Remote sensing surface temperature angle effect correction method based on emissivity kernel driving model

Through the method based on the emissivity nuclear drive model, the direction emissivity model and the GSW algorithm model are constructed, combined with atmospheric profile data and observation zenith angle and other parameters, the problem of low surface temperature inversion accuracy in satellite remote sensing technology is solved, and the problem of low surface temperature inversion accuracy is realized. Higher precision data acquisition.

CN118916585AActive Publication Date: 2024-11-08CHINA UNIV OF MINING & TECH (BEIJING) +1

Patent Information

Application Number
CN202411012125.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-08
Estimated Expiration
2044-07-26

AI Technical Summary

Technical Problem

When obtaining surface temperature data, existing satellite remote sensing technology ignores the angular effect of emissivity, resulting in a large difference in the accuracy of surface temperature inversion from the truth value, and high-precision monitoring of surface temperature data in large areas or large areas cannot be achieved.

Method used

Using a method based on emissivity nuclear drive model, the direction emissivity model and GSW algorithm model are constructed, combined with atmospheric profile data and observed zenith angle parameters, the direction emissivity and split window coefficient are calculated, and then the inversion is obtained is obtained from the research area. The surface temperature of the pixel by cell.

Benefits of technology

The accuracy of surface temperature inversion is improved, the uncertainty caused by the angle effect is reduced, and data acquisition is achieved with higher accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118916585A_ABST
    Figure CN118916585A_ABST
Patent Text Reader

Abstract

The invention discloses a remote sensing surface temperature angle effect correction method based on an emissivity kernel-driven model, and the method comprises the steps: S1, building a direction emissivity model based on a parameter kernel-driven model, collecting the data of a research region, inputting the data into the direction emissivity model, and carrying out the calculation to obtain the direction emissivity; s2, collecting atmosphere profile data to obtain channel atmosphere parameters of two channels of a wave band 31 and a wave band 32; grouping by adopting a grouping scheme and obtaining the on-satellite brightness temperature of the channel; s3, constructing a GSW algorithm model, and constructing a coefficient lookup table under various grouping conditions; and S4, in combination with the directional emissivity, the observation zenith angle, the atmospheric water vapor content and the brightness temperature, traversing from the coefficient lookup table to extract split window coefficients under the corresponding groups, and performing inversion calculation by using a GSW algorithm model to obtain the surface temperature of each pixel of the research area. According to the method, the directional emissivity is innovatively obtained, the coefficient lookup table under the grouping condition is constructed, the land surface temperature LST is obtained pixel by pixel, and the precision of land surface temperature LST data products is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of surface temperature inversion, in particular to a remote sensing surface temperature angle effect correction method based on an emissivity kernel driven model, and in particular to the technical field of quantitative remote sensing. Background Art

[0002] At present, land surface temperature (LST) is the key driving force of energy balance and water cycle in the interaction between the land surface and the atmosphere. There are generally two ways to obtain land surface temperature data. The first is to set up an observation station on the ground (it is expensive to set up an observation station, which is not conducive to widespread deployment) and obtain it through the station data measured by the observation station. This method is only applicable to the acquisition of land surface temperature data in the area where the observation station is located. The application surface is very limited and cannot be applied to land surface temperature data monitoring over a large area or region. The second is to simulate based on satellite remote sensing technology (sensor products associated with satellites, etc.). LST is an indispensable input element in the study of climate change, environmental monitoring and evapotranspiration estimation; however, when thermal infrared sensors observe ground objects, the radiation observed from different observation angles is different. This phenomenon is called thermal radiation directivity (TRD); thermal radiation directivity (TRD) will cause the accuracy of the inverted land surface temperature (LST) to be greatly different from the true value; studies have found that the difference in LST observed at different observation angles can reach 6-12K. The sensor products of the associated satellite use the observed zenith angle VZA and the emissivity based on classification (such as feature classification: in general, there are 8 representative features: evergreen coniferous forest (ENF), deciduous broad-leaved forest (DBF), open shrub forest (OS), woody savanna (WS), grassland (GL), cultivated land (CL), urban and built-up land (UBL), and wasteland (BR)). The observed zenith angle VZA and classification also have mutual influence (for example, the angular effect of emissivity is ignored). For example: emissivity is a directional quantity, which is related to the observed zenith angle (VZA). At the same time, the natural surface is highly heterogeneous, and its components, roughness, structure and spatial heterogeneity lead to anisotropy in emissivity. For non-isothermal heterogeneous pixels, the directionality of thermal radiation has two reasons: 1) components with different temperatures and emissivities change with the change of observation angle; 2) multiple scattering within different components. For isothermal pixels, the anisotropy of thermal radiation comes from the anisotropy of the ground surface emissivity (LSE). Ignoring the anisotropy of emissivity will bring greater uncertainty to the subsequent surface temperature inversion, so the angle effect of emissivity cannot be ignored. Therefore, the second method of obtaining surface temperature data has many technical difficulties (such as angle effect), and there will still be unsatisfactory accuracy. The acquisition of higher-precision surface temperature data is a current technical problem and a key direction of technical research and development in this field. Summary of the invention

[0003] The purpose of the present invention is to provide a remote sensing surface temperature angle effect correction method based on an emissivity kernel driven model. A directional emissivity model is constructed based on a parameter kernel driven model. A hot spot kernel or / and a basic shape kernel are used to establish a directional emissivity model for study area data, and the directional emissivity is calculated. At the same time, channel atmospheric parameters of two channels, band 31 and band 32, are obtained according to atmospheric profile data. The observed zenith angle, the atmospheric bottom layer boundary temperature, the average emissivity, and the emissivity difference are grouped and the channel satellite brightness temperature is calculated. Then, a GSW algorithm model is constructed to obtain a coefficient lookup table under multiple grouping conditions; then, the directional emissivity, the observed zenith angle, the atmospheric water vapor content, and the brightness temperature are combined and the split window coefficients under the corresponding group are traversed and extracted from the coefficient lookup table, and the split window coefficients under the corresponding group are input into the GSW algorithm model to invert the surface temperature LST of each pixel in the study area, thereby improving the inversion accuracy.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A remote sensing surface temperature angle effect correction method based on an emissivity kernel driven model, the method comprising:

[0006] S1. A directional emissivity model is constructed based on a parameter kernel driven model. The parameter kernel driven model includes a hot spot kernel and a basic shape kernel. Data of the study area including MYD21, MYD03, MYD10_L2, MYD021km, MYD11_L2, and MCD12Q1 products are collected (satellite thermal infrared data at each observation zenith angle VZA within 16 days are used as multi-angle observation data) and input into the parameter kernel driven model. The directional emissivity model is calculated according to the following formula:

[0007] Directional emissivity = P + hot spot kernel coefficient × hot spot kernel data + basic shape kernel coefficient × basic shape kernel data. Where P is the isotropy coefficient;

[0008] S2, collect atmospheric profile data and input them into MODTRAN 5.2 software to simulate atmospheric parameters and convolve them with the spectral response function of the channel to obtain the channel atmospheric parameters of band 31 and band 32, the channel atmospheric parameters include the effective transmittance of the atmosphere, the atmospheric downward radiation and the atmospheric upward radiation when the observation zenith angle is θ; group the observation zenith angle, the boundary temperature of the bottom layer of the atmosphere, the average emissivity and the emissivity difference, and then perform Planck inverse operation to obtain the channel on-board brightness temperature under each group;

[0009] S3, constructing a GSW algorithm model, inputting the bottom boundary temperature of the atmosphere, the brightness temperature on the channel satellite, the average emissivity, and the emissivity difference into the GSW algorithm model to obtain multiple sets of equations according to the following algorithm formula, performing multiple linear regression on them, and obtaining the split window coefficients a0 to a7 of the general split window algorithm, and then constructing a coefficient lookup table under various grouping conditions;

[0010]

[0011] Among them, T s represents the bottom boundary temperature of the atmosphere, T 31 and T 32 represents the on-board brightness temperature of the channel of bands 31 and 32, respectively; Δε represents the difference in emissivity between bands 31 and 32; ε represents the average emissivity of bands 31 and 32; a0 to a7 are the split window coefficients in the coefficient lookup table;

[0012] S4, extract the observation zenith angle, atmospheric water vapor content, brightness temperature from the study area data and combine them with the directional emissivity obtained in step S1, traverse and extract the split window coefficient under the corresponding group from the coefficient lookup table, and then input the directional emissivity, split window coefficient and brightness temperature into the GSW algorithm model, and obtain the surface temperature LST of each pixel in the study area according to the following inversion formula:

[0013] Among them, ε p represents the average emissivity of bands 31 and 32, Δε p represents the difference in emissivity between bands 31 and 32, T p31 and T p32 They represent the brightness temperatures of bands 31 and 32 respectively, and a0 to a7 are the split window coefficients obtained from the coefficient lookup table.

[0014] In order to better implement the present invention, the present invention also includes the following method:

[0015] S5. Collect data in the study area including SURFRAD station data and / or ERA5_LAND data and obtain the true value of LST, and construct a loss constraint function in the GSW algorithm model to constrain the loss of the surface temperature LST and the true value of LST.

[0016] Preferably, in step S5, the method of obtaining the LST true value from the SURFRAD station data of the present invention is as follows: the SURFRAD station data is uniformly converted to the Coordinated Universal Time, and then the LST true value is obtained according to the following formula:

[0017] Among them LST s真 represents the true value of LST, R ↑ and R ↓ denote the upward and downward radiation from the surface, σ SB represents the Stefan Boltzmann constant, ε bb Represents the broadband emissivity; ε bb The calculation method is as follows: bb =0.095+0.329*ε 29 +0.572*ε 31, where ε 29 and ε 31 They respectively represent the narrow-band emissivity of band 29 and band 31 obtained from the MYD21 product.

[0018] Preferably, the parameter kernel driven model constructs several parameter kernel driven units according to the hot spot kernel and / or the basic shape kernel, selects the parameter kernel driven unit with the smallest root mean square error (RMSE) to establish the directional emissivity model, inputs the study area data, and calculates the directional emissivity.

[0019] Preferably, the hotspot nuclear data of the present invention is obtained according to the following method:

[0020]

[0021] in, represents the hotspot kernel data, θ s represents the solar zenith angle, θ v represents the observation zenith angle, represents the relative azimuth angle between the sun and the sensor, and D represents the angular distance between the sun and the sensor.

[0022] Preferably, the basic shape core data of the present invention is obtained according to the following method:

[0023]

[0024] in represents the basic shape kernel data, and ξ represents the phase angle between the sun, the ground surface, and the sensor.

[0025] Preferably, in step S2, the brightness temperature T(θ) on the channel star of the present invention is obtained according to the following method:

[0026] B i (T i (θ))=ε i (θ)·B i (T s1 )·τ i (θ)+(1-ε i (θ))·R atm↓,i ·τ i +R atm↑,i (θ)

[0027] Where B() represents the Planck equation, T i (θ) represents the brightness temperature of channel i when the observation zenith angle is θ; ε i (θ) represents the channel emissivity of channel i when the observation zenith angle is θ, B i (T s ) represents channel i, temperature is T s1The emission radiance of a black body under Planck's law, τ i represents the atmospheric effective transmittance of channel i, R atm↓,i represents the atmospheric downward radiation of channel i, R atm↑,i (θ) represents the atmospheric upward radiation of channel i when the observation zenith angle is θ.

[0028] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0029] The present invention constructs a directional emissivity model based on a parameter kernel driven model, inputs the study area data, and calculates the directional emissivity. At the same time, the atmospheric profile data is input into MODTRAN 5.2 software to simulate the atmospheric parameters and convolve with the spectral response function of the channel to obtain the channel atmospheric parameters of the two channels of band 31 and band 32, the observed zenith angle, the atmospheric bottom boundary temperature, the average emissivity, and the emissivity difference are grouped, and the channel on-board brightness temperature is calculated, and then the GSW algorithm model is constructed to perform multiple sets of equations and multivariate linear regression to obtain the split window coefficients of the general split window algorithm and then construct a coefficient lookup table under multiple grouping conditions; then the directional emissivity, the observed zenith angle, the atmospheric water vapor content, and the brightness temperature are combined and extracted from the coefficient lookup table. The split window coefficients under the corresponding grouping are input into the GSW algorithm model to invert the surface temperature LST of each pixel in the study area, thereby improving the inversion accuracy of the surface temperature LST. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the process of the present invention;

[0031] Figure 2 Schematic maps of eight IGBP land types in the study area are selected as examples in the embodiments;

[0032] Figure 3 It is a schematic diagram of the principle of grouping according to the observed zenith angle, the bottom boundary temperature of the atmosphere, the average emissivity, and the emissivity difference in the embodiment;

[0033] Figure 4 A schematic diagram of the principle of hierarchical extension grouping is given as an example in an embodiment;

[0034] Figure 5 For example, the polar plot of directional emissivity of woody savanna in the study area during the daytime in summer in band 31 is shown;

[0035] Figure 6 For example, polar plots of directional emissivity of urban and built-up areas in band 32 during summer daytime in the study area are shown;

[0036] Figure 7 For the embodiment, the RMSE of the directional emissivity of the kernel-driven unit of each parameter in the 31 band of the study area is displayed;

[0037] Figure 8 For the embodiment, the RMSE of the directional emissivity of the kernel-driven unit of each parameter in the 32 bands of the study area is displayed;

[0038] Fig. 9 For example, the comparison diagram of LST of three types of inversions is verified using SURFRAD station data in the study area as an example;

[0039] Fig.10 As an example, a comparison chart of LST of three types of inversions is verified using ERA5-Land data taking the study area as an example. DETAILED DESCRIPTION

[0040] The present invention is further described in detail below in conjunction with embodiments:

[0041] Example

[0042] like Figures 1 to 8 As shown, a remote sensing surface temperature angle effect correction method based on an emissivity kernel driven model includes:

[0043] S1. Construct a directional emissivity model based on a parameter kernel driven model. The parameter kernel driven model includes a hot spot kernel and a basic shape kernel. Collect the study area data including MYD21, MYD03, MYD10_L2, MYD021km, MYD11_L2, and MCD12Q1 products (using satellite thermal infrared data at each observation zenith angle VZA within 16 days as multi-angle observation data) and input them into the parameter kernel driven model. Figure 2 Taking the study area as an example (taking the satellite data from 2019 to 2021 covering the mainland of the study area as an example), some data (including observation azimuth angle VAA, observation zenith angle VZA, solar azimuth angle SAA, solar zenith angle SZA; quality assurance QA, atmospheric precipitable water PWV) were obtained using MYD21, MYD03, MYD10_L2, MYD021km, MYD11_L2, MCD12Q1 and other products as follows:

[0044]

[0045] See also Figure 1, the land cover of different colors corresponds to 8 IGBP land types, of which the triangulation points are 7 SURFRAD stations (can be used for data verification). The MYD21 product is reprojected according to the geographic location information of the MYD03 product; in order to reduce the uncertainty of LSE (Land Surface Emissivity in remote sensing and earth sciences), only pixels with mandatory QA flags and good quality are used, and the threshold of precipitable water volume (PWV) is limited to less than 1.5 cm to filter out pixels in relatively dry areas. In addition, MYD10A1 can be used to remove pixels covered by snow. The spatial resolution of the IGBP data extracted from the MCD12Q1 product is 500 meters. In order to match the MYD21 data with a spatial resolution of 1 km, the IGBP can be clustered from 500 meters to 1 km. The rule is that only when four 500-meter pixels have the same land cover type, they are aggregated into a new 1 km pixel, otherwise they are set to null values.

[0046] The directional emissivity model is calculated according to the following formula:

[0047] Directional emissivity = P + hot spot kernel coefficient × hot spot kernel data + basic shape kernel coefficient × basic shape kernel data. Where P is the isotropy coefficient.

[0048] Preferably, the hotspot core data of this embodiment is obtained according to the following method:

[0049]

[0050] in, represents the hotspot kernel data, θ s represents the solar zenith angle, θ v represents the observation zenith angle, represents the relative azimuth angle between the sun and the sensor, and D represents the angular distance between the sun and the sensor.

[0051] Preferably, the basic shape kernel data of this embodiment is obtained according to the following method:

[0052]

[0053] in represents the basic shape kernel data, and ξ represents the phase angle between the sun, the ground surface, and the sensor.

[0054] In some embodiments, the directional emissivity model constructs a number of parameter kernel drive units according to the hot spot kernel and / or the basic shape kernel, selects the parameter kernel drive unit with the smallest root mean square error RMSE to establish the directional emissivity model, inputs the study area data, and calculates the directional emissivity. As a more detailed technical content of this embodiment, the parameter kernel drive model includes 11 parameter kernel drive units, and the directional emissivity formula of the first parameter kernel drive unit (ROU) is The second parameter, the directional emissivity formula of the core driver unit (VIN), is

[0055] The third parameter, the directional firing rate of the nuclear driver unit (RTH), is given by

[0056] The directional firing rate formula of the fourth parameter core driver unit (R1K) is The fifth parameter, the directional firing rate of the nuclear driver unit (RVI), is given by The formula for the directional firing rate of the sixth parameter nuclear drive unit (RVUS) is The seventh parameter, the directional firing rate of the core driver unit (VTH), is given by The directional emission rate formula of the eighth parameter kernel driver unit (VTK) is

[0057] The directional firing rate formula of the ninth parameter core driver unit (VVI) is The directional firing rate formula of the eleventh parameter core drive unit (VUS) is

[0058] The directional emissivity formula of the twelfth parameter kernel driver unit (GUTA_sparse) is where f iso represents the isotropy coefficient, f Hotspot Both represent hot spot kernel coefficients, f BaseShape Represents the basic shape kernel coefficient.

[0059] The hotspot nuclear data K Hotspot_rou Obtained as follows:

[0060]

[0061] where θ s represents the solar zenith angle, θ v represents the observation zenith angle, represents the relative azimuth angle between the sun and the sensor, and D represents the angular distance between the sun and the sensor.

[0062] Hot spot nuclear data Obtained as follows:

[0063]

[0064] The basic shape kernel data K BaseShape thick Obtained as follows:

[0065]

[0066] where ξ is the phase angle between the sun, the ground and the sensor.

[0067] Basic shape kernel data K BaseShape_thin Obtained as follows:

[0068]

[0069] Basic shape kernel data K BaseShape_vin Obtained as follows:

[0070] K BaseShape_vin (θ v )=1-cosθ v

[0071] Basic shape kernel data K BaseShape_usea Obtained as follows:

[0072] K BaseShape_usea (θ v )=sinθ v .

[0073] exist In the formula, each parameter is calculated as follows:

[0074]

[0075] f iso , f bgd , f ori , and f shw is the nuclear coefficient, which is determined by the surface structure and component temperature; K bgd , K ori , and K shw There are three kernels that weight the contributions of surface background, directional effects, and shadows respectively.

[0076] In an embodiment, the parameter kernel driving model constructing the directional emissivity model can select the directional emissivity with the smallest root mean square error RMSE (smallest on a time scale) among the above 11 parameter kernel driving units as the final directional emissivity.

[0077] Taking the data from 2019 to 2021 in the study area as an example, Figure 5As shown, (a) is the firing rate of MYD21, (b)-(I) are the directional firing rates simulated by 11 parameter nuclear driven unit types, and the label in the upper right corner is the RMSE between the simulated firing rate and the original firing rate. Figure 5 , the root mean square error RMSE of the directional emissivity obtained by using the ninth parameter kernel drive unit and the eleventh parameter kernel drive unit is relatively minimal, and a directional emissivity with the smallest root mean square error RMSE can be selected as the final directional emissivity (the directional emissivity of the woody savanna in band 31 during the day in summer). Taking the data from 2019 to 2021 in the study area as an example, Figure 6 As shown in the figure, the directional emissivity simulation effect of band 32 for urban and built-up areas during summer daytime is displayed. The tenth parameter kernel driving unit is used to obtain the minimum root mean square error RMSE of the directional emissivity. It is selected as the directional emissivity model, and the data of the study area are input to obtain the directional emissivity of the city and built-up area in band 32 during summer daytime in the study area.

[0078] Taking the data from 2019 to 2021 in the study area as an example, we can Figure 7 , Figure 8 The RMSE of the directional emissivity of each parameter kernel driver unit in 8 land types, band 31 or band 32, day and night, and four seasons are displayed separately for quick selection.

[0079] S2. Collect atmospheric profile data and input them into MODTRAN 5.2 software to simulate atmospheric parameters and convolve them with the spectral response function of the channel to obtain the channel atmospheric parameters (τ, R) of the two channels of band 31 and band 32. atm↑ and R atm↓ ), the channel atmospheric parameters include the effective transmittance of the atmosphere, the downward atmospheric radiation and the upward atmospheric radiation when the observation zenith angle is θ; the observation zenith angle, the bottom boundary temperature of the atmosphere, the average emissivity and the emissivity difference are grouped (see Figure 3 , observation zenith angle VZA, bottom atmospheric boundary temperature T s, average emissivity ε, and emissivity difference Δε are grouped together), and then the Planck inverse operation is performed to obtain the channel satellite brightness temperature under each group. The grouping of average emissivity ε and emissivity difference Δε is as follows: (0.90, 0.96), (0.94, 1.0), 2 groups. The grouping of observation zenith angle is as follows: [1, 2.8], step size is 0.2, 10 groups. The grouping of the bottom boundary temperature of the atmosphere is as follows: ≤280K, [275, 295), [290, 310), [305, 325), ≥320, 5 groups. If the atmospheric water vapor content WVC is introduced, the grouping of the atmospheric water vapor content WVC is as follows: [0, 1.5], [1.0, 2.5], [2.0, 3.5], [3.0, 4.5], [4.0, 5.5], [5.0, 6.5], 6 groups.

[0080] In some embodiments, Figure 4 As shown, in addition to "grouping the observed zenith angle, bottom atmospheric boundary temperature, average emissivity, and emissivity difference", the grouping in this embodiment can also be based on Figure 4 The data are divided into eight land surface types, four seasons, day and night (i.e., the data are divided into eight land surface types, four seasons, day and night).

[0081] In some embodiments, the brightness temperature T(θ) on the channel star is obtained as follows:

[0082] B i (T i (θ))=ε i (θ)·B i (T s1 )·τ i (θ)+(1-ε i (θ))·R atm↓,i ·τ i +R atm↑,i (θ)

[0083] Where B() represents the Planck equation, T i (θ) represents the brightness temperature of channel i when the observation zenith angle is θ; ε i (θ) represents the channel emissivity of channel i when the observation zenith angle is θ, B i (T s ) represents channel i, temperature is T s1 The emission radiance of a black body under Planck's law, τ i represents the atmospheric effective transmittance of channel i, R atm↓,i represents the atmospheric downward radiation of channel i, R atm↑,i (θ) represents the atmospheric upward radiation of channel i when the observation zenith angle is θ.

[0084] S3, constructing a GSW algorithm model, inputting the bottom boundary temperature of the atmosphere, the brightness temperature on the channel satellite, the average emissivity, and the emissivity difference into the GSW algorithm model, obtaining multiple sets of equations according to the following algorithm formula, performing multiple linear regression on them, and obtaining split window coefficients a0 to a7 of a general split window algorithm (the split window algorithm in this embodiment preferably uses two split window channels), and then constructing coefficient lookup tables under various grouping conditions;

[0085]

[0086] Among them, T s represents the bottom boundary temperature of the atmosphere, T 31 and T 32 They represent the on-board brightness temperatures of the channels in bands 31 and 32 respectively, Δε represents the emissivity difference between bands 31 and 32, ε represents the average emissivity of bands 31 and 32, and a0 to a7 are the split window coefficients in the coefficient lookup table.

[0087] S4, extract the observation zenith angle, atmospheric water vapor content, brightness temperature from the study area data and combine them with the directional emissivity obtained in step S1, traverse and extract the split window coefficient under the corresponding group from the coefficient lookup table, and then input the directional emissivity, split window coefficient and brightness temperature into the GSW algorithm model to obtain the surface temperature LST of each pixel in the study area according to the following inversion formula:

[0088] Among them, ε p represents the average emissivity of pixel bands 31 and 32 in the study area, Δε p It represents the difference in emissivity between bands 31 and 32 of the pixel in the study area, T p31 and T p32 represents the brightness temperature of bands 31 and 32 (brightness temperature on the same channel), and a0 to a7 are split window coefficients obtained in the coefficient lookup table. Multiple groups of the coefficient lookup table (grouping can further introduce atmospheric water vapor content WVC) and split window coefficients under the grouping have been obtained in step S3. This embodiment performs grouping, and a rough LST can be estimated. coarse Then based on LST coarse , emissivity, etc., find the corresponding split window coefficient in the lookup table, and obtain a0 to a7 corresponding to VZA by interpolation, and finally invert the accurate LST accurate .

[0089] Taking the study area as an example, see Fig. 9 , Fig.10, where LST_GSW_DE is the surface temperature LST considering the angle effect based on the method of the present invention. It is the LST data obtained by inputting into the GSW algorithm based on the directional reflectivity inverted by the method of the present invention combined with the coefficients in the coefficient lookup table under grouping conditions; LST_TES is the surface temperature LST considering the angle effect, which is derived from the LST product of MYD21; LST_GSW_CE does not consider the angle effect, and is the LST data obtained by inputting into the GSW algorithm based on the classified emissivity (LSE product of MYD11_L2) combined with the coefficients in the coefficient lookup table under grouping conditions.

[0090] S5. Collect data including SURFRAD station data and / or ERA5_LAND data in the study area and obtain the true value of LST. Construct a loss constraint function in the GSW algorithm model to constrain the loss of surface temperature LST and the true value of LST. Take SURFRAD station data as an example. SURFRAD station data provides accurate, continuous and long-term surface radiation budget data, which consists of seven observation stations with different climates covering the study area; SURFRAD provides minute-by-minute surface radiation observation data, which can be used to evaluate satellite products. The geographic information of SURFRAD stations and IGBP surface coverage are shown in the following table:

[0091]

[0092]

[0093] When performing ground authenticity verification, the ground observation and satellite imaging times must be unified to UTC (Coordinated Universal Time), and the site longitude is used as an auxiliary condition to convert between the ground observation point time and UTC. In addition, this embodiment also uses the 3-times standard deviation method to eliminate outliers in the site and inverted LST. In some preferred embodiments, the method for obtaining the true value of LST from SURFRAD site data is as follows: SURFRAD site data is uniformly converted to Coordinated Universal Time, and then the true value of LST is obtained according to the following formula: Among them LST s真 represents the true value of LST, R ↑ and R ↓ denote the upward and downward radiation from the surface, σ SB represents the Stefan Boltzmann constant, ε bb Represents the broadband emissivity; ε bb The calculation method is as follows: bb =0.095+0.329*ε 29 +0.572*ε 31 , where ε 29 and ε 31They represent the narrow band emissivity of band 29 and band 31 obtained from the MYD21 product. ERA5_LAND data is a climate reanalysis data that provides high-resolution surface variable information with a grid spacing of about 9 kilometers per hour, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF); in ERA5-Land data, the surface temperature represents the temperature of the uppermost surface, in K, with a temporal resolution of hours and a spatial resolution of 11132 meters.

[0094] The present invention uses the data including SURFRAD station data and / or ERA5_LAND data to obtain the true value of LST. First, the surface temperature LST corresponding to the LST true value area can be obtained based on the method of the present invention to verify the data accuracy. After actual verification, the data accuracy is high. Second, a loss constraint function can be constructed in the GSW algorithm model to perform loss constraints on the surface temperature LST and the true value of LST. Through the loss constraint of the loss constraint function, the inversion accuracy of the surface temperature LST obtained by the method of the present invention is further improved.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for correcting the angle effect of remote sensing surface temperature based on an emissivity kernel driven model, characterized in that: The methods include: S1. A directional emissivity model is constructed based on a parameter kernel driven model. The parameter kernel driven model includes a hot spot kernel and a basic shape kernel. Data of the study area including MYD21, MYD03, MYD10_L2, MYD021km, MYD11_L2, and MCD12Q1 products are collected and input into the parameter kernel driven model. The directional emissivity model is calculated according to the following formula: Directional emissivity = P + hot spot kernel coefficient × hot spot kernel data + basic shape kernel coefficient × basic shape kernel data. Where P is the isotropy coefficient; S2. Collect atmospheric profile data and input them into MODTRAN 5.2 software to simulate atmospheric parameters and convolve them with the spectral response function of the channel to obtain the channel atmospheric parameters of band 31 and band 32. The channel atmospheric parameters include the effective transmittance of the atmosphere, the atmospheric downward radiation and the atmospheric upward radiation when the observation zenith angle is θ; group the observation zenith angle, the boundary temperature of the bottom layer of the atmosphere, the average emissivity and the emissivity difference, and then perform Planck inverse operation to obtain the channel on-board brightness temperature under each group; S3, constructing a GSW algorithm model, inputting the bottom boundary temperature of the atmosphere, the brightness temperature on the channel satellite, the average emissivity, and the emissivity difference into the GSW algorithm model to obtain multiple sets of equations according to the following algorithm formula, performing multiple linear regression on them, and obtaining the split window coefficients a0 to a7 of the general split window algorithm, and then constructing a coefficient lookup table under various grouping conditions; Among them, T s represents the bottom boundary temperature of the atmosphere, T 31 and T 32 represents the on-board brightness temperature of the channel of bands 31 and 32, respectively; Δε represents the difference in emissivity between bands 31 and 32; ε represents the average emissivity of bands 31 and 32; a0 to a7 are the split window coefficients in the coefficient lookup table; S4, extract the observation zenith angle, atmospheric water vapor content, brightness temperature from the study area data and combine them with the directional emissivity obtained in step S1, traverse and extract the split window coefficient under the corresponding group from the coefficient lookup table, and then input the directional emissivity, split window coefficient and brightness temperature into the GSW algorithm model to obtain the surface temperature LST of each pixel in the study area according to the following inversion formula: Average emissivity of bands 31 and 32, △ε p represents the difference in emissivity between bands 31 and 32, T p31 and T p32 They represent the brightness temperatures of bands 31 and 32 respectively, and a0 to a7 are the split window coefficients obtained from the coefficient lookup table.

2. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 1, characterized in that: Also includes the following methods: S5. Collect data in the study area including SURFRAD station data and / or ERA5_LAND data and obtain the true value of LST, and construct a loss constraint function in the GSW algorithm model to constrain the loss of the surface temperature LST and the true value of LST.

3. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 2, characterized in that: In step S5, the method for obtaining the LST true value from the SURFRAD station data is as follows: the SURFRAD station data is uniformly converted to the Coordinated Universal Time, and then the LST true value is obtained according to the following formula: Among them LST s True means LST truth value, R ↑ and R ↓ denote the upward and downward radiation from the surface, σ SB represents the Stefan Boltzmann constant, ε bb Represents the broadband emissivity; ε bb The calculation method is as follows: bb =0.095+0.329*ε 29 +0.572*ε 31 , where ε 29 and ε 31 They respectively represent the narrow-band emissivity of band 29 and band 31 obtained from the MYD21 product.

4. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 1 or 2, characterized in that: The parameter kernel driven model constructs several parameter kernel driven units according to the hot spot kernel and / or the basic shape kernel, and selects the parameter kernel driven unit with the smallest root mean square error (RMSE) to establish the directional emissivity model. The data of the study area is input to calculate the directional emissivity.

5. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 1, characterized in that: The hotspot nuclear data is obtained as follows: in, represents the hotspot kernel data, θ s represents the solar zenith angle, θ v represents the observation zenith angle, represents the relative azimuth angle between the sun and the sensor, and D represents the angular distance between the sun and the sensor.

6. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 1, characterized in that: The basic shape kernel data is obtained as follows: in represents the basic shape kernel data, and ξ represents the phase angle between the sun, the ground, and the sensor.

7. The method for correcting the angle effect of remote sensing surface temperature based on the emissivity kernel driven model according to claim 1, characterized in that: In step S2, the brightness temperature T(θ) on the channel star is obtained as follows: B i (T i (i))=e i (θ)·B i (T s1 )·t i (θ)+(1-ε i (i))·R atm↓,i ·t i +R atm↑,i (i) Where B() represents the Planck equation, T i (θ) represents the brightness temperature of channel i when the observation zenith angle is θ; ε i (θ) represents the channel emissivity of channel i when the observation zenith angle is θ, B i (T s ) represents channel i, temperature is T s1 The emission radiance of a black body under Planck's law, τ i represents the effective atmospheric transmittance of channel i, R atm↓,i represents the atmospheric downward radiation of channel i, R atm↑,i (θ) represents the atmospheric upward radiation of channel i when the observation zenith angle is θ.

Citation Information

Patent Citations

  • Surface temperature inversion method for broadband thermal imager of unmanned aerial vehicle

    CN113447137A

  • Atmospheric water vapor content inversion method and system based on MODIS data

    CN114581791A

  • Surface temperature product angle normalization method based on kernel-driven model

    CN117371333A

  • Method for calculating land surface temperature

    KR1020110096301A

Cited By

  • Surface temperature split window remote sensing inversion method and system based on virtual channel

    CN119442917A

  • Environmental radiation correction method and system based on three-dimensional angle coefficient mapping

    CN120869360A