CN116049768B - Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model - Google Patents
Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model Download PDFInfo
- Publication number
- CN116049768B CN116049768B CN202310343942.9A CN202310343942A CN116049768B CN 116049768 B CN116049768 B CN 116049768B CN 202310343942 A CN202310343942 A CN 202310343942A CN 116049768 B CN116049768 B CN 116049768B
- Authority
- CN
- China
- Prior art keywords
- soil moisture
- microwave
- active
- fusion
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, and relates to the field of microwave remote sensing, wherein the method comprises the following steps: acquiring current active microwave soil moisture data and current passive microwave soil moisture data; inputting current active microwave soil moisture data and current passive microwave soil moisture data into a soil moisture data fusion model, and acquiring active and passive microwave soil moisture fusion data output by the soil moisture data fusion model, wherein the soil moisture data fusion model is determined according to active microwave fusion weights and passive microwave fusion weights; the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model. The invention can comprehensively utilize the respective advantages of the active microwave soil moisture product and the passive microwave soil moisture product, and improve the precision of soil moisture data.
Description
Technical Field
The invention relates to the field of microwave remote sensing, in particular to an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model.
Background
The performance of active microwave remote sensing and passive microwave remote sensing on soil moisture inversion accuracy is advantageous, however, the accuracy of a single active soil moisture product or passive soil moisture product is low, and the soil moisture monitoring requirement cannot be met.
Disclosure of Invention
The invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, which is used for solving the technical problem of lower accuracy of soil moisture data in the prior art and provides a technical scheme of active and passive microwave soil moisture fusion.
In a first aspect, the invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, which comprises the following steps:
acquiring current active microwave soil moisture data and current passive microwave soil moisture data;
inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model;
The soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight;
the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model;
the generalized linear regression model is determined by performing generalized linear regression on the active microwave back scattering coefficient after the standardized regression treatment, the passive microwave bright temperature after the standardized regression treatment and the model soil moisture data;
the model soil moisture data is obtained according to a preset soil moisture model.
According to the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model provided by the invention, the acquisition of current active microwave soil moisture data and current passive microwave soil moisture data comprises the following steps:
transmitting a first indication instruction to an active microwave soil moisture product, and receiving current active microwave soil moisture data output by the active microwave soil moisture product;
and sending a second instruction to the passive microwave soil moisture product, and receiving current passive microwave soil moisture data output by the passive microwave soil moisture product.
According to the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model provided by the invention, the current active microwave soil moisture data and the current passive microwave soil moisture data are input into the soil moisture data fusion model, and the active and passive microwave soil moisture fusion data output by the soil moisture data fusion model are obtained, comprising the following steps:
inputting the current active microwave soil moisture data to a first output layer of the soil moisture data fusion model to obtain a first fusion value output by the first output layer according to the current active microwave soil moisture data and the active microwave fusion weight;
inputting the current passive microwave soil moisture data to a first output layer of the soil moisture data fusion model to obtain a second fusion value output by the first output layer according to the current passive microwave soil moisture data and the passive microwave fusion weight;
and inputting the first fusion value and the second fusion value to a second output layer of the soil moisture data fusion model to obtain the active and passive microwave soil moisture fusion data output by the second output layer according to the first fusion value and the second fusion value.
According to the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model, before inputting the current active microwave soil moisture data and the current passive microwave soil moisture data into the soil moisture data fusion model, the active and passive microwave soil moisture fusion algorithm further comprises the following steps:
performing generalized linear regression on the normalized back scattering coefficient, the normalized passive microwave bright temperature and the model soil moisture data corresponding to each historical time in all historical times, determining a generalized linear regression model, and obtaining a back scattering coefficient normalized regression coefficient and a bright temperature normalized regression coefficient in the generalized linear regression model;
and determining the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient.
According to the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model, before the active microwave backward scattering coefficient after the standardized regression processing, the passive microwave brightness temperature after the standardized regression processing and the model soil moisture data corresponding to each moment in all moments are subjected to generalized linear regression processing, the algorithm further comprises:
Acquiring active microwave backward scattering coefficients, passive microwave brightness temperatures and model soil moisture data corresponding to each historical moment;
the active microwave back scattering coefficient and the passive microwave bright temperature are processed through standardized regression, and the active microwave back scattering coefficient after standardized regression and the passive microwave bright temperature after standardized regression are obtained;
and traversing all historical time until the active microwave backward scattering coefficient after the standardized regression processing and the passive microwave bright temperature after the standardized regression processing corresponding to each time are obtained.
According to the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model provided by the invention, the determining of the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient comprises the following steps:
determining a normalized regression total coefficient according to the backscatter coefficient normalized regression coefficient and the bright temperature normalized regression coefficient;
determining the active microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the standardized regression total coefficient;
And determining the passive microwave fusion weight according to the brightness temperature standardized regression coefficient and the standardized regression total coefficient.
In a second aspect, an active-passive microwave soil moisture fusion device based on a generalized linear regression model is provided, including:
a first acquisition unit: the method comprises the steps of acquiring current active microwave soil moisture data and current passive microwave soil moisture data;
a second acquisition unit: the method comprises the steps of inputting current active microwave soil moisture data and current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model;
the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight;
the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model;
the generalized linear regression model is determined by performing generalized linear regression on the active microwave back scattering coefficient after the standardized regression treatment, the passive microwave bright temperature after the standardized regression treatment and the model soil moisture data;
The model soil moisture data is obtained according to a preset soil moisture model.
In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model as described in any one of the above when executing the program.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model as described in any one of the above.
The invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, which is characterized in that the generalized linear regression model is determined after generalized linear regression is carried out on an active microwave back scattering coefficient after standardized regression processing, passive microwave bright temperature after standardized regression processing and model soil moisture data, active microwave fusion weights and passive microwave fusion weights are determined according to the back scattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient in the generalized linear regression model, active and passive microwave soil moisture fusion data are determined according to current active microwave soil moisture data and the corresponding active microwave fusion weights thereof, current passive microwave soil moisture data and the corresponding passive microwave fusion weights thereof, and the active and passive microwave soil moisture fusion data have definite physical significance.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is one of the flow diagrams of the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model provided by the invention;
FIG. 2 is a schematic flow chart of acquiring current active microwave soil moisture data and current passive microwave soil moisture data provided by the invention;
FIG. 3 is a schematic flow chart of acquiring active and passive microwave soil moisture fusion data provided by the invention;
FIG. 4 is a second schematic flow chart of the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model;
FIG. 5 is a third schematic flow chart of the active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model provided by the invention;
FIG. 6 is a schematic flow chart of an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model;
FIG. 7 is a schematic structural diagram of an active and passive microwave soil moisture fusion device based on a generalized linear regression model;
fig. 8 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Soil moisture plays an important role in processes and feedback of surface energy, moisture circulation, atmospheric circulation, climate change and the like, and is one of the most important parameters in applications such as flood forecasting, drought monitoring, crop yield forecasting, surface and climate model evaluation and the like. By accurately monitoring the space-time distribution of the earth surface soil moisture, the method is vital to grain safety, water resource optimization management and ecological protection, and compared with other soil moisture observation modes, microwave remote sensing is considered as the most effective means for acquiring large-scale soil moisture data, and the performance of active microwave and passive microwave remote sensing on soil moisture inversion precision is advantageous at present.
The method for researching the active and passive microwave soil moisture fusion can adopt an average weight method, is simple but has no clear physical meaning, and the respective advantages of active microwave and passive microwave soil moisture products are not fully utilized. Therefore, in order to solve the above problems, the present invention provides an active-passive microwave soil moisture fusion algorithm based on a generalized linear regression model, and fig. 1 is one of flow diagrams of the active-passive microwave soil moisture fusion algorithm based on the generalized linear regression model, provided by the present invention, and includes:
102, inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model;
the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight;
the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model;
The generalized linear regression model is determined by performing generalized linear regression on the active microwave back scattering coefficient after the standardized regression treatment, the passive microwave bright temperature after the standardized regression treatment and the model soil moisture data;
the model soil moisture data is obtained according to a preset soil moisture model.
In step 101, the present invention acquires current active microwave soil moisture data and current passive microwave soil moisture data in real time, and acquires soil moisture data with physical significance by taking the current active microwave soil moisture data and the current passive microwave soil moisture data as fusion parameters, wherein the physical significance is that the present invention can fully consider sensitivity to soil moisture when the backscattering coefficient and the bright temperature are cooperated, and comprehensively utilizes the high precision advantages of active microwave soil moisture products in medium vegetation and passive microwave soil moisture products in sparse vegetation to obtain soil moisture data with high precision under different vegetation coverage conditions.
According to the invention, the current active microwave soil moisture data is obtained through the active microwave soil moisture product, and the current passive microwave soil moisture data is obtained through the passive microwave soil moisture product.
In step 102, mainly for constructing the soil moisture data fusion model, the construction of the soil moisture data fusion model needs to determine an active microwave fusion weight and a passive microwave fusion weight, where the active microwave fusion weight is used for reflecting the importance degree of the current active microwave soil moisture data in the active and passive microwave soil moisture fusion data, and the passive microwave fusion weight is used for reflecting the importance degree of the current passive microwave soil moisture data in the active and passive microwave soil moisture fusion data.
In order to determine the active microwave fusion weight and the passive microwave fusion weight, the active microwave fusion weight and the passive microwave fusion weight in the invention are determined by the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient in the generalized linear regression model, and can be intuitively reflected by the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient, so the invention further needs to determine the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient.
Because the active microwave backward scattering coefficient and the passive microwave bright temperature have different units, the sensitivity degree of the active microwave backward scattering coefficient and the passive microwave bright temperature to soil moisture cannot be directly compared and analyzed, standardized regression processing is carried out on the backward scattering coefficient and the bright temperature before modeling of a generalized linear regression model, the generalized linear regression model is determined after generalized linear regression is carried out on the active microwave backward scattering coefficient after the standardized regression processing, the passive microwave bright temperature after the standardized regression processing and model soil moisture data, and the active microwave fusion weight and the passive microwave fusion weight are determined according to the backward scattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient in the generalized linear regression model.
According to the invention, by establishing a generalized linear regression model between the active microwave backward scattering coefficient and the passive microwave brightness and soil moisture, the sensitivity and the cooperative response of the backward scattering coefficient and the brightness to the soil moisture are analyzed, and an active and passive microwave soil moisture fusion algorithm with clear physical significance based on the generalized linear regression model is established, so that the advantages of an active and passive microwave soil moisture product are fully utilized. The invention can effectively fuse the existing active and passive microwave soil moisture products, not only can improve the precision of single active or passive soil moisture products, but also can improve the effective observation quantity of soil moisture.
Optionally, the model soil moisture data is obtained according to a preset soil moisture model, and the preset soil moisture model can be constructed by a water balance equation or obtained by calculation and analysis according to a model soil moisture product.
The method solves the problems that the conventional average weight method has no definite physical meaning and does not fully utilize the advantages of active microwave and passive microwave soil moisture products, models the active microwave backward scattering coefficient, passive microwave brightness and soil moisture by utilizing a generalized linear regression model, and performs weighted fusion on active and passive microwave soil moisture data from the angle of sensitivity of a microwave observation signal to the soil moisture.
The invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, which is characterized in that the generalized linear regression model is determined after generalized linear regression is carried out on an active microwave back scattering coefficient after standardized regression processing, passive microwave bright temperature after standardized regression processing and model soil moisture data, active microwave fusion weights and passive microwave fusion weights are determined according to the back scattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient in the generalized linear regression model, active and passive microwave soil moisture fusion data are determined according to current active microwave soil moisture data and the corresponding active microwave fusion weights thereof, current passive microwave soil moisture data and the corresponding passive microwave fusion weights thereof, and the active and passive microwave soil moisture fusion data have definite physical significance.
Fig. 2 is a schematic flow chart of acquiring current active microwave soil moisture data and current passive microwave soil moisture data, where the acquiring the current active microwave soil moisture data and the current passive microwave soil moisture data includes:
In step 1011, the present invention sends the first instruction at a preset time, where the first instruction is used to instruct an active microwave soil moisture product to obtain current active microwave soil moisture data, and after receiving the first instruction, the active microwave soil moisture product calculates, analyzes and determines the current active microwave soil moisture data, sends the current active microwave soil moisture data to the present device, and the present device receives the current active microwave soil moisture data.
In step 1012, the present invention sends the second instruction at the preset time, that is, while sending the first instruction, where the second instruction is used to instruct a passive microwave soil moisture product to obtain current passive microwave soil moisture data, and after receiving the second instruction, the passive microwave soil moisture product calculates, analyzes and determines the current passive microwave soil moisture data, sends the current passive microwave soil moisture data to the present device, and the present device receives the current passive microwave soil moisture data.
Fig. 3 is a schematic flow chart of acquiring active and passive microwave soil moisture fusion data, inputting the current active microwave soil moisture data and the current passive microwave soil moisture data into a soil moisture data fusion model, and acquiring active and passive microwave soil moisture fusion data output by the soil moisture data fusion model, where the flow chart includes:
In step 1021, a first layer of the soil moisture data fusion model is configured to determine a first fusion value according to a product of the current active microwave soil moisture data and the active microwave fusion weight.
In step 1022, the first layer of the soil moisture data fusion model is further configured to determine a second fusion value according to a product of the current passive microwave soil moisture data and the passive microwave fusion weight.
In step 1023, a second layer of the soil moisture data fusion model is configured to determine the active and passive microwave soil moisture fusion data based on a sum of the first fusion value and the second fusion value.
The skilled person understands that the active and passive microwave soil moisture fusion data refer to high-precision soil moisture fusion data obtained by weighting and fusing active microwave soil moisture products and passive microwave soil moisture products by using different weight schemes. The acquisition mode of the active and passive microwave soil moisture fusion data is as follows:
in the formula (1), the components are as follows,
representing the active and passive microwave soil moisture fusion data, < > water>
Is the current active microwave soil moisture data, ++>
Is the current passive microwave soil moisture data, < > water >
For the active microwave fusion weight, +.>
And (5) fusing the weight for the passive microwaves.
Fig. 4 is a second schematic flow chart of the active and passive microwave soil moisture fusion algorithm based on the generalized linear regression model, and before inputting the current active microwave soil moisture data and the current passive microwave soil moisture data into the soil moisture data fusion model, the method further includes:
and 202, determining the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient.
In step 201, the historical time is a time before the current time, and the active microwave backward scattering coefficient, the passive microwave brightness temperature and the model soil moisture data corresponding to each historical time in all the historical times are stored in a database or a storage medium.
The active microwave backward scattering coefficient and the passive microwave bright temperature are obtained through remote sensing of the soil moisture product, and model soil moisture data which corresponds to each historical moment and is determined by the model soil moisture product are obtained while the active microwave backward scattering coefficient and the passive microwave bright temperature which correspond to each historical moment are obtained.
In an optional embodiment, generalized linear regression processing is performed on the active microwave back scattering coefficient, the passive microwave bright temperature and the model soil moisture data corresponding to each of all the historical moments, a generalized linear regression model is determined, the back scattering coefficient regression coefficient and the bright Wen Huigui coefficient in the generalized linear regression model are obtained, and the active microwave fusion weight and the passive microwave fusion weight are determined according to the back scattering coefficient regression coefficient and the bright Wen Huigui coefficient.
Specifically, the generalized linear regression model is composed of a random part, a system part and a connection function, wherein the random part refers to a response variable (soil moisture) and probability distribution thereof. The response variable obeys various distributions, and the soil moisture is more suitable for beta distribution. The system part refers to an interpretation variable, which can be an attribute variable or a continuous variable. The join function refers to a function that joins the model predictive value and the system part. And modeling the active microwave back scattering coefficient, the passive microwave brightness and the soil moisture by using a generalized linear regression model. The formula of the generalized linear regression model before improvement is as follows:
In the formula (2), the amino acid sequence of the compound,
for model predictive value, +.>
For model soil moisture data, ++>
Is an averaging operator.
In the formula (3), the amino acid sequence of the compound,
as a connection function, the connection function under the beta distribution is
，/>
Is the active microwave back scattering coefficient, +.>
For the passive microwave light-up temperature,
is the regression coefficient of the model.
As an improvement, in order to overcome the technical barriers that the active microwave back scattering coefficient and the passive microwave bright temperature have different magnitude units, the invention adopts the parameters after standardized regression to construct a generalized linear regression model, thereby realizing the direct comparative analysis of the sensitivity degree of the two to soil moisture, specifically, the standardized regression is used for processing the active microwave back scattering coefficient and the passive microwave bright temperature, performing generalized linear regression on the normalized back scattering coefficient, the normalized passive microwave bright temperature and the model soil moisture data corresponding to each historical time in all historical times, determining a generalized linear regression model, and obtaining a back scattering coefficient normalized regression coefficient and a bright temperature normalized regression coefficient in the generalized linear regression model; and determining the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient.
The formula of the improved generalized linear regression model is as follows:
in the formula (4), the amino acid sequence of the compound,
for model predictive value, +.>
For model soil moisture data, ++>
Is an averaging operator.
In the formula (5), the amino acid sequence of the compound,
as a connection function, the connection function under the beta distribution is
，/>
For the normalized regression-processed active microwave back-scattering coefficient,/>
Is normalized and returned to the passive microwave bright temperature +.>
Is a normalized regression coefficient of the model.
In step 202, the determining the active microwave fusion weight and the passive microwave fusion weight according to the backscatter coefficient normalization regression coefficient and the bright temperature normalization regression coefficient includes:
determining a normalized regression total coefficient according to the backscatter coefficient normalized regression coefficient and the bright temperature normalized regression coefficient;
determining the active microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the standardized regression total coefficient;
and determining the passive microwave fusion weight according to the brightness temperature standardized regression coefficient and the standardized regression total coefficient.
Assuming that the backscatter coefficient normalized regression coefficient is
The standard regression coefficient of the brightness temperature is +. >
Determining the normalized regression total coefficient as +.f according to the sum of the backscattering coefficient normalized regression coefficient and the bright temperature normalized regression coefficient>
。
Determining the active microwave fusion weight according to the backscattering coefficient normalized regression coefficient and the normalized regression total coefficient can refer to the following formula:
in the formula (6), the amino acid sequence of the compound,
normalization of regression coefficients for backscattering coefficients, < >>
For the bright temperature standard regression coefficient,
is the active microwave fusion weight.
Optionally, the passive microwave fusion weight is determined according to the brightness temperature standardized regression coefficient and the standardized regression total coefficient, and the following formula can be referred to:
in the formula (7), the amino acid sequence of the compound,
normalization of regression coefficients for backscattering coefficients, < >>
For the bright temperature standard regression coefficient,
is a passive microwave fusion weight.
Fig. 5 is a third schematic flow chart of an active-passive microwave soil moisture fusion algorithm based on a generalized linear regression model, where before performing generalized linear regression on an active microwave back scattering coefficient after normalized regression processing, a passive microwave bright temperature after normalized regression processing, and model soil moisture data corresponding to each of all moments, the algorithm further includes:
In step 301, it is understood by those skilled in the art that the active microwave back-scattering coefficient and the passive microwave bright temperature are satellite observed data for inverting the soil moisture, the active microwave soil moisture product is obtained by combining the active microwave back-scattering coefficient with a corresponding algorithm, the passive microwave soil moisture product is obtained by combining the passive microwave bright temperature with a corresponding algorithm, the active microwave back-scattering coefficient and the passive microwave bright temperature are satellite observed data for obtaining the active microwave soil moisture product and the passive microwave soil moisture product, respectively, and the model soil moisture data corresponding to the active back-scattering coefficient and the passive microwave bright temperature at the historical time are obtained.
In step 302, the active microwave back scattering coefficient and the passive microwave bright temperature are normalized and regression processed, so as to obtain the active microwave back scattering coefficient and the passive microwave bright temperature after normalization regression processing.
The formula for performing standardized regression processing on the active microwave back scattering coefficient and the passive microwave bright temperature can be referred to as follows:
in the formula (8), the amino acid sequence of the compound,
is the active microwave backward scattering coefficient or the passive microwave bright temperature, +.>
Is the standard deviation of the active microwave back scattering coefficient or the standard deviation of the passive microwave bright temperature, +.>
Is normalized to the active microwave back scattering coefficient or passive microwave bright temperature, ++>
Is the average value of the active microwave back scattering coefficient and the passive microwave bright temperature, +.>
For different moments in the time series.
In step 303, according to step 301 and step 302, the present invention calculates the active microwave backward scattering coefficient and the passive microwave bright temperature corresponding to all the historical moments respectively, obtains the average value and standard deviation of the active microwave backward scattering coefficient and the average value and standard deviation of the passive microwave bright temperature, and then traverses each moment according to the average value and standard deviation to perform standardization processing, and obtains the active microwave backward scattering coefficient after the standardization regression processing and the passive microwave bright temperature after the standardization regression processing corresponding to each moment.
FIG. 6 is a schematic flow chart of an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, wherein the generalized linear regression model is constructed through an active microwave backward scattering coefficient, a passive microwave bright temperature and model soil moisture, a standardized regression coefficient is determined according to the generalized linear regression model, so that active and passive microwave fusion weights are determined, active and passive microwave soil moisture fusion data are determined according to current active microwave soil moisture data and current passive microwave soil moisture data which are acquired currently and corresponding active and passive microwave fusion weights thereof.
Modeling the standardized active microwave backward scattering coefficient and the microwave passive lighting temperature and model soil moisture data by using generalized linear regression, calculating a fusion weight by using the standardized regression coefficient of the model, weighting and fusing the current active microwave soil moisture data and the current passive microwave soil moisture data to obtain active and passive microwave soil moisture fusion data, finally performing accuracy verification on the active and passive microwave soil moisture fusion data by using the third-party soil moisture actual measurement data, and comparing and analyzing the active and passive microwave soil moisture fusion data based on an average weight method and original active microwave and passive microwave soil moisture products.
Specifically, the method also comprises the steps of performing accuracy verification on the active and passive microwave soil moisture fusion data, performing accuracy verification on the active and passive microwave soil moisture fusion product based on the method by utilizing the third-party soil moisture actual measurement data, and performing comparison analysis on an average weight method and the original active microwave and passive microwave soil moisture product. The evaluation indexes selected by the invention comprise Root Mean Square Error (RMSE), average deviation Bias, unbiased root mean square error (ubRMSE) and correlation coefficientRAs shown in table 1:
TABLE 1
Table 1 shows the average value of the measured accuracy verification results on the grid scale according to the method provided by the invention, the average weight method fusion, the active microwave soil moisture product acquisition method and the passive microwave soil moisture product acquisition method. It can be seen from table 1 that the method provided by the invention has the highest precision, the precision is superior to that of the fusion of the traditional average weight method, and the absolute precision indexes, namely Root Mean Square Error (RMSE) and unbiased root mean square error (ubRMSE), of the invention are also superior to that of the original, and the fusion method provided by the invention can effectively improve the precision of the original active or passive microwave soil moisture products according to the active microwave soil moisture product acquisition method and the passive microwave soil moisture product acquisition method.
Fig. 7 is a schematic structural diagram of an active and passive microwave soil moisture fusion device based on a generalized linear regression model, and the invention also provides an active and passive microwave soil moisture fusion device based on a generalized linear regression model, which comprises a first acquisition unit 1: for obtaining the current active microwave soil moisture data and the current passive microwave soil moisture data, the working principle of the first obtaining unit 1 may refer to the foregoing step 101, which is not described herein again.
The active and passive microwave soil moisture fusion device based on the generalized linear regression model further comprises a second acquisition unit 2: the working principle of the second obtaining unit 2 may refer to the foregoing step 102, and will not be repeated herein.
The soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight;
the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model;
The generalized linear regression model is determined by performing generalized linear regression on the active microwave back scattering coefficient after the standardized regression treatment, the passive microwave bright temperature after the standardized regression treatment and the model soil moisture data;
the model soil moisture data is obtained according to a preset soil moisture model.
The invention provides an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model, which is characterized in that the generalized linear regression model is determined after generalized linear regression is carried out on an active microwave back scattering coefficient after standardized regression processing, passive microwave bright temperature after standardized regression processing and model soil moisture data, active microwave fusion weights and passive microwave fusion weights are determined according to the back scattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient in the generalized linear regression model, active and passive microwave soil moisture fusion data are determined according to current active microwave soil moisture data and the corresponding active microwave fusion weights thereof, current passive microwave soil moisture data and the corresponding passive microwave fusion weights thereof, and the active and passive microwave soil moisture fusion data have definite physical significance.
Fig. 8 is a schematic structural diagram of an electronic device provided by the present invention. As shown in fig. 8, the electronic device may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may call logic instructions in the memory 830 to perform a generalized linear regression model-based active-passive microwave soil moisture fusion algorithm, the method comprising: acquiring current active microwave soil moisture data and current passive microwave soil moisture data; inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model; the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight; the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model; the generalized linear regression model is determined by performing generalized linear regression on an active microwave back scattering coefficient after standardized regression treatment, a passive microwave bright temperature after standardized regression treatment and model soil moisture data, wherein the model soil moisture data is obtained according to a preset soil moisture model.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, where the computer program product includes a computer program, where the computer program can be stored on a non-transitory computer readable storage medium, where the computer program, when executed by a processor, can perform an active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model provided by the above methods, where the method includes: acquiring current active microwave soil moisture data and current passive microwave soil moisture data; inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model; the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight; the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model; the generalized linear regression model is determined by performing generalized linear regression on an active microwave back scattering coefficient after standardized regression treatment, a passive microwave bright temperature after standardized regression treatment and model soil moisture data, wherein the model soil moisture data is obtained according to a preset soil moisture model.
In yet another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model provided by the above methods, the method comprising: acquiring current active microwave soil moisture data and current passive microwave soil moisture data; inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model; the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight; the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model; the generalized linear regression model is determined by performing generalized linear regression on an active microwave back scattering coefficient after standardized regression treatment, a passive microwave bright temperature after standardized regression treatment and model soil moisture data, wherein the model soil moisture data is obtained according to a preset soil moisture model.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (3)
1. An active and passive microwave soil moisture fusion algorithm based on a generalized linear regression model is characterized by comprising the following steps:
acquiring current active microwave soil moisture data and current passive microwave soil moisture data;
inputting the current active microwave soil moisture data and the current passive microwave soil moisture data to a soil moisture data fusion model, and obtaining active and passive microwave soil moisture fusion data output by the soil moisture data fusion model;
the soil moisture data fusion model is determined according to the active microwave fusion weight and the passive microwave fusion weight;
the active microwave fusion weight and the passive microwave fusion weight are determined according to a backscattering coefficient standardized regression coefficient and a bright temperature standardized regression coefficient in a generalized linear regression model;
The generalized linear regression model is determined by performing generalized linear regression on the active microwave back scattering coefficient after the standardized regression treatment, the passive microwave bright temperature after the standardized regression treatment and the model soil moisture data;
the model soil moisture data are obtained according to a preset soil moisture model;
before inputting the current active microwave soil moisture data and the current passive microwave soil moisture data into the soil moisture data fusion model, the method further comprises:
performing generalized linear regression on the normalized back scattering coefficient, the normalized passive microwave bright temperature and the model soil moisture data corresponding to each historical time in all historical times, determining a generalized linear regression model, and obtaining a back scattering coefficient normalized regression coefficient and a bright temperature normalized regression coefficient in the generalized linear regression model;
determining the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient;
before generalized linear regression is performed on the normalized back-scattering coefficient of the active microwave after the regression process, the normalized passive microwave bright temperature after the regression process and the model soil moisture data corresponding to each time, the method further comprises:
Acquiring active microwave backward scattering coefficients, passive microwave brightness temperatures and model soil moisture data corresponding to each historical moment;
the active microwave back scattering coefficient and the passive microwave bright temperature are processed through standardized regression, and the active microwave back scattering coefficient after standardized regression and the passive microwave bright temperature after standardized regression are obtained;
traversing all historical moments until the active microwave backward scattering coefficient after standardized regression processing and the passive microwave bright temperature after standardized regression processing corresponding to each moment in all moments are obtained;
the determining the active microwave fusion weight and the passive microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the bright temperature standardized regression coefficient comprises the following steps:
determining a normalized regression total coefficient according to the backscatter coefficient normalized regression coefficient and the bright temperature normalized regression coefficient;
determining the active microwave fusion weight according to the backscattering coefficient standardized regression coefficient and the standardized regression total coefficient;
and determining the passive microwave fusion weight according to the brightness temperature standardized regression coefficient and the standardized regression total coefficient.
2. The generalized linear regression model-based active and passive microwave soil moisture fusion algorithm of claim 1, wherein the obtaining current active and passive microwave soil moisture data includes:
transmitting a first indication instruction to an active microwave soil moisture product, and receiving current active microwave soil moisture data output by the active microwave soil moisture product;
and sending a second instruction to the passive microwave soil moisture product, and receiving current passive microwave soil moisture data output by the passive microwave soil moisture product.
3. The generalized linear regression model-based active and passive microwave soil moisture fusion algorithm of claim 1, wherein the inputting the current active microwave soil moisture data and the current passive microwave soil moisture data into the soil moisture data fusion model to obtain the active and passive microwave soil moisture fusion data output by the soil moisture data fusion model comprises:
inputting the current active microwave soil moisture data to a first output layer of the soil moisture data fusion model to obtain a first fusion value output by the first output layer according to the current active microwave soil moisture data and the active microwave fusion weight;
Inputting the current passive microwave soil moisture data to a first output layer of the soil moisture data fusion model to obtain a second fusion value output by the first output layer according to the current passive microwave soil moisture data and the passive microwave fusion weight;
and inputting the first fusion value and the second fusion value to a second output layer of the soil moisture data fusion model to obtain the active and passive microwave soil moisture fusion data output by the second output layer according to the first fusion value and the second fusion value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310343942.9A CN116049768B (en) | 2023-04-03 | 2023-04-03 | Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310343942.9A CN116049768B (en) | 2023-04-03 | 2023-04-03 | Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116049768A CN116049768A (en) | 2023-05-02 |
| CN116049768B true CN116049768B (en) | 2023-07-11 |
Family
ID=86113640
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310343942.9A Active CN116049768B (en) | 2023-04-03 | 2023-04-03 | Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116049768B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111610201B (en) * | 2020-04-30 | 2021-08-10 | 中国科学院空天信息创新研究院 | Passive microwave multi-channel synergistic soil moisture inversion method and device |
| US11592431B2 (en) * | 2021-04-20 | 2023-02-28 | Indigo Ag, Inc. | Addressing incomplete soil sample data in soil enrichment protocol projects |
| CN114140591B (en) * | 2021-11-29 | 2024-04-26 | 中国科学院南京土壤研究所 | Soil organic matter remote sensing mapping method combining machine learning and ground statistics |
| CN113888386B (en) * | 2021-12-03 | 2022-02-15 | 中国科学院、水利部成都山地灾害与环境研究所 | High-resolution time-space seamless earth surface soil moisture estimation method and system |
| CN115479960B (en) * | 2022-09-13 | 2023-06-13 | 安徽师范大学 | Method for monitoring soil humidity in crop growth process by combining SAR (synthetic aperture radar) with optical remote sensing data |
-
2023
- 2023-04-03 CN CN202310343942.9A patent/CN116049768B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN116049768A (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107765347B (en) | Short-term wind speed prediction method based on Gaussian process regression and particle filtering | |
| US8041597B2 (en) | Self-calibrating outlier model and adaptive cascade model for fraud detection | |
| CN112965146B (en) | Quantitative precipitation estimation method combining meteorological radar and rainfall barrel observation data | |
| CN114254561A (en) | Waterlogging prediction method, waterlogging prediction system and storage medium | |
| US20210048557A1 (en) | Method and device for simulating discharge, and computer device | |
| CN104573876A (en) | Wind power plant short-period wind speed prediction method based on time sequence long memory model | |
| CN109784528A (en) | Water quality prediction method and device based on time series and support vector regression | |
| CN104102836A (en) | Method for quickly estimating robust state of power system | |
| CN112819312A (en) | Method and system for evaluating drought socioeconomic exposure degree under climate change scene | |
| CN101883425A (en) | Target tracking and identification device and method based on entropy-weighted gray correlation | |
| CN113592143A (en) | Rainfall forecast-based medium and long term runoff ensemble forecasting method and system | |
| CN109816165B (en) | Wind power ultra-short term power prediction method and system | |
| CN116049768B (en) | Active and passive microwave soil moisture fusion algorithm based on generalized linear regression model | |
| CN111597692B (en) | Method, system, electronic equipment and storage medium for estimating surface net radiation | |
| JP2020171217A (en) | Method, apparatus and program for predicting yield | |
| CN115183884A (en) | Infrared temperature measurement compensation method and device of electric heating cooperative system | |
| JP2023174067A (en) | Methane emission amount estimation system, method and program | |
| CN114415157A (en) | Underwater target multi-model tracking method based on underwater acoustic sensor network | |
| CN110874662B (en) | Real-time verification method and system for water consumption in water supply pipe network model | |
| CN117895920B (en) | Distributed consistency Kalman filtering method for sensor network under communication link fault | |
| Gao et al. | Dynamic State Estimation for power system based on M-UKF algorithm | |
| CN114563045B (en) | Detection method and device for hospital logistics system, electronic equipment and medium | |
| CN113177702B (en) | Meteorological input data matching method and system | |
| CN110907977A (en) | Information processing method and device and computer storage medium | |
| CN114756805B (en) | Sea surface emissivity correction method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |