CN108509836B - Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation - Google Patents

Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation Download PDF

Info

Publication number
CN108509836B
CN108509836B CN201810084248.9A CN201810084248A CN108509836B CN 108509836 B CN108509836 B CN 108509836B CN 201810084248 A CN201810084248 A CN 201810084248A CN 108509836 B CN108509836 B CN 108509836B
Authority
CN
China
Prior art keywords
crop
lai
model
data
dual
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
Application number
CN201810084248.9A
Other languages
Chinese (zh)
Other versions
CN108509836A (en
Inventor
黄健熙
李俐
卓文
朱德海
张晓东
苏伟
刘峻明
刘哲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinzhinong Beijing Risk Management Technology Co ltd
Original Assignee
China Agricultural University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Agricultural University filed Critical China Agricultural University
Priority to CN201810084248.9A priority Critical patent/CN108509836B/en
Publication of CN108509836A publication Critical patent/CN108509836A/en
Application granted granted Critical
Publication of CN108509836B publication Critical patent/CN108509836B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G06F2218/16—Classification; Matching by matching signal segments
    • G06F2218/20—Classification; Matching by matching signal segments by applying autoregressive analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Image Processing (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention belongs to the field of agricultural remote sensing, and relates to a crop yield estimation method for assimilation of dual-polarized synthetic aperture radar and crop model data, which comprises the following specific steps: collecting satellite data of the dual-polarization synthetic aperture radar, carrying out polarization decomposition on the dual-polarization SAR data obtained by preprocessing, and selecting an LAI inversion model of the scattering component relation combination with the highest precision for inversion to obtain a remote sensing observation LAI; calibrating a WOFOST model LAI of the crops in the research area; assimilating the two LAIs by using a particle filter algorithm; and (4) adopting the optimized LAI track of the crop growth period to redrive the WOFOST model one by one, and carrying out spatial mapping. The method disclosed by the invention integrates the advantages of SAR remote sensing data and a crop model, fully utilizes rich information provided by multi-polarization SAR data, overcomes the problem of missing of optical remote sensing data in the key growth period of corn, improves the yield simulation of the crop model, optimizes the LAI track in the growth period of the crop with precision, and can estimate the crop yield on a regional scale.

Description

Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation
Technical Field
The invention belongs to the field of agricultural remote sensing, and particularly relates to a crop yield estimation method based on dual-polarization synthetic aperture radar and crop model data assimilation.
Background
The traditional crop yield estimation method mainly comprises a statistical investigation method, a forecasting method based on a crop model, an agricultural weather forecasting method and the like. These methods have inherent limitations that make it difficult to estimate regional crop yield with high accuracy. The estimation method based on the satellite remote sensing technology has the advantage of being unique in regional crop yield estimation by virtue of the characteristics of spatial continuity and temporal dynamics. Meanwhile, the remote sensing technology is combined with a crop growth model based on the mechanism processes of crop photosynthesis, respiration, transpiration, nutrition and the like, so that the aim of regional high-precision yield estimation can be achieved. The data assimilation method can combine the advantages of the crop growth model on point and remote sensing observation, and becomes a hotspot of research on agricultural quantitative remote sensing at home and abroad in recent years.
However, optical remote sensing is greatly restricted by weather factors, and radar remote sensing has the characteristic of being less affected by cloud and fog compared with optical remote sensing data, and can be monitored all day long, so that continuous and long-time observation data can be obtained in the growing season of crops, and the method is very helpful for monitoring and estimating the growth of crops. The data assimilation method is used for combining the radar remote sensing and the crop model, so that the defect of the crop model in the aspect of regional crop yield estimation can be overcome, the limitation that common optical remote sensing data are greatly limited by weather factors such as cloud and fog can be overcome, and the method can be suitable for crop yield estimation research in a regional range.
Disclosure of Invention
In order to solve the following problems in the prior art: the invention provides a crop yield estimation method for assimilating dual-polarized synthetic aperture radar and crop model data, and aims to 'how to assimilate remote sensing data of a large-range high-timeliness dual-polarized synthetic aperture radar and a mechanism model for accurately simulating crop growth so as to accurately estimate the yield in a large range'.
The invention provides a crop yield estimation method for assimilation of dual-polarized synthetic aperture radar and crop model data, which comprises the following specific steps:
s1, collecting satellite data of the dual-polarized synthetic aperture radar in the growth period of the crop to be detected in the research area, and preprocessing the satellite data to obtain VH and VV dual-polarized backscattering coefficients of a time sequence C waveband (C band), namely dual-polarized SAR data;
s2, carrying out polarization decomposition on the dual-polarization SAR data, and analyzing different scattering component characteristics and the correlation with the LAI value;
s3, selecting an LAI inversion model of the scattering component relation combination with the highest precision to invert the LAI to obtain a remote sensing observation LAI, and setting uncertainty of evaluation observation;
s4, collecting meteorological, crop, soil and crop management parameters in the research area and taking the meteorological, crop, soil and crop management parameters as input parameters, and calibrating a WOFOST (world food students) model of the crops in the research area to obtain a WOFOST simulation LAI;
S5, assimilating the remote sensing observation LAI and the WOFOST simulation LAI by using the LAI as an assimilation variable and utilizing a particle filter algorithm to obtain an optimized crop growth period LAI track;
and S6, operating the step S5 one by one crop grid, redriving the WOFOST model by adopting the optimized LAI track of the crop growth period, simulating the crop yield of an output area, performing space mapping and guiding the crop production.
In step S1, the satellite data of the dual-polarization synthetic aperture radar is preferably Sentinel 1 (Sentinel-1) satellite data, which is slc (single look complete) data of a Sentinel 1 (Sentinel-1) satellite, and the data is complex data including amplitude and phase information.
The preprocessing in step S1 refers to self-correction of orbit data, thermal noise removal, oblique ground conversion, single view complex data generation of polarization scattering matrix S, radiometric calibration, terrain correction, and speckle noise filtering.
The thermal noise is removed, and the influence of the thermal noise caused by the sensor is eliminated by a bilinear interpolation method.
Wherein, the speckle noise filtering is realized by Lee filtering.
In the dual-polarization SAR data described in step S1, a target vectorized scattering matrix is defined as:
Figure BDA0001561938940000031
where k is the target vector, SVV、SVHThe components represent the scattering information of the linear polarization states of the dual-polarization data VV, VH, respectively.
In the step S2, the dual-polarization SAR data is subjected to polarization decomposition, and in the polarization decomposition process, a covariance matrix is calculated:
Figure BDA0001561938940000032
natural ground objects generally maintain the symmetry of the flight direction, so that it can be assumed that the homopolar and cross-polar components are uncorrelated, as follows:
Figure BDA0001561938940000033
then
Figure BDA0001561938940000034
Where < … > represents the spatial statistical average of the case where the random scattering medium is assumed to be isotropic. T denotes a transposed matrix. | … | represents the complex magnitude.
By using the method of Freeman-Durden decomposition for reference, the polarization decomposition of dual-polarization SAR data can decompose a covariance matrix or a coherent matrix into two components: bulk scattering components from a series of vegetation canopy oriented dipoles; single and second scattering components resulting from first order Bragg (Bragg) surface scattering and dihedral reflections.
Step S2, performing polarization decomposition on the dual-polarization SAR data, where the decomposition process is as follows:
Figure BDA0001561938940000035
in the formula (2) fvCorresponding to the volume scatter component (corresponding to the volume scatter contribution above), fs+dComponents corresponding to the combined effect of odd and even scattering (corresponding to the single and second scattering components obtained by the above first-order bragg surface scattering and dihedral angle reflection);
from this, the power of various components can be deduced:
Ps+d＝fs+d (3)
PV＝4fv/3 (4)
w＝Ps+d+PV＝|SHV|2+|SVV|2 (5)
In the formula, parameter Ps+dPower, parameter P, representing the component of the co-action of odd and even scatterVRepresenting the power of the bulk scattering component, and the parameter w is the total scattering power of the two scattering components.
Step S3 is to select the LAI inversion model of the scattering component relationship combination with the highest precision for LAI inversion, and first select the polarization indexes respectively
Figure BDA0001561938940000041
PV/W、Ps+dSensitivity of/W to changes in LAI values, and then using the measured data to establish LAI and PV/w、Ps+dMultivariate regression model between/w:
LAI＝f(PV/w,Ps+d/W) (6)
thereby realizing the LAI inversion of the regional crops.
In the step S4, the WOFOST model of the crop in the research area is calibrated, and the regional parameter calibration is completed by using an Inverse Distance Weight (IDW) interpolation algorithm with reference to the weather station as a reference for the weather parameters and the accumulated temperature parameters required by the crop model.
In step S5, the remote sensing observation LAI and the wocost simulation LAI are assimilated by a particle filter algorithm, and are calculated by equations (7), (8) and (9):
Figure BDA0001561938940000042
Figure BDA0001561938940000043
Figure BDA0001561938940000044
Figure BDA0001561938940000045
a model simulation state variable representing the ith particle at time k;
Figure BDA0001561938940000046
a model simulation state variable representing the ith particle at time k + 1; m is a nonlinear operator, namely a WOFOST model; u. ofkIs a model driving parameter;
Figure BDA0001561938940000047
represents an observed state variable of the ith particle at the moment k + 1; h is an observation operator, and epsilon is observation noise; x is the number of k+1The optimal estimated value at the k +1 moment is obtained;
Figure BDA0001561938940000048
the weight of each particle after the normalization resampling is carried out; n represents the number of particles;
Figure BDA0001561938940000051
Figure BDA0001561938940000052
Figure BDA0001561938940000053
Figure BDA0001561938940000054
in the formula (I), the compound is shown in the specification,
Figure BDA0001561938940000055
representing a particle importance weight;
Figure BDA0001561938940000056
represents the observed value as yk+1Likelihood probability density of time, i.e.
Figure BDA0001561938940000057
In the case of yk+1The probability of occurrence; in the same way, the method for preparing the composite material,
Figure BDA0001561938940000058
to represent
Figure BDA0001561938940000059
In the event of occurrence of
Figure BDA00015619389400000510
The probability of occurrence;
Figure BDA00015619389400000511
a sampling function representing importance; k is a radical ofiIs a resampling coefficient;
Figure BDA00015619389400000512
express get
Figure BDA00015619389400000513
The integer part of (2).
Wherein the crop is preferably corn.
The invention also provides application of the bipolar synthetic aperture radar and the crop yield estimation method for crop model data assimilation in guiding crop production.
Compared with the prior art, the invention has the beneficial effects that:
the method disclosed by the invention integrates the advantages of SAR remote sensing data and a crop model, fully utilizes rich information provided by multi-polarization SAR data, adopts a dual-polarization signal decomposition mode to invert LAI, and utilizes a particle filter assimilation LAI to a WOFOST model, so that the problem of lacking of optical remote sensing data in the growth period of corn is solved, the crop yield simulation precision of the crop model can be improved, the LAI track in the growth period of the crop can be optimized, and the crop yield can be estimated on an area scale.
Drawings
FIG. 1 is a schematic flow chart of a method for crop estimation of yield by applying dual polarized synthetic aperture radar and crop model data assimilation to corn according to example 1 of the present invention;
FIG. 2 is a graph of the results of corn yield estimation from the crop estimation method of example 1 using the assimilation of polarized synthetic aperture radar and crop model data.
Detailed Description
The following describes in further detail specific embodiments of the present invention with reference to examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Example 1
The schematic flow chart of the crop yield estimation method for assimilating the dual-polarized synthetic aperture radar and the crop model data according to the invention for the corn is shown in the attached figure 1.
S1, collecting Sentinel-1 data in the corn growth period of a research area, and carrying out pretreatment such as terrain correction to obtain backscattering coefficients of two polarizations (VH and VV) of C band of a time sequence, namely dual-polarization SAR data;
the city of Hebei province Hebei Heshui was selected as the study area between the east longitude 115 ° 10 '-116 ° 34' and the north latitude 37 ° 03 '-38 ° 23'. The total area of the research area 8815km2, the terrain is mainly plain, the cultivated land occupies more than 60% of the total area, and belongs to the warm zone semi-humid monsoon climate, the annual sunshine hours 2400-.
Selecting time sequence Sentinel-1 satellite remote sensing data of summer corn in Hebei province, Hebei province and Heshui city in 2017 in the key growth period of 6 months to 10 months. At the same time, the image is preprocessed to obtain the backscattering coefficients of two polarizations (VH, VV) of the C band in time series.
The main preprocessing steps further comprise self-correction of orbit data, removal of thermal noise, inclined ground conversion, generation of a polarization scattering matrix S from single-view complex data, radiometric calibration, terrain correction and speckle noise filtering. The thermal noise is removed, and the influence of the thermal noise caused by the sensor is eliminated by a bilinear interpolation method. And filtering coherent speckle noise by adopting Lee filtering. For dual-polarization SAR, a target vectorized scattering matrix is defined as:
Figure BDA0001561938940000061
where k is the target vector, SVV,SVHThe components represent the scattering information of the linear polarization states of the dual-polarization data VV, VH, respectively.
S2, polarization decomposition of dual-polarized backscattering coefficients of VH and VV in the growth period (namely polarization decomposition of dual-polarized SAR data) is carried out, and characteristics of different scattering components and correlation relation with LAI values are analyzed;
the polarization decomposition process first calculates the covariance matrix:
Figure BDA0001561938940000062
Figure BDA0001561938940000063
natural ground objects generally maintain the symmetry of the flight direction, so that it can be assumed that the homopolar and cross-polar components are uncorrelated, as follows:
Figure BDA0001561938940000064
Figure BDA0001561938940000071
Then
Figure BDA0001561938940000072
Where < … > represents the spatial statistical average of the case where the random scattering medium is assumed to be isotropic. T denotes a transposed matrix. | … | represents the complex magnitude.
By using the method of Freeman-Durden decomposition for reference, the polarization decomposition of dual-polarization SAR data can decompose a covariance matrix or a coherent matrix into two components: bulk scattering components from a series of vegetation canopy oriented dipoles; single and second scattering components resulting from first order Bragg surface scattering and dihedral reflections. The decomposition process is as follows:
Figure BDA0001561938940000073
in the formula fvCorresponding to the bulk scattering component, fs+dCorresponding to the component of the odd and even scatter contributions.
From this, the power of various components can be deduced:
Ps+d＝fs+d (3)
PV＝4fv/3 (4)
w＝Ps+d+PV＝|SHV|2+|SVV|2 (5)
parameter Ps+dPower, parameter P, representing the component of the co-action of odd and even scatterVRepresenting the power of the bulk scattering component, and the parameter w is the total scattering power of the two scattering components.
S3, selecting an LAI inversion model of the scattering component relation combination with the highest precision, inverting the LAI, and giving uncertainty of estimation observation;
firstly, selecting polarization indexes respectively
Figure BDA0001561938940000074
PV/w、Ps+dSensitivity of/w to changes in LAI values, and then using the measured data to establish LAI and PV/w、Ps+dW, multiple regression model between:
LAI＝f(PV/w,Ps+d/w) (6)
Thereby achieving LAI inversion.
S4 collecting weather, crop, soil and crop management parameters in the research area as input parameters, calibrating the WOFOST model of the corn in the research area
The following data were obtained: selecting data of 6 meteorological elements such as the highest air temperature, the lowest air temperature, the sunshine hours, the water vapor pressure, the wind speed, the precipitation and the like of 21 national-level meteorological observation stations 2017 day by day according to the outer envelope range of the research area; acquiring collected soil parameters, crop parameters and phenological data from agricultural meteorological sites in a research area; acquiring control parameters such as longitude and latitude, elevation and the like; agricultural meteorological data and summer corn yield data of each county of Hebei province, Hebei province and the balance Water City of 17 years.
The method comprises the steps of obtaining a summer corn planting area by utilizing long-time sequence images and field survey data of a research area through interpretation, generating a 1-kilometer grid, calculating the planting percentage of summer corn in grid units, and setting a threshold value to remove units with the planting proportion of the summer corn lower than 20%. And meanwhile, calibrating the WOFOST model by using the collected weather, crops, soil and management parameters at each agricultural gas station. And uniformly measuring the output of the model and meteorological data, and performing inverse distance weight interpolation on the meteorological data and crop parameters according to the result of primary calibration to generate raster data of each pixel of 1 kilometer.
S5, assimilating the remote sensing observation LAI and the WOFOST simulation LAI by using the LAI as an assimilation variable and utilizing a particle filter algorithm, and optimizing an LAI track in a corn growth period;
in the running process of the model, if remote sensing observation data exist in the same day, assimilation is carried out. The assimilation method is represented as follows:
Figure BDA0001561938940000081
Figure BDA0001561938940000082
Figure BDA0001561938940000083
Figure BDA0001561938940000084
model simulation state change of ith particle representing k timeAn amount;
Figure BDA0001561938940000085
a model simulation state variable representing the ith particle at time k + 1; m is a nonlinear operator, namely a WOFOST model; u. ofkIs a model driving parameter;
Figure BDA0001561938940000086
represents an observed state variable of the ith particle at the moment k + 1; h is an observation operator, and epsilon is observation noise; x is the number ofk+1The optimal estimated value at the k +1 moment is obtained;
Figure BDA0001561938940000091
the weight of each particle after the normalization resampling is carried out; n represents the number of particles;
Figure BDA0001561938940000092
Figure BDA0001561938940000093
Figure BDA0001561938940000094
Figure BDA0001561938940000095
in the formula (I), the compound is shown in the specification,
Figure BDA0001561938940000096
representing a particle importance weight;
Figure BDA0001561938940000097
represents the observed value as yk+1Likelihood probability density of time, i.e.
Figure BDA0001561938940000098
In the case of yk+1Take place ofProbability; in the same way, the method for preparing the composite material,
Figure BDA0001561938940000099
to represent
Figure BDA00015619389400000910
In the event of occurrence of
Figure BDA00015619389400000911
The probability of occurrence;
Figure BDA00015619389400000912
a sampling function representing importance; k is a radical ofiIs a resampling coefficient;
Figure BDA00015619389400000913
express get
Figure BDA00015619389400000914
The integer part of (2).
The basic process of assimilation is that the number of particles is set to 300, and in the prediction stage, the initial LAI x of the model is perturbed by the given Gaussian distribution random noise kObtaining an initial particle swarm at the kth time
Figure BDA00015619389400000915
In the update phase, the importance weight of each particle is calculated by equation (14) using the model predicted state, the remote sensing observations, and the initial importance probability density
Figure BDA00015619389400000916
After each re-sampling, the data is re-sampled,
Figure BDA00015619389400000917
specifically selecting the initial importance probability density function of the particles, namely, assuming that the condition of the predicted particles deviating from the observed value conforms to normal distribution, calculating the weight value of the predicted particles by using the probability density function of the normal distribution:
Figure BDA00015619389400000918
in the formula, Rk+1Is the covariance matrix of the observed errors at time k + 1. Then, the state estimation value at the k +1 th time is obtained by equation (13). Thereby optimizing the LAI trajectory during the corn growth period.
And S6, operating the step S5 one corn grid by one, redriving the WOFOST model by adopting the optimized LAI, simulating the corn yield of an output area, and performing spatial mapping.
The results of the corn yield estimated by the crop estimation method using the bipolar synthetic aperture radar and the crop model data assimilation in this example are shown in fig. 2.
The method disclosed by the invention integrates the advantages of SAR remote sensing data and a crop model, fully utilizes rich information provided by multi-polarization SAR data, adopts a dual-polarization signal decomposition mode to invert LAI, and utilizes a particle filter assimilation LAI to a WOFOST model, so that the problem of lacking of optical remote sensing data in the growth period of corn is solved, the crop yield simulation precision of the crop model can be improved, the LAI track in the growth period of the crop can be optimized, and the crop yield can be estimated on an area scale.
Although the invention has been described in detail hereinabove with respect to a general description and specific embodiments thereof, it will be apparent to those skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.

Claims (7)

1. A crop yield estimation method for assimilation of dual-polarized synthetic aperture radar and crop model data is characterized by comprising the following specific steps:
s1, collecting satellite data of the dual-polarized synthetic aperture radar in the growth period of the crop to be detected in the research area, and preprocessing the satellite data to obtain VH and VV dual-polarized backscattering coefficients of a time sequence C wave band, namely dual-polarized SAR data;
s2, carrying out polarization decomposition on the dual-polarization SAR data, and analyzing different scattering component characteristics and the correlation with the LAI value;
s3, selecting an LAI inversion model of the scattering component relation combination with the highest precision to invert the LAI to obtain a remote sensing observation LAI, and setting uncertainty of evaluation observation;
the specific method for selecting the LAI inversion model of the scattering component relation combination with the highest precision to invert the LAI is as follows: firstly, selecting polarization indexes respectively
Figure FDA0003162430650000011
PV/w、Ps+dSensitivity of/w to changes in LAI values, and then using the measured data to establish LAI and PV/w、Ps+dMultivariate regression model between/w:
LAI＝f(PV/w,Ps+d/w) (6)
thereby realizing LAI inversion of regional crops; parameter Ps+dPower, parameter P, representing the component of the co-action of odd and even scatterVThe power of the volume scattering component is represented, and the parameter w is the total scattering power of the two scattering components;
s4, collecting meteorological, crop, soil and crop management parameters in the research area and taking the meteorological, crop, soil and crop management parameters as input parameters, and calibrating a WOFOST model of the crops in the research area to obtain a WOFOST simulation LAI; the WOFOST model for calibrating the crops in the research area needs to complete regional parameter calibration by taking a meteorological site as a reference and adopting an inverse distance weight interpolation algorithm for meteorological parameters and accumulated temperature parameters needed by the crop model;
s5, assimilating the remote sensing observation LAI and the WOFOST simulation LAI by using the LAI as an assimilation variable and utilizing a particle filter algorithm to obtain an optimized crop growth period LAI track;
assimilating the remotely sensed observation LAI and the WOFOST simulated LAI using a particle filtering algorithm, and calculating with equations (7), (8) and (9):
Figure FDA0003162430650000021
Figure FDA0003162430650000022
Figure FDA0003162430650000023
Figure FDA0003162430650000024
a model simulation state variable representing the ith particle at time k;
Figure FDA0003162430650000025
a model simulation state variable representing the ith particle at time k + 1; m is a nonlinear operator, namely a WOFOST model; u. of kIs a model driving parameter;
Figure FDA0003162430650000026
represents an observed state variable of the ith particle at the moment k + 1; h is an observation operator, and epsilon is observation noise; x is the number ofk+1The optimal estimated value at the k +1 moment is obtained;
Figure FDA0003162430650000027
the weight of each particle after the normalization resampling is carried out; n represents the number of particles;
Figure FDA0003162430650000028
Figure FDA0003162430650000029
Figure FDA00031624306500000210
Figure FDA00031624306500000211
in the formula (I), the compound is shown in the specification,
Figure FDA00031624306500000212
representing a particle importance weight;
Figure FDA00031624306500000213
represents the observed value as yk+1Likelihood probability density of time, i.e.
Figure FDA00031624306500000214
In the case of yk+1The probability of occurrence; in the same way, the method for preparing the composite material,
Figure FDA00031624306500000215
to represent
Figure FDA00031624306500000216
In the event of occurrence of
Figure FDA00031624306500000217
The probability of occurrence;
Figure FDA00031624306500000218
a sampling function representing importance; k is a radical ofiIs a resampling coefficient;
Figure FDA00031624306500000219
express get
Figure FDA00031624306500000220
The integer part of (1);
and S6, operating the step S5 one by one crop grid, redriving the WOFOST model by adopting the optimized LAI track of the crop growth period, simulating the crop yield of an output area, performing space mapping and guiding the crop production.
2. The method of claim 1, wherein the preprocessing in step S1 includes self-correction, thermal noise removal, oblique earth transformation, single view complex data generation of polarization scattering matrix S, radiometric calibration, terrain correction, and speckle noise filtering.
3. The method of claim 2, wherein removing thermal noise uses bilinear interpolation to remove thermal noise effects from the sensor.
4. The method of claim 2, wherein speckle noise filtering is performed using Lee filtering.
5. The method of claim 2, wherein the dual polarized SAR data of step S1, defines a target vectorized scattering matrix as:
Figure FDA0003162430650000031
where k' is the target vector, SVV、SVHThe components represent the scattering information of the linear polarization states of the dual-polarization data VV, VH, respectively.
6. The method of claim 5, wherein the step S2 is implemented by performing polarization decomposition on dual-polarized SAR data, the decomposition process is as follows:
Figure FDA0003162430650000032
in the formula (2) fvCorresponding to the bulk scattering component, fs+dComponents corresponding to the combined effect of odd and even scatter;
from this, the power of various components can be deduced:
Ps+d＝fs+d (3)
PV＝4fv/3 (4)
w＝Ps+d+PV＝|SHV|2+|SVV|2 (5)
in the formula, parameter Ps+dPower, parameter P, representing the component of the co-action of odd and even scatterVRepresenting the power of the bulk scattering component, and the parameter w is the total scattering power of the two scattering components.
7. Use of a method of crop assessment by assimilation of bipolar synthetic aperture radar with crop model data as claimed in any one of claims 1 to 6 for directing crop production.
CN201810084248.9A 2018-01-29 2018-01-29 Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation Active CN108509836B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810084248.9A CN108509836B (en) 2018-01-29 2018-01-29 Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810084248.9A CN108509836B (en) 2018-01-29 2018-01-29 Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation

Publications (2)

Publication Number Publication Date
CN108509836A CN108509836A (en) 2018-09-07
CN108509836B true CN108509836B (en) 2021-10-08

Family

ID=63374526

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810084248.9A Active CN108509836B (en) 2018-01-29 2018-01-29 Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation

Country Status (1)

Country Link
CN (1) CN108509836B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109344737A (en) * 2018-09-12 2019-02-15 中国农业大学 One-dimensional DCNN land use classes method based on SAR polarization decomposing
CN109800921B (en) * 2019-01-30 2019-12-20 北京师范大学 Regional winter wheat yield estimation method based on remote sensing phenological assimilation and particle swarm optimization
CN110287457B (en) * 2019-07-02 2023-02-17 吉林大学 Corn biomass inversion measurement method based on satellite radar remote sensing data
CN110633841B (en) * 2019-08-13 2022-04-01 中国农业大学 Provincial range plot scale data assimilation yield prediction method based on set sampling
CN110532967A (en) * 2019-09-02 2019-12-03 中国科学院遥感与数字地球研究所 A kind of Crop classification method based on No. 1 RVI time series of sentry
CN111368736B (en) * 2020-03-04 2020-12-01 北京四象爱数科技有限公司 Rice refined estimation method based on SAR and optical remote sensing data
CN112052988B (en) * 2020-08-18 2024-03-22 中国农业大学 Crop yield estimation method coupling multi-objective optimization and collection assimilation and application
CN112991247B (en) * 2021-03-04 2022-06-03 河南省气象科学研究所 Winter wheat evapotranspiration remote sensing inversion and crop model assimilation method
CN113129328B (en) * 2021-04-22 2022-05-17 中国电子科技集团公司第二十九研究所 Target hotspot area fine analysis method
CN113435566B (en) * 2021-06-22 2022-04-12 布瑞克农业大数据科技集团有限公司 Method, system and device for fertilizing farm crops and storage medium
CN114282609B (en) * 2021-12-23 2022-09-20 西南交通大学 Method for extracting indexes of area and phenology of crops

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001188084A (en) * 1999-09-23 2001-07-10 Astrium Ltd Radar for space-born use
CN102968640A (en) * 2012-10-25 2013-03-13 西安电子科技大学 Polarized synthetic aperture radar (SAR) image classification method based on Freeman decomposition and data distribution characteristics
CN103323846A (en) * 2013-05-15 2013-09-25 中国科学院电子学研究所 Inversion method based on polarization interference synthetic aperture radar and device
CN104330798A (en) * 2014-11-03 2015-02-04 北京农业信息技术研究中心 Synthetic aperture radar based crop seeding date monitoring method and device through remote sensing image
CN106258686A (en) * 2016-08-11 2017-01-04 中国科学院遥感与数字地球研究所 The water-cloud model of a kind of improvement and apply the rice parameters retrieval method of this model

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2617119A1 (en) * 2008-01-08 2009-07-08 Pci Geomatics Enterprises Inc. Service oriented architecture for earth observation image processing
CN104134095B (en) * 2014-04-17 2017-02-15 中国农业大学 Crop yield estimation method based on scale transformation and data assimilation

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001188084A (en) * 1999-09-23 2001-07-10 Astrium Ltd Radar for space-born use
CN102968640A (en) * 2012-10-25 2013-03-13 西安电子科技大学 Polarized synthetic aperture radar (SAR) image classification method based on Freeman decomposition and data distribution characteristics
CN103323846A (en) * 2013-05-15 2013-09-25 中国科学院电子学研究所 Inversion method based on polarization interference synthetic aperture radar and device
CN104330798A (en) * 2014-11-03 2015-02-04 北京农业信息技术研究中心 Synthetic aperture radar based crop seeding date monitoring method and device through remote sensing image
CN106258686A (en) * 2016-08-11 2017-01-04 中国科学院遥感与数字地球研究所 The water-cloud model of a kind of improvement and apply the rice parameters retrieval method of this model

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Regional winter wheat yield prediction by integrating MODIS LAI into the WOFOST model with sequential assimilation techniqu;Liu J等;《Journal of Food, Agriculture & Environment》;20140131;第12卷(第1期);第182页左栏第2段、右栏第2-3段，第181页右栏第4段、左栏第三段，第183页右栏最后1段，图4，第184页左栏第1段，图6 *
基于改进的SVM方法对极化雷达数据估算小麦叶面积指数;王舒;《中国优秀硕士学位论文全文数据库信息科技辑》;20170315;第4-7，38-43页 *
粒子滤波算法在数据同化中的应用研究进展;毕海芸等;《遥感技术与应用》;20141031;第29卷(第5期);第702页左栏第2段 *

Also Published As

Publication number Publication date
CN108509836A (en) 2018-09-07

Similar Documents

Publication Publication Date Title
CN108509836B (en) Crop yield estimation method based on double-polarized synthetic aperture radar and crop model data assimilation
CN109829234B (en) A kind of across scale Dynamic High-accuracy crop condition monitoring and yield estimation method based on high-definition remote sensing data and crop modeling
CN106908415B (en) A kind of big region crops time of infertility Soil Moisture Monitoring method based on amendment NDVI time series
CN108802728B (en) Crop irrigation guiding method for assimilation of bipolar synthetic aperture radar and crop model
Børgesen et al. A probabilistic assessment of climate change impacts on yield and nitrogen leaching from winter wheat in Denmark
CN103955860A (en) Regional crop yield estimation method based on ensemble Kalman filter assimilation
Wu et al. Assimilating MODIS-LAI into crop growth model with EnKF to predict regional crop yield
Yang et al. In-season biomass estimation of oilseed rape (Brassica napus L.) using fully polarimetric SAR imagery
CN105321120A (en) Assimilation evapotranspiration and LAI (leaf area index) region soil moisture monitoring method
CN115479960B (en) Method for monitoring soil humidity in crop growth process by combining SAR (synthetic aperture radar) with optical remote sensing data
CN108537679B (en) Remote sensing and crop model fused region scale crop emergence date estimation method
Cheng et al. Cold damage risk assessment of double cropping rice in Hunan, China
CN110599360A (en) High-resolution remote sensing estimation method for evapotranspiration of crops in arid region
Shi et al. Estimation of sugarcane yield by assimilating UAV and ground measurements via ensemble Kalman filter
Lubis et al. Physical condition of the ocean to global climate change variability: case study in the Batam waters, Indonesia
Chaudhari et al. Spatial wheat yield prediction using crop simulation model, GIS, remote sensing and ground observed data
Tian et al. Comparison of two optimization algorithms for estimating regional winter wheat yield by integrating MODIS leaf area index and world food studies model
Lykhovyd Study of climate impact on vegetation cover in Kherson Oblast (Ukraine) using normalized difference and enhanced vegetation indices
Mattia et al. Time series of COSMO-SkyMed data for landcover classification and surface parameter retrieval over agricultural sites
Zhang et al. Enhanced Feature Extraction From Assimilated VTCI and LAI With a Particle Filter for Wheat Yield Estimation Using Cross-Wavelet Transform
CN111175784A (en) Satellite remote sensing monitoring method for cotton canopy moisture content
CN107316017B (en) Non-point source pollution source intensity estimation method based on crop information remote sensing accurate inversion
Zhuo et al. Assimilating SAR and optical remote sensing data into WOFOST model for improving winter wheat yield estimation
Yang et al. Biomass estimation of oilseed rape using simulated compact polarimtric SAR imagery
Oleson et al. Evaluating land surface parameters in the Biosphere‐Atmosphere Transfer Scheme using remotely sensed data sets

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
TR01 Transfer of patent right

Effective date of registration: 20220505

Address after: 100005 room 07-162, 101, floor 7, No. 219, Wangfujing Street, Dongcheng District, Beijing

Patentee after: Jinzhinong (Beijing) Risk Management Technology Co.,Ltd.

Address before: 100193 No. 2 Old Summer Palace West Road, Beijing, Haidian District

Patentee before: CHINA AGRICULTURAL University

TR01 Transfer of patent right