Soil moisture standard quantitative detection method and soil moisture control method construction method

Abstract

The invention provides a soil moisture standard quantitative detection method and a soil moisture control party construction method, and belongs to the technical field of remote sensing, and the method comprises the following steps: defining soil moisture standard quantitative products which have unified specification and quality requirements, so that different satellite soil moisture remote sensing products have consistency and comparability; constructing a soil moisture standard quantitative product three-value evaluation system, wherein three values comprise a truth value, an agreed truth value and a relative truth value; constructing a soil moisture control party, and taking the soil moisture control party as a relative true value of a soil moisture standard quantitative product; and comparing the difference between a soil moisture standard quantitative product and a soil moisture control party as an authenticity inspection index. The method comprehensively unifies concepts such as a space-time framework parameter field, a standard quantitative product and a truth value, solves the problems of consistency and equivalence of an existing soil moisture remote sensing product, and improves the application level of a remote sensing satellite.

Description

Technical Field

[0001] The present invention belongs to the field of remote sensing technology, and particularly relates to the definition, construction method and truthfulness verification technology of soil moisture standard quantitative products. In particular, aiming at the problem of improving the quality consistency and equivalence of multi-source satellite remote sensing soil moisture products, a soil moisture standard quantitative detection method and a soil moisture control method construction method are proposed. Background Art

[0002] Soil moisture is an important and indispensable part of the global water cycle system, and is of great significance to agricultural production, ecosystem health and climate change research. Since the 1970s, a series of soil moisture product algorithms based on different satellite sensors (such as passive microwave, radar remote sensing, thermal infrared optics, etc.) have been developed one after another. These products vary in spatial resolution from 1-meter level to 100-meter level and even to the coarse grid resolution at the global scale (such as 9 km, 25 km), providing users with a variety of choices. However, due to the influence of multiple error sources (such as algorithms, sensors and physical processes), the quality of these satellite soil moisture products varies greatly, each with its own advantages and disadvantages. For example, there are differences in many aspects such as spatial resolution and spectral band settings between different satellites, which makes it difficult to collect ground verification data for each product, posing challenges to the consistency and equivalence of soil moisture remote sensing products. The existing technologies have the following problems:

[0003] High data acquisition cost: Although ground measured data has high accuracy, its acquisition cost is high, and the site distribution is sparse and uneven, making it difficult to meet the needs of large-scale and continuous monitoring. The ground verification method is costly and difficult to match the satellite pixel scale; due to differences in sensor parameters, inversion algorithms and error sources (such as surface heterogeneity) of different satellites, the product consistency is insufficient.

[0004] Complex multi-source data fusion: Traditional ground measurement methods (such as TDR, cosmic ray neutron meter) are costly and difficult to match the satellite pixel scale; there are differences in resolution, temporal resolution and physical meaning between different data sources, and problems such as scale conversion and data assimilation need to be solved during the fusion process, with high technical difficulty. The specifications of multi-source soil moisture products are not unified and cannot be directly fused and applied.

[0005] Limited model simulation accuracy: Model simulation requires a large amount of auxiliary data and is highly dependent on underlying surface conditions such as soil type and vegetation cover, and the accuracy and reliability of the simulation results are limited. There is a lack of standardized true value definition and test index system; there is a lack of a unified standardized test framework, resulting in limited product interoperability and operational application. Summary of the Invention

[0006] The present invention aims to propose a method for quantitatively detecting soil moisture standards and a method for constructing a soil moisture control party, and a method for authenticating the truthfulness of concepts such as a comprehensive unified spatio-temporal framework parameter field, a standard quantitative product, and a true value, so as to solve the problems of consistency and equivalence of existing soil moisture remote sensing products and improve the application level of remote sensing satellites. Specifically, it includes:

[0007] A method for quantitatively detecting soil moisture standards, comprising the following steps:

[0008] Step 1: Define a soil moisture standard quantitative product, which has unified specifications and quality requirements, so that different satellite soil moisture remote sensing products have consistency and comparability;

[0009] Step 2: Construct a three-value evaluation system for the soil moisture standard quantitative product. The three values include the true value, the conventional true value, and the relative true value. Among them, the true value is the theoretical value, and the conventional true value under certain measurement conditions is the relative true value, and the relative true value is used to replace the true value;

[0010] Step 3: Construct a soil moisture control party, and use the soil moisture control party as the relative true value of the soil moisture standard quantitative product;

[0011] Step 4: Use the difference between the soil moisture standard quantitative product and the soil moisture control party as an authenticity test index to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product..

[0012] A device for quantitatively detecting soil moisture standards, comprising:

[0013] A definition module that defines a soil moisture standard quantitative product, which has unified specifications and quality requirements, so that different satellite soil moisture remote sensing products have consistency and comparability;

[0014] An evaluation module that constructs a three-value evaluation system for the soil moisture standard quantitative product. The three values include the true value, the conventional true value, and the relative true value. Among them, the true value is the theoretical value, and the conventional true value under certain measurement conditions is the relative true value, and the relative true value is used to replace the true value;

[0015] A soil moisture control party construction module that constructs a soil moisture control party and uses the soil moisture control party as the relative true value of the soil moisture standard quantitative product;

[0016] An authenticity test module that uses the difference between the soil moisture standard quantitative product and the soil moisture control party as an authenticity test index to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product.

[0017] An electronic device, comprising: one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0018] A computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor implements the above-mentioned method.

[0019] A method for constructing a soil moisture control plot, comprising:

[0020] Clarify the specification requirements of the soil moisture standard quantitative product;

[0021] According to different specification requirements, divide the soil moisture control plot into different levels;

[0022] Select a corresponding area for sampling for plot design;

[0023] According to the designed plot, measure and obtain soil moisture data;

[0024] Construct the soil moisture control plot. According to the characteristics of the soil moisture data and the specification requirements, select a suitable model to construct different levels of soil moisture control plots to form plots;

[0025] Evaluate the soil moisture control plot by comparing with known true value data, and optimize and adjust the model to improve the quality of the soil moisture control plot.

[0026] A device for constructing a soil moisture control plot, comprising:

[0027] A specification requirement clarification module for clarifying the specification requirements of the soil moisture standard quantitative product;

[0028] A level division module for dividing the soil moisture control plot into different levels according to different specification requirements;

[0029] A plot design module for selecting a corresponding area for sampling for plot design;

[0030] A data measurement and acquisition module for measuring and obtaining soil moisture data according to the designed plot;

[0031] A soil moisture control plot construction module for selecting a suitable model according to the data characteristics and specification requirements to construct different levels of soil moisture control plots to form plots;

[0032] An optimization and adjustment module for comparing with known true value data and optimizing and adjusting the model to improve the quality of the soil moisture control plot.

[0033] An electronic device, characterized in that it comprises: one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method.

[0034] A computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor implements the above-mentioned method.

[0035] The present invention has the following beneficial technical effects:

[0036] (1) The definition of the soil moisture standard quantitative product proposed in this application unifies the product specifications and quality requirements, and solves the problems of large differences in the quality of different satellite soil moisture products and difficulty in comparing consistency and equivalence in the prior art. Through the standardized product definition, the soil moisture data obtained by different satellites can be processed and analyzed under the same framework, and a method for defining the soil moisture standard quantitative product with unified specification requirements and quality requirements is proposed to ensure the consistency and comparability of different satellite soil moisture products. By multi-source data fusion and machine learning models, the inspection accuracy is improved, and the problems of consistency and equivalence of remote sensing products are solved; the comparability and interoperability of data are improved, providing a solid data foundation for large-scale and multi-field remote sensing applications.

[0037] (2) The constructed soil moisture control party is used as the relative true value. A soil moisture control party with a clear hierarchy is constructed as the relative true value for the inspection of the soil moisture standard quantitative product, improving the accuracy and reliability of the verification data; providing an objective and accurate benchmark for the inspection of the soil moisture standard quantitative product. Compared with the uncertainty of the true value caused by the limitations of measurement means and methods in the prior art, the control party construction method of this application is more systematic and comprehensive, supports multi-level resolution inspections from 1m to 1000m, adapts to different satellite data, fully considers various factors such as soil type, terrain, and vegetation cover, and ensures the representativeness and accuracy of the control party. Through the comparative analysis of the control party and the remote sensing product, the errors and biases of the product can be evaluated more effectively, providing a clear direction for the improvement and optimization of the product, thereby improving the quality and application level of the remote sensing product.

[0038] (3)This application provides multiple methods for constructing soil moisture control parties, including those based on ground measured data, multi-source data fusion models, and intelligent models, etc. These methods each have their own advantages and can be flexibly selected according to different product specifications and application scenario requirements. Compared with the prior art, this method system is more perfect, can adapt to diverse data sources and complex surface conditions, and improves the feasibility and efficiency of constructing control parties. At the same time, when using an intelligent model to construct a control party, it can give full play to the advantages of machine learning algorithms, automatically select the most relevant features, capture the non-linear relationship between remote sensing data and soil moisture, and perform efficient soil moisture estimation, further improving the accuracy and stability of the control party. For example, machine learning is used to reduce the ground sampling density (e.g., only 1 control party is required for a 1000m resolution).

[0039] (4)The present invention defines the true value, conventional true value, and relative true value of the soil moisture standard quantitative product, and gives the calculation method of specific authenticity test indexes. These indexes can comprehensively and quantitatively evaluate the accuracy and uncertainty of the product, providing a scientific basis for product quality control and user selection. Compared with the prior art, the test indexes of this application are more systematic and comprehensive, can more accurately reflect the error size of the product and its correlation with the true value, and help improve the credibility of remote sensing products and their in-depth application in various fields. Provide authenticity test indexes, clarify the relevant definitions of product true values, give product test indexes, and provide a scientific basis for the quality assessment of soil moisture remote sensing products. Description of the Drawings

[0040] Figure 1 It is a schematic diagram of the data cube of the present invention;

[0041] Figure 2 It is a flowchart for constructing the soil moisture control party of the present invention. Detailed Embodiments

[0042] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, the present invention adopts the following technical solutions.

[0043] The present invention proposes a method for standard quantitative detection of soil moisture, including the following steps:

[0044] Step 1: Definition of the soil moisture standard quantitative product; including:

[0045] Specified soil moisture products are quantitative soil moisture datasets defined based on remote sensing technology according to unified specifications and quality requirements. Using observational data obtained from different satellite sensors (such as passive microwave, radar remote sensing, thermal infrared optics, etc.), through remote sensing inversion algorithms, these data are converted into quantitative information that can reflect soil moisture content. These products usually describe the spatial variation details of soil moisture at specific spatial resolutions (ranging from 1m to 1000m), meeting the refined requirements for soil moisture information in different application scenarios. Their application scenarios are extensive, including irrigation decision support in agricultural production, water cycle analysis in meteorological forecasting, and vegetation growth status assessment in ecological environment monitoring. In terms of computer hardware and internal performance, their processing process has certain requirements for computing resources. Especially when dealing with large amounts of remote sensing data and complex algorithm operations, high-performance computing devices are needed to ensure processing efficiency and accuracy.

[0046] Specified Soil Moisture Products (SSMP for short), as an important part of the land surface physical, chemical, and biological parameter product system, follow the conversion path of "radiometric calibration → parameter inversion → spatio-temporal reconstruction" in their production process. Relying on the input of optical microwave radiation products, through inversion algorithms driven by physical mechanisms, a data cube of soil volumetric water content with clear physical meaning is generated. SSMP adopts a standardized process design, and through an error propagation control model in the production process, effectively eliminates external algorithm differences, ensuring that the spatio-temporal consistency of product specifications meets the engineering constraint conditions of Δx ≤ 0.05° and Δt ≤ 6 h.

[0047] (1) Variety setting

[0048] Specified soil moisture products are classified based on the interaction mechanism between electromagnetic waves and soil media, constructing four categories of standardized product lineages: (1) Optical remote sensing products, based on the spectral absorption characteristics in the visible - near-infrared band; (2) Passive microwave remote sensing products, relying on the penetration characteristics of radiometers to achieve surface soil moisture inversion; (3) Active microwave remote sensing products, using the response relationship between the backscattering coefficient of C / X-band synthetic aperture radar and dielectric constant; (4) Multi-source fusion products, integrating the inversion results of multiple physical mechanisms through the D-S evidence theory framework.

[0049] (2) Specification setting

[0050] The spatio-temporal reference system adopts a dual-constraint standardization framework: (1) Spatial standard field. Based on a five-layer and fifteen-level spatial resolution hierarchical model, a multi-scale spatial reference from 0.01° to 1° is achieved through a double longitude-latitude grid system; (2) Temporal standard field. A UTC time coding system is constructed, and a sliding time window algorithm is used to achieve seamless connection of daily-scale, dekadal-scale, and inter-annual-scale products. This specification system realizes seamless integration of multi-source data through a geospatial transformation library.

[0051] (3)Quality requirements

[0052] A four-element quality evaluation model Q = (A, P, U, C) is established, where: Accuracy includes an absolute accuracy requirement of RMSE ≤ 0.04 m 3 / m 3 ; Precision is measured by the correlation coefficient R 2 ≥ 0.85 of cross-validation between products; Uncertainty quantifies the error propagation in the inversion process using the Monte Carlo method; Consistency requires Bias ≤ ±0.02 m 3 / m 3 in the overlapping area of multi-source products. A value traceability chain from sensors to end products is established in the quality evaluation process.

[0053] (4)Timeliness and scale requirements

[0054] The timeliness index adopts a time response function, and it is required that the operational products meet the lag thresholds of τ ≤ 12 h (daily products) and τ ≤ 72 h (dekadal products). It effectively supports cross-scale application requirements from local hydrological model assimilation to global water cycle research.

[0055] Step 2: Definition and construction of the true value, conventional true value, and relative true value of the soil moisture standard quantitative product; including:

[0056] The true value of the soil moisture standard quantitative product refers to the true connotation and appearance of soil moisture at the pixel scale, which is the real-world data that objectively exists and does not depend on the data and algorithms of remote sensing products. Since the absolute true value is difficult to obtain in actual measurements, the concepts of conventional true value and relative true value are introduced. The conventional true value is based on the true value definition as the standard and continuously approaches the true value through the improvement of measurement methods. It has clear sampling regulations, and its value changes with the equipment and the observation process. The improvement of technology can change the value of the conventional true value. The conventional true value is the value obtained within the ideal number of observations under the allowable deviation. This true value is accessible and infinitely close to the true value definition, but it cannot reach the true value definition and there is a certain degree of uncertainty to some extent. It can be obtained through multiple samplings using observation means without systematic errors. The relative true value is the true value that is as accurate as possible to approach the absolute true value based on the existing measurement means. The methods for constructing these true values include methods based on ground measured data, multi-source data fusion models, and intelligent models, etc. For example, when constructing the control value based on ground measured data, the ground multi-point observation values or footprint / patch observation values are directly used, but this method is limited by factors such as observation cost and site distribution. The multi-source data fusion model constructs a more accurate control value by integrating data with different accuracies and qualities, such as other satellite soil moisture products, terrain, vegetation products, and ground measurements. The intelligent model uses machine learning algorithms to model the non-linear relationship between multi-source remote sensing data and ground observation data to generate a spatially continuous true value of the soil moisture control value.

[0057] (1) True value

[0058] The numerical value of the observed object obtained by remote sensing is obtained by sampling by the remote sensor at a certain spatio-temporal scale. The parameter field value that is consistent with the specifications of the standard quantitative product and has a clear spatio-temporal range can be used as the true value definition of the standard quantitative product. This true value is the intrinsic attribute value of the measurement object in the ideal interference-free state, satisfying absolute objectivity, mathematical uniqueness, and physical certainty, and is the convergence limit of all actual measurement values.

[0059] The true value is the true connotation and appearance of soil moisture at the pixel scale. It objectively exists and is the data of the real world, not depending on the data and algorithms of any remote sensing product. The true value not only reflects the spatio-temporal field characteristics of soil moisture but also reflects the product specification requirements, and has various characteristics such as definition, authenticity, accuracy, and feasibility. Therefore, the true value with the spatio-temporal sampling characteristics of soil moisture and meeting the product specification requirements is the true value of the soil moisture standard quantitative product.

[0060] (2) Conventional true value

[0061] In actual work, by stipulating standard specifications, the spatio-temporal information sets of soil moisture obtained by different measurement methods can be optimized, or a certain certified measurement method can be specified. For example, the data accuracy of obtaining the true value of pixels multiple times within a certain error range is used to determine the true value of the soil moisture product at the pixel scale, that is, the conventional true value. This is a measurable numerical value of the spatio-temporal distribution characteristics of soil moisture that is regarded as accurate. We regard this conventional true value as the true value. The conventional true value is a true value at the experimental level, with accuracy and its accessibility, and is used as a substitute for the true value in actual work.

[0062] The conventional true value is based on the definition of the true value as the standard and continuously approaches the true value through the improvement of measurement methods. Its value changes with the equipment and the observation process, and the improvement of technology can change the value of the conventional true value. The conventional true value is the value obtained within the allowable deviation and within the ideal number of observations. This conventional true value has accessibility, infinitely approaching the true value, but cannot reach the theoretical true value, and there is a certain degree of uncertainty to a certain extent, that is, theoretically, it is obtained by multiple sampling using observation means without systematic errors. Its simplest form can be expressed by a mathematical equation:

[0063] (1)

[0064] (2)

[0065] represents the true value of the standard quantitative product of soil moisture, represents the observation data obtained by a certain observation means, β represents the weight factor, and t represents an infinite number of samplings under ideal conditions.

[0066] (3) Relative true value

[0067] Ground measured data is the main source of the true value at the pixel scale in authenticity verification. However, limited by measurement means and methods, the absolute true value cannot be obtained, and only the best approximation value can be obtained. The relative true value of the remote sensing standard quantitative product is to obtain a true value that is as accurate as possible and close to the absolute true value. Like the conventional true value, this true value has accessibility, infinitely approaching the true value, but cannot reach the true value definition, and there is a certain degree of uncertainty to a certain extent.

[0068] In actual measurement, due to different measurement conditions, such as different measurement methods, measurement equipment, measurement personnel, and measurement environment, etc., different true values may occur. This also precisely illustrates the immeasurability of the true value definition. There is only a conventional true value under certain measurement conditions, that is, the commonly used relative true value at present. In the extreme case, the average value of ideal measurements without systematic errors and with a sufficient number of measurements can be considered as the pixel-level conventional true value of the standard quantitative product. However, in limited actual measurements, there are both systematic errors and accidental errors. The difference that always exists between the measured value and the actual value is the error. The relative true value needs to improve its accuracy by continuously improving the measurement method, and its value will change with the improvement of the method.

[0069] In summary, the conventional true value is achievable, and the relative true value is measurable. The relative true value is an optimal solution in the actual situation, which is equivalent to a higher-precision observation. Generally, it is used as the true value to analyze and evaluate remote sensing information products. The relative true value of soil moisture as a benchmark is the measured true value under a semi-constrained intensity between indoor physical experiments and field experiments.

[0070] Step 3: Construction of the soil moisture control party; including:

[0071] Definition of the "five controls" data party; including: The data square (Data Square, abbreviated as data party DS) is a geographic spatio-temporal raster data organization model that spatially divides the Earth's surface based on the latitude-longitude grid and maps specific indicator data to grid cells. Its core lies in constructing a unified geographic grid framework, and through multi-dimensional information networks such as spatio-temporal relationships (temporal sequence, spatial distribution), attribute semantic associations, and evolution processes, it normalizes and integrates multi-source spatio-temporal data. This structured organization mode significantly improves the efficiency of computer artificial intelligence processing and big data analysis, so it is called the "intelligent cube of geographic spatio-temporal raster data". The data square (Data Square) is as Figure 1 shown, showing in the form of a geographic spatio-temporal raster how the Earth's surface is divided according to the latitude-longitude grid, marking different resolution levels (such as 1m, 100m, 1000m) and forming an indicator data mapping, reflecting the grid point sizes and data organizational structures at different resolutions.

[0072] Remote sensing sampling discretizes the continuous ground surface through finite-resolution imaging, and can generate multi-level grid data parties with a resolution of 10 -4 degrees to 10 degrees (the equator corresponds to about 1 cm to 1000 m). Its spatial resolution adopts a standardized grading system: 0.1m, 0.25m, 0.5m, 1m, 2.5m, 5m, 10m, 25m, 50m, 100m, 250m, 500m, 1000m, 2500m, 5000m.

[0073] The specific definition of the soil moisture control party is:

[0074] Based on remote sensing inversion technologies using multi-source satellite data such as optical / near-infrared, thermal infrared, and active / passive microwave, the dynamic monitoring of soil moisture has been realized. However, the significant differences in the spatial resolution of multi-satellite payloads have made it difficult to match ground validation with each product one by one. Therefore, it is necessary to construct a standardized and multi-level ground benchmark verification sample plot system for soil moisture, that is, the soil moisture control plot.

[0075] The soil moisture control plot is a high-precision relative true value generated through ground measurement and data processing. Based on known geographical coordinates and geometric forms, it represents the soil moisture field with spatial distribution characteristics, is used to quantitatively evaluate the deviation of remote sensing products, and optimizes the inversion algorithm through truthfulness verification to approach the true value. The resolution levels of the control plot are set as 1m, 5m, 10m, 25m, 50m, 100m, 250m, 500m, 1000m, etc. according to the geographical longitude and latitude grid and mainstream satellite parameters, with both standardization and full-domain coverage characteristics.

[0076] With the progress of intelligent observation technologies, soil moisture measurement devices have expanded from point scale (such as needle-type time domain reflectometer TDR) to surface scale (cosmic ray observatory, with a scale of hundreds of meters) and kilometer scale (large aperture scintillometer LAS). When the sample area is divided into m control plot units, its true average water content can be characterized by the mean value of each unit. For example, when validating the SSMP product with a 1000m spatial resolution, theoretically, 106 (1000×1000) 1m control plots, 104 (100×100) 10m control plots, 102 (10×10) 100m control plots, or 1 1000m control plot are required. The scale of the control plot is inversely proportional to the sampling frequency, that is, the larger the scale, the fewer the required sampling units.

[0077] The soil moisture control plot is an objective entity representing the true state of soil moisture at the pixel scale, and its attributes are independent of the data and inversion algorithm of remote sensing products. The construction of the control plot should follow the following principles: The soil moisture control plot should fully consider factors such as soil type, terrain, and vegetation cover type, select areas with representativeness of the underlying surface, and should cover the local main land cover types (except water bodies, ice and snow, towns, and construction land). The spatial range of the control plot should not be less than 3×3 times the spatial resolution of the SSMP to be tested. And it is advisable to select flat areas, avoid military restricted areas, airport air traffic control areas, and other dangerous areas, and try to avoid ground objects such as roads, shelter forests, and high-voltage lines, while ensuring relatively convenient transportation conditions. The construction of the soil moisture control plot uses the observation value closest to the transit time of the satellite data of the SSMP to be tested, and it is recommended to be within ±24h. Try to stagger the periods after precipitation, after irrigation, and during soil freezing. If conditions permit, auxiliary information should be obtained synchronously, including meteorological elements such as air temperature, humidity, and wind speed at a short distance above the soil surface.

[0078] The construction of the soil moisture control plot includes:

[0079] It is recommended to adopt the following three categories of core methods to construct the soil moisture control method. The first category is the method of constructing with in-pixel scale measured data. High-precision instruments (such as COSMOS cosmic ray observatory) are used to directly obtain the control data at the 500m level, or the control is constructed by converting the point and surface scales using multi-point observation values. The second category is the multi-source data fusion modeling method. Based on the fine-grid observation data, the control is constructed through optimization algorithms. The third category is to construct the control based on intelligent learning models.

[0080] (1) Construction based on in-pixel scale ground measured data

[0081] Directly use the multi-point ground observation values or footprint / patch observation values to construct the soil moisture control. As the in-pixel scale increases, using the measured data of the dense observation network can obtain verification results with relatively high reliability. However, this method has a high observation cost, and in practice, the ground observation stations are sparsely and unevenly distributed. Moreover, in actual field sampling, under certain error and confidence conditions, generally uniform [underlying surface sample plots] are selected. In fact, these measurement methods include various biases such as instrument measurement errors, ground sampling errors, scale conversion errors, and in-pixel spatial registration errors, and it is difficult to meet the requirements of large-scale, continuous, and dynamic monitoring of soil moisture, and it is difficult to meet the need for obtaining the true value.

[0082] The core of the authenticity test lies in obtaining the true value data, and related methods are continuously iterated and innovated: for example, it includes the optimization of ground sampling layout based on the spatial simulated annealing algorithm, the geostatistical layout method, the estimation sampling of the spatial non-uniform surface mean, the hierarchical optimization sampling, and the construction of the cross-validation reference data set, etc.

[0083] This construction method is limited by the measurement means and methods. In the actual acquisition of the true value in the authenticity test, the agreed true value cannot be obtained, and only the best approximation value can be obtained. Establishing a typical ecosystem soil automatic observation network (abbreviated as "space-based soil water molecule network"), covering major ecosystems (including grasslands, farmlands, deserts, and forests) and dry and wet climate zones, with unified data specifications and quality requirements, can obtain the relative true value of soil moisture at the 100m scale, which is a typical 100m-level soil moisture control.

[0084] (2) Construction based on multi-source data fusion model

[0085] With the increase of pixel scale, the surface heterogeneity also increases. Especially when the amount of ground measured data is small and it is difficult to represent the true value at the pixel scale, it is recommended to use methods such as the heterogeneous surface spatial inference model, regression Kriging, Bayesian maximum entropy, etc., and introduce auxiliary information with strong correlation with the estimated variable or similar spatial distribution pattern to improve the estimation accuracy. With the support of relevant auxiliary information and high-resolution remote sensing observations, a hierarchical Bayesian model can be constructed or the spatio-temporal Kriging method can be used to integrate various prior information and spatio-temporal observation data to achieve accurate estimation of soil control volume and its uncertainty evaluation under the constraint of the variable time-varying law. Such methods can be used for large-scale spatial heterogeneity research, can integrate various data with different accuracies and qualities, and construct soil control volume at the levels of 250m, 500m, and 1000m.

[0086] (3) Construction based on intelligent learning models

[0087] Remote sensing images usually have the characteristics of high dimension and large scale, and traditional statistical methods are difficult to process such complex data sets. Machine learning models (such as random forest, gradient boosting tree, etc.) can effectively process big data, quickly extract useful information from multi-dimensional data, and generate spatially continuous true values of soil moisture control volume. The spatial distribution and dynamic changes of soil moisture are affected by many factors, such as vegetation, terrain, climate, etc. Machine learning models can model the non-linear relationship between multi-source remote sensing data and ground observation data, automatically select the most relevant features, help researchers understand the main driving factors of soil moisture, and facilitate further optimization of the model. At the same time, during the model training process, the accuracy and stability of the final product can be improved by continuously adjusting hyperparameters and model structures.

[0088] When using machine learning models to construct the true value of soil moisture remote sensing products, first, multi-source remote sensing data and ground measured data need to be collected as model inputs and training targets. At the same time, auxiliary data such as meteorology and terrain are combined for expansion. Model selection usually includes classic machine learning algorithms such as random forest and gradient boosting tree, and the model performance is improved through feature selection and hyperparameter optimization. These models can capture the non-linear relationship between remote sensing data and soil moisture and perform efficient soil moisture estimation. Finally, cross-validation is used to evaluate the model accuracy, and a spatially continuous true value product of soil moisture is generated through interpolation methods.

[0089] The construction process of soil moisture control volume is mainly divided into six stages, and the construction process is as Figure 2As shown, it demonstrates the specific steps and interrelationships of six stages from requirements analysis to verification and optimization. The arrows indicate the data flow direction, helping to understand the logic and sequence of the entire construction process. The flowchart of soil moisture control construction includes modules of requirements analysis, hierarchical standardization, model training, and verification and optimization. First is requirements analysis, which clarifies the specification requirements of the soil moisture standard quantitative product, such as spatial resolution, temporal resolution, product accuracy, etc. Secondly is hierarchical standardization. According to different resolution requirements, the control is divided into different levels, such as 1m, 2.5m, 5m, etc. In the quadrat design stage, factors such as soil type, terrain, and vegetation cover type need to be considered, and representative areas are selected for sampling. In the data collection stage, according to the designed quadrats, various measurement means are used to obtain soil moisture data. In the control model construction stage, according to the data characteristics and product requirements, a suitable model is selected, such as a mean model based on ground-measured data, a multi-source data fusion model, or an intelligent model, etc., to construct the soil moisture control. Finally is the verification and optimization stage. By comparing with known true value data, the accuracy and reliability of the control are evaluated, and the model is optimized and adjusted to improve the quality of the control.

[0090] Step 4: Authenticity test of SSMP based on soil moisture control; including:

[0091] Taking into account the specification requirements of the soil moisture standard quantitative product, using the constructed soil moisture control as the relative true value, conduct in-orbit and ground verification of SSMP. The main operation process includes: conducting synchronous measurements for the SSMP to be verified; determining the control construction model to be carried out according to the hierarchical scale and measurement data; producing the soil moisture control for the selected level; and completing the inspection and evaluation of SSMP.

[0092] The authenticity test indicators of SSMP include accuracy evaluation indicators and uncertainty evaluation indicators;

[0093] Quality inspection includes the precision and reliability of values. Only when reaching a certain level of precision can the information meet its specification requirements and truthfully reflect the state and change law of the observed object. Moreover, this compliance must withstand the inspection of the entire transmission process, overcoming the complexity, instability of the objective target itself, and the incompleteness of information, as well as the systematic errors and random errors generated during data collection, processing, and application analysis, ensuring reliability in a certain time and space, and having a stable and traceable production process at the same time.

[0094] (1) Accuracy evaluation indicators

[0095] The available indicators for quantitatively expressing the accuracy of SSMP include: mean error (ME), mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE), and correlation coefficient (R). The calculations are shown in formulas (3)-(7).

[0096] (3)

[0097] (4)

[0098] (5)

[0099] (6)

[0100] (7)

[0101] (2)Uncertainty evaluation indicators

[0102] The uncertainty of SSMP is quantitatively expressed by indicators such as standard deviation ( ), covariance (COV), etc. The calculation formulas of the indicators are shown in (8)-(9).

[0103] (8)

[0104] (9)

[0105] In the formula:

[0106] m i ——The value of the i-th SSMP pixel, cm 3 cm -3 ,

[0107] d i ——The value of the i-th soil moisture control value, cm 3 cm -3 ,

[0108] n——The number of samples, pieces,

[0109] ——The mean value of the SSMP sample points, cm 3 cm -3 ,

[0110] ——The mean value of the soil moisture control data, cm 3 cm -3 .

[0111] The authenticity test indicators are used to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product. The accuracy evaluation indicators include mean error (ME), mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE), and correlation coefficient (R). These indicators are calculated by comparing the differences between the pixel values of the soil moisture standard quantitative product and the soil moisture control values, and can comprehensively reflect the error size of the product and its correlation with the true value. The uncertainty evaluation indicators use the standard deviation ( ), indicators such as covariance (COV), etc., are used to quantitatively express the uncertainty of the soil moisture standard quantitative product, reflecting the dispersion degree and stability of the data. In practical applications, these indicators can help users understand the quality status of the product, provide a basis for product improvement and optimization, and also provide a reference for users to select appropriate products in different application scenarios.

[0112] The present invention also provides a soil moisture standard quantitative detection device, including:

[0113] A definition module, which defines the soil moisture standard quantitative product. The soil moisture standard quantitative product has unified specifications and quality requirements, making different satellite soil moisture remote sensing products have consistency and comparability;

[0114] An evaluation module, which constructs a three - value evaluation system for the soil moisture standard quantitative product. The three values include the true value, the conventional true value, and the relative true value. Among them, the true value is the theoretical value, and the conventional true value under certain measurement conditions is the relative true value, and the relative true value is used to replace the true value;

[0115] A soil moisture control party construction module, which constructs the soil moisture control party, and uses the soil moisture control party as the relative true value of the soil moisture standard quantitative product;

[0116] A authenticity verification module, which uses the difference between the soil moisture standard quantitative product and the soil moisture control party as the authenticity verification index to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product.

[0117] The present invention also provides an electronic device, including: one or more processors; a memory for storing one or more programs. Among them, when the one or more programs are executed by the one or more processors, the one or more processors implement a soil moisture standard quantitative detection method of the present invention.

[0118] The present invention also provides a computer - readable storage medium, on which executable instructions are stored. When the instructions are executed by a processor, the processor implements a soil moisture standard quantitative detection method of the present invention.

[0119] On the other hand, the present invention also provides a soil moisture control party construction device, including:

[0120] A specification requirement clarification module, which clarifies the specification requirements of the soil moisture standard quantitative product;

[0121] A hierarchical division module, which divides the soil moisture control party into different levels according to different specification requirements;

[0122] A quadrat design module, which selects corresponding areas for sampling for quadrat design;

[0123] Data measurement and acquisition module, which measures and acquires soil moisture data according to the designed quadrats;

[0124] Soil moisture control square construction module, which selects appropriate models according to data characteristics and specification requirements to construct soil moisture control squares at different levels to form quadrats;

[0125] Optimization and adjustment module, which improves the quality of soil moisture control squares by comparing with known true value data and optimizing and adjusting the models.

[0126] The present invention also provides an electronic device, including: one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method for constructing a soil moisture control square.

[0127] The present invention also provides a computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor implements the above-mentioned method for constructing a soil moisture control square.

Claims

1. A method for standard quantitative detection of soil moisture, characterized in that, It includes the following steps: Step 1: Define the soil moisture standard quantitative product, which has unified specifications and quality requirements, so that different satellite soil moisture remote sensing products have consistency and comparability; Step 2: Construct a three-value evaluation system for the soil moisture standard quantitative product. The three values include the true value, the conventional true value, and the relative true value. Among them, the true value is the theoretical value, and the conventional true value under certain measurement conditions is the relative true value, and the relative true value is used to replace the true value; Step 3: Construct the soil moisture control side, and use the soil moisture control side as the relative true value of the soil moisture standard quantitative product; Step 4: Use the difference between the soil moisture standard quantitative product and the soil moisture control side as the authenticity test index to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product.

2. The soil moisture standard quantitative detection method according to claim 1, wherein In Step 1: The soil moisture standard quantitative product is a soil moisture quantitative data set defined based on remote sensing data according to unified specifications and quality requirements; the remote sensing data is the data obtained by using satellite sensors, and through the remote sensing inversion algorithm, the remote sensing data is converted into quantitative information that can reflect the soil moisture content.

3. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, The varieties of the soil moisture standard quantitative product include: Optical remote sensing product, based on the spectral absorption characteristics in the visible-near infrared band; Passive microwave remote sensing product, relying on the penetration characteristics of the radiometer to realize the inversion of surface soil moisture; Active microwave remote sensing product, using the response relationship between the backscattering coefficient of the C / X-band synthetic aperture radar and the dielectric constant; Multi-source fusion product, integrating the inversion results of multiple physical mechanisms.

4. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, The specifications of the soil moisture standard quantitative product include: Spatial standard field, based on the spatial resolution hierarchical model, and realizing the multi-scale spatial benchmark of 0.01°-1° through the double longitude and latitude grid; Temporal standard field, constructing the UTC time coding system, and using the sliding time window algorithm to realize the seamless connection of products at daily, dekadal, and interannual scales.

5. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, The quality evaluation method of the soil moisture standard quantitative product is: Establish a four-element quality evaluation model Q=(A,P,U,C), where: A is the accuracy; P is the reliability; U is the uncertainty; C is the consistency.

6. The soil moisture standard quantitative detection method according to claim 1, characterized in that, The timeliness and scale requirements of the soil moisture standard quantitative product are specifically: The timeliness index adopts the time response function τ, and the lag thresholds of τ≤12 h for daily products and τ≤72 h for dekadal products.

7. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, In Step 2, the conventional true value is represented by a mathematical equation: (1) (2) Represents the true value of the soil moisture standard quantitative product, represents the observed data obtained by a certain observation method, β represents the weight factor, and t represents an infinite number of samplings under ideal conditions.

8. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, The soil moisture control side is the relative true value generated through ground measurement and data processing. It is based on known geographical coordinates and geometric forms, and represents the soil moisture field with spatial distribution characteristics, which is used to quantitatively evaluate the product deviation, and optimize the inversion algorithm through the authenticity test to approach the true value.

9. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, The resolution levels of the soil moisture control side are set to 1m, 5m, 10m, 25m, 50m, 100m, 250m, 500m, 1000m levels according to the geographical longitude and latitude grid and the mainstream satellite parameters.

10. A method for quantitatively detecting soil moisture standards according to claim 1, characterized in that, Step 4 includes: The authenticity test indexes of the soil moisture standard quantitative product include the accuracy evaluation index and the uncertainty evaluation index; The accuracy evaluation indicators include: Mean Error (ME), Mean Absolute Error (MAE), Mean Relative Error (MRE), Root Mean Square Error (RMSE), and Correlation Coefficient (R); the calculations are shown in formulas (3)-(7): (3) (4) (5) (6) (7) The uncertainty evaluation indicators include: standard deviation , covariance COV. The calculation formulas for the indicators are shown in (8)-(9): (8) (9) Where: m i —— the pixel value of the i-th soil moisture standard quantitative product, d i —— The moisture control value of the i-th soil, n—the number of samples, —— The mean value of the sample points of the soil moisture standard quantitative product, —— Mean value of soil moisture control data.

11. A device for quantitatively detecting soil moisture standards, characterized in that, including: A definition module that defines the soil moisture standard quantitative product. The soil moisture standard quantitative product has unified specifications and quality requirements, enabling different satellite soil moisture remote sensing products to have consistency and comparability; An evaluation module that constructs a three-value evaluation system for the soil moisture standard quantitative product. The three values include the true value, the conventional true value, and the relative true value. Among them, the true value is the theoretical value, and the conventional true value under certain measurement conditions is the relative true value, and the relative true value is used to replace the true value; A soil moisture control party construction module that constructs the soil moisture control party and uses the soil moisture control party as the relative true value of the soil moisture standard quantitative product; A authenticity verification module that uses the difference between the soil moisture standard quantitative product and the soil moisture control party as the authenticity verification index to quantitatively evaluate the accuracy and uncertainty of the soil moisture standard quantitative product.

12. An electronic device, characterized in that, including: One or more processors; A memory for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, It stores executable instructions thereon, and when the instructions are executed by the processor, the processor implements the method described in any one of claims 1 to 10.

14. A method for constructing a soil moisture control system, characterized in that, including: Clarify the specification requirements of the soil moisture standard quantitative product; According to different specification requirements, divide the soil moisture control party into different levels; Select corresponding areas for sampling for plot design; According to the designed plots, measure and obtain soil moisture data; Soil moisture control party construction. According to the characteristics of the soil moisture data and the specification requirements, select a suitable model to construct different levels of soil moisture control parties to form plots; Evaluate the soil moisture control party and optimize and adjust the model to improve the quality of the soil moisture control party.

15. A method for constructing a soil moisture control system according to claim 14, characterized in that, The specification requirements include spatial resolution, temporal resolution, and product accuracy.

16. A method for constructing a soil moisture control system according to claim 14, characterized in that, Suitable models include the mean model based on ground measured data, the multi-source data fusion model, or the intelligent learning model.

17. A method for constructing a soil moisture control system according to claim 16, characterized in that, Construct the soil moisture control party based on the mean model of ground measured data; including: Directly use ground multi-point observation values or footprint / patch observation values to construct the soil moisture control party; used to construct the soil moisture control party at the 100m level.

18. A method for constructing a soil moisture control system according to claim 16, characterized in that Construct the soil moisture control party based on the multi-source data fusion model; including: Adopt the heterogeneous surface spatial inference model, regression Kriging, or Bayesian maximum entropy method, and introduce auxiliary information. With the support of the auxiliary information and remote sensing observation data, construct a hierarchical Bayesian model or adopt the spatio-temporal Kriging method, and comprehensively use the prior information and spatio-temporal observation data to construct the soil control parties at the 250m, 500m, and 1000m levels under the constraint of the variable time-varying law. The auxiliary information includes the atmospheric temperature, humidity, and wind speed in the vicinity above the soil surface.

19. A method for constructing a soil moisture control system according to claim 16, characterized in that, Construct the soil moisture control party based on the intelligent learning model; including: Collect multi-source remote sensing data and ground measured data as the input and training targets of the intelligent learning model; at the same time, expand by combining various auxiliary data such as meteorology and terrain, select machine learning algorithms, improve the model performance through feature selection and hyperparameter optimization, finally evaluate the model accuracy using cross-validation, and generate spatially continuous soil moisture control squares through interpolation methods.

20. A device for constructing a soil moisture control system, characterized in that, Including: Specification requirement clarification module, which clarifies the specification requirements of the soil moisture standard quantitative product; Hierarchical division module, which divides the soil moisture control squares into different levels according to different specification requirements; Quadrat design module, which selects corresponding areas for sampling for quadrat design; Data measurement and acquisition module, which measures and acquires soil moisture data according to the designed quadrats; Soil moisture control square construction module, which selects a suitable model according to the characteristics of the soil moisture data and specification requirements to construct different levels of soil moisture control squares to form quadrats; Optimization and adjustment module, which optimizes and adjusts the model to improve the quality of the soil moisture control squares.

21. An electronic device, characterized in that, Including: One or more processors; A memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 14 to 19.

22. A computer-readable storage medium, characterized in that, Stored thereon are executable instructions, which when executed by a processor cause the processor to implement the method according to any one of claims 14 to 19.

Images