same measurement area at different times, and the instrument automatically records multiple best inversion models according to time.

[0110] The instrument is deployed in the measurement area, and the instrument obtains the latest remote sensing satellite image every time. The image is input into the best inversion model with the closest time stored in the instrument, and the soil moisture value of each pixel point in the entire measurement area is inverted, and the soil moisture result of the measurement area is corrected in combination with the soil moisture sensor data at the time of image shooting.

[0111] The calibration steps are as follows: 1) The instrument records the soil moisture values ​​at different depths of the deployed points every 30 minutes.

[0112] 2) Use the latest remote sensing images combined with the best inversion model to invert the soil moisture results at the time when the image was acquired.

[0113] 3) Obtain the latest remote sensing image time, and use the soil moisture value at the closest recording time as the correction value. If the time is the same, take the average value of the data at two times as the correction value.

[0114] 4) Obtain the inversion result of the soil moisture at the location of the instrument and obtain the correction factor, which is = the inversion result of the soil moisture at the location of the instrument / correction value.

[0115] 5) Dividing the data of each pixel point of the entire inversion result by the correction coefficient can obtain the corrected true value of all pixel points, thereby obtaining the corrected soil moisture sensor data to correct the soil moisture result value of the measurement area.

[0116] Furthermore, in the time interval between the two images, the instrument will record the soil moisture values ​​at different depths of the deployed points every 30 minutes to correct the soil moisture result values ​​of the entire area at that time, thereby obtaining regional soil moisture data every 30 minutes.

[0117] The calibration steps are as follows: 1) The instrument records the soil moisture values ​​at different depths of the deployed points every 30 minutes.

[0118] 2) The soil moisture value obtained each time is used as the correction value.

[0119] 3) Obtain the soil moisture inversion result obtained from the latest remote sensing image at the location of the instrument, and obtain the correction coefficient, which is the soil moisture inversion result obtained from the latest remote sensing image at the location of the instrument / correction value.

[0120] 4) Dividing the data of each pixel point of the soil moisture inversion results obtained from all the latest remote sensing images by the correction coefficient can obtain the corrected true value of all pixels, thereby obtaining the soil moisture result value of the measurement area during that period.

[0121] The target area soil moisture inversion model training method provided by the embodiment of the present invention is through obtaining soil moisture measurement data in sample soil and a first remote sensing satellite image in the corresponding measurement time period; performing band calculation on the first remote sensing satellite image to obtain multiple band operation results reflected in the first remote sensing satellite image, wherein each band calculation result includes multiple parameter indexes; randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the multiple band calculation results, respectively, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; inputting the inversion data set and the multiple band operation results into a full subset screening algorithm to screen out the best variable combination; inputting the best variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; inputting the test data set into the inversion model established by each algorithm, and performing model testing on the inversion models established by the plurality of algorithms by calculating the determination coefficient and the root mean square error, so as to determine the best target area soil moisture inversion model. This method is equipped with ground moisture sensors, remote sensing image receivers and multiple algorithms at the same time, so that the advantages of the two measurement methods are combined and the disadvantages are supplemented. For each measurement area, the process can be re-modeled conveniently, which can make the instrument measurement results more accurate and avoid the influence of various factors such as topography, soil properties, vegetation coverage, and vegetation types in different regions on the results. Different regions, different soil properties, different types of vegetation coverage and canopy density, and different topography will lead to different sensitive band combinations used in soil moisture inversion. Therefore, for the measured area, point measurement is first carried out, and then characteristic variables are screened, so that the inversion input parameters most suitable for the area can be obtained according to local conditions. Since the actual monitoring after the instrument is deployed will be somewhat different from the point sampling, such as date, temperature, air pressure, humidity, etc., which may affect the inversion accuracy, the addition of ground sensors can correct the results of remote sensing inversion and make them closer to the current actual values. The use of remote sensing images fundamentally solves the limitations of the monitoring range of ground sensors, and the use of ground sensors fundamentally solves the problem of no data between two scene images. The actual value of each pixel of the remote sensing image can be accurately and quantitatively inverted.

[0122] The regional soil moisture detection device provided by the present invention is described below. The regional soil moisture detection device described below and the regional soil moisture detection method described above can be referred to each other.

[0123] Figure 5 : is a schematic diagram of the structure of the regional soil moisture detection device provided by the present invention, which specifically includes: The acquisition module 501 is used to acquire the current remote sensing satellite image and soil moisture sensor data in the target area. For detailed description, please refer to the relevant description corresponding to the above method embodiment, which will not be repeated here.

[0124] The detection module 502 is used to input the current remote sensing satellite image into the latest target area soil moisture inversion model, so that the latest target area soil moisture inversion model inverts the soil moisture data of each pixel in the target area. For detailed description, please refer to the relevant description corresponding to the above method embodiment, which will not be repeated here.

[0125] The correction module 503 is used to correct the soil moisture data of each pixel based on the soil moisture sensor data to obtain the soil moisture detection result of the target area. For detailed description, please refer to the relevant description corresponding to the above method embodiment, which will not be repeated here.

[0126] The model training module 504 is used to obtain the soil moisture measurement data in the sample soil and the first remote sensing satellite image within the corresponding measurement time period; perform band calculation on the first remote sensing satellite image to obtain multiple band operation results reflected in the first remote sensing satellite image, wherein each band calculation result includes multiple parameter indexes; randomly extract an inversion data set and a test data set from the soil moisture measurement data and the multiple band calculation results, respectively, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; input the inversion data set and the multiple band operation results into the full subset screening algorithm to screen out the best variable combination; input the best variable combination and the inversion data set into a preset plurality of algorithm models to obtain the inversion model established by each algorithm; input the test data set into the inversion model established by each algorithm, and perform model testing on the inversion models established by the multiple algorithms by calculating the determination coefficient and the root mean square error to determine the best target area soil moisture inversion model. For detailed description, please refer to the relevant description corresponding to the above method embodiment, which will not be repeated here.

[0127] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor (processor) 810, a communication interface (Communications Interface) 820, a memory (memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call the logic instructions in the memory 830 to execute the regional soil moisture detection method, which includes: obtaining the current remote sensing satellite image and soil moisture sensor data in the target area; inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area; correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain the soil moisture detection result of the target area; wherein the soil moisture inversion model of the target area is trained based on the following steps: obtaining the soil moisture measurement data in the sample soil and the first remote sensing satellite image in the corresponding measurement time period; performing band calculation on the first remote sensing satellite image to obtain multiple bands reflected in the first remote sensing satellite image. The invention relates to a method for determining a soil moisture inversion model in a target area by using a plurality of inversion datasets and a plurality of band calculation results, wherein each band calculation result includes a plurality of parameter indexes; randomly extracting an inversion dataset and a test dataset from the soil moisture measurement data and the plurality of band calculation results, wherein the inversion dataset is used to train a soil moisture inversion model, and the test dataset is used to test the trained soil moisture inversion model; inputting the inversion dataset and the plurality of band calculation results into a full subset screening algorithm to screen out an optimal variable combination; inputting the optimal variable combination and the inversion dataset into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; inputting the test dataset into the inversion models established by each algorithm, and performing model tests on the inversion models established by the plurality of algorithms by calculating the determination coefficient and the root mean square error to determine an optimal soil moisture inversion model for the target area.

[0128] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0129] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the regional soil moisture detection method provided by the above-mentioned methods, which includes: obtaining the current remote sensing satellite image and soil moisture sensor data in the target area; inputting the current remote sensing satellite image into the latest target area soil moisture inversion model, so that the latest target area soil moisture inversion model inverts the soil moisture data of each pixel in the target area; correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain the soil moisture detection result of the target area; wherein the target area soil moisture inversion model is trained based on the following steps: obtaining the soil moisture measurement data in the sample soil and the first remote sensing satellite image in the corresponding measurement time period; and correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain the soil moisture detection result of the target area. A remote sensing satellite image is subjected to band calculation to obtain a plurality of band calculation results reflected in the first remote sensing satellite image, wherein each band calculation result includes a plurality of parameter indexes; an inversion data set and a test data set are randomly extracted from the soil moisture measurement data and the plurality of band calculation results, respectively, wherein the inversion data set is used to train a soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; the inversion data set and the plurality of band calculation results are input into a full subset screening algorithm to screen out an optimal variable combination; the optimal variable combination and the inversion data set are input into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; the test data set is respectively input into the inversion models established by each algorithm, and the inversion models established by the plurality of algorithms are tested by calculating the determination coefficient and the root mean square error to determine the optimal soil moisture inversion model of the target area.

[0130] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the regional soil moisture detection method provided by the above-mentioned methods, the method comprising: obtaining current remote sensing satellite images and soil moisture sensor data in the target area; inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel point in the target area; correcting the soil moisture data of each pixel point based on the soil moisture sensor data to obtain the soil moisture detection result of the target area; wherein the soil moisture inversion model of the target area is trained based on the following steps: obtaining soil moisture measurement data in the sample soil and a first remote sensing satellite image in the corresponding measurement time period; performing band calculation on the first remote sensing satellite image, The calculation results of multiple bands reflected in the first remote sensing satellite image are obtained, wherein each band calculation result includes multiple parameter indexes; an inversion data set and a test data set are randomly extracted from the soil moisture measurement data and the calculation results of the multiple bands, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; the inversion data set and the calculation results of the multiple bands are input into the full subset screening algorithm to screen out the best variable combination; the best variable combination and the inversion data set are input into the preset multiple algorithm models to obtain the inversion model established by each algorithm; the test data set is respectively input into the inversion model established by each algorithm, and the inversion models established by the multiple algorithms are tested by calculating the determination coefficient and the root mean square error to determine the best soil moisture inversion model of the target area.

[0131] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0132] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0133] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting regional soil moisture, characterized in that: include: Obtain current remote sensing satellite images and soil moisture sensor data in the target area; Inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area; Correcting the soil moisture data of each pixel based on the soil moisture sensor data to obtain a soil moisture detection result of the target area; The target area soil moisture inversion model is trained based on the following steps: Acquire soil moisture measurement data in the sample soil and the first remote sensing satellite image corresponding to the measurement time period; Performing band calculation on the first remote sensing satellite image to obtain a plurality of band calculation results reflected in the first remote sensing satellite image, wherein each band calculation result includes a plurality of parameter indexes; Randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the calculation results of the multiple bands, respectively, wherein the inversion data set is used to train the soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; Inputting the inversion data set and the plurality of band operation results into a full subset screening algorithm to screen out an optimal variable combination; Inputting the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; The test data set is input into the inversion model established by each algorithm respectively, and the inversion models established by the multiple algorithms are tested by calculating the determination coefficient and the root mean square error to determine the best soil moisture inversion model for the target area.

2. The method according to claim 1, characterized in that The step of correcting the soil moisture data based on the soil moisture sensor data to obtain a soil moisture detection result of the target area includes: The ratio of the soil moisture data of each pixel point to the corresponding soil moisture sensor data is used as a correction coefficient; The ratio of the soil moisture data of each pixel point to the correction coefficient is used as the soil moisture detection result of the target area; or, If the current remote sensing satellite image in the target area is not acquired within a preset time period, the soil moisture data obtained last time is corrected based on the soil moisture sensor data.

3. The method according to claim 1, characterized in that The step of obtaining the soil moisture measurement data in the sample soil comprises: Determine the number and location of soil sampling plots within the target area based on preset rules to obtain multiple soil sampling plots; The soil moisture measurement data are respectively obtained in the plurality of soil sampling plots through soil moisture sensors.

4. The method according to claim 1, characterized in that: The performing band calculation on the first remote sensing satellite image to obtain a plurality of band calculation results reflected in the first remote sensing satellite image includes: Obtaining the blue light band reflectivity, green light band reflectivity, red light band reflectivity and near infrared band reflectivity displayed in the first remote sensing satellite image; Based on the blue light band reflectivity, green light band reflectivity, red light band reflectivity and near infrared band reflectivity and the soil line coefficient of the sample soil, multiple band operation results reflected in the first remote sensing satellite image are calculated.

5. The method according to claim 3 or 4, characterized in that: The randomly extracting an inversion data set and a test data set from the soil moisture measurement data and the calculation results of the multiple bands respectively comprises: Based on a preset ratio and a preset number of extractions, an inversion data set and a test data set are randomly extracted from the soil moisture measurement data and the calculation results of the multiple bands to obtain an inversion data set and a test data set extracted each time.

6. The method according to claim 5, characterized in that The inversion data set and the plurality of band operation results are input into a full subset screening algorithm to screen out the best variable combination, including: The model is trained based on the inversion dataset and test dataset extracted each time; The goodness of fit of the model trained based on the inversion data set and the test data set extracted each time is evaluated, and the best variable combination with the best goodness of fit is screened out.

7. The method according to claim 6, characterized in that The step of inputting the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by a plurality of algorithms includes: The optimal variable combination and the inversion data set extracted each time are input into a plurality of preset algorithm models to obtain a plurality of inversion models established by each algorithm.

8. The method according to claim 7, characterized in that The test data set is input into the inversion model established by each algorithm respectively, and the inversion models established by the multiple algorithms are tested by calculating the determination coefficient and the root mean square error to determine the best soil moisture inversion model of the target area, including: Inputting the test data set into the multiple inversion models established by each algorithm respectively, and verifying the inversion models based on goodness of fit and root mean square error; The inversion model with the highest goodness of fit and the smallest root mean square error is selected as the best target area soil moisture inversion model; Alternatively, the relative error coefficient of each inversion model is calculated based on the goodness of fit and the root mean square error, and the inversion model with the highest relative error coefficient is selected as the optimal target area soil moisture inversion model.

9. A regional soil moisture detection device, characterized in that: include: The acquisition module is used to obtain the current remote sensing satellite image and soil moisture sensor data in the target area; A detection module, used for inputting the current remote sensing satellite image into the latest soil moisture inversion model of the target area, so that the latest soil moisture inversion model of the target area inverts the soil moisture data of each pixel in the target area; A correction module, used for correcting the soil moisture data of each pixel point based on the soil moisture sensor data to obtain a soil moisture detection result of the target area; A model training module is used to obtain soil moisture measurement data in a sample soil and a first remote sensing satellite image within a corresponding measurement time period; perform band calculation on the first remote sensing satellite image to obtain multiple band calculation results reflected in the first remote sensing satellite image, wherein each band calculation result includes multiple parameter indexes; randomly extract an inversion data set and a test data set from the soil moisture measurement data and the multiple band calculation results, respectively, wherein the inversion data set is used to train a soil moisture inversion model, and the test data set is used to test the trained soil moisture inversion model; input the inversion data set and the multiple band calculation results into a full subset screening algorithm to screen out an optimal variable combination; input the optimal variable combination and the inversion data set into a plurality of preset algorithm models to obtain an inversion model established by each algorithm; input the test data set into the inversion models established by each algorithm, respectively, and perform model testing on the inversion models established by the plurality of algorithms by calculating the determination coefficient and the root mean square error to determine the optimal target area soil moisture inversion model.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the regional soil moisture detection method according to any one of claims 1 to 8 is implemented.

Images