A method for optical satellite multi-satellite radiation reference correction and consistency evaluation

Abstract

The application discloses a kind of optical satellite multi-star radiation reference correction and consistency evaluation method, using the field observation data of global radiation field network respectively to selected optical reference satellite, each optical contrast satellite is re-corrected, and the radiation reference of the optical reference satellite, each optical contrast satellite is corrected to the same level;After re-correcting, optical reference satellite and each optical contrast satellite are combined two by two, and the following operations are completed for each pair of combination: spatial matching and spectral matching factor matching are carried out, and star matching data set is obtained;The consistency evaluation coefficient of star matching data set is calculated, and the results of optical reference satellite and optical contrast satellite are evaluated.The application can accurately evaluate the consistency of multi-star radiation by optimizing the calibration difference of satellite itself when evaluating the consistency of optical satellite.

Description

Technical Field

[0001] This invention relates to a method for multi-satellite radiation reference calibration and consistency evaluation of optical satellites, belonging to the field of satellite application technology. Background Technology

[0002] Currently, in order to fully compare the observation capabilities of various satellites during their on-orbit period and form a long-term, sustainable Earth observation dataset, relevant organizations at home and abroad, such as NASA in the United States and EAS in the European Union, attach great importance to the consistency assessment of satellite-acquired data. For the long-term study of the changes and trends of products from optical satellite series, including basic radiation and derivative products, it is necessary to assess the consistency of these products.

[0003] Common methods for assessing the radiometric consistency of multiple optical satellites typically involve directly performing quality control on the radiometric products of the optical satellite sequence, then performing spatiotemporal matching of the products, and finally conducting a consistency analysis on the matched datasets. However, due to factors such as payload attenuation during the on-orbit period of optical satellites and differences in data processing methods, this method often results in varying systemic differences in the radiometric products of optical satellites. Summary of the Invention

[0004] The technical problem solved by this invention is to overcome the shortcomings of the prior art and propose a method for multi-satellite radiation reference calibration and consistency evaluation of optical satellites. This method uses a global radiation field network to calibrate the product accuracy of optical satellites, optimizes the calibration differences of the satellites themselves when re-evaluating the consistency of optical satellites, and accurately evaluates the radiation consistency of multiple satellites.

[0005] The technical solution of this invention is:

[0006] A method for multi-satellite radiation reference calibration based on a global radiation field network includes: selecting one optical satellite as the optical reference satellite from among multiple optical satellites to be compared, and using the remaining optical satellites as optical comparison satellites; recalibrating the optical reference satellite and each optical comparison satellite using field observation data from the global radiation field network, so as to calibrate the radiation reference of the optical reference satellite and each optical comparison satellite to the same level.

[0007] Preferably, the optical reference satellite and each optical comparison satellite are recalibrated using in-situ observation data from the global radiation field network. The recalibration method is the same, specifically as follows:

[0008] Spatiotemporal matching was performed on the field observation data of the global radiation field network and the observation data of the satellites to be recalibrated to obtain the satellite-ground matching dataset;

[0009] Quality control is performed on the satellite-ground matching dataset. Centered on the satellite imaging pixels corresponding to the latitude and longitude of the field data acquisition, a set of pixels with more than 50% of the effective pixels of (2n-1)×(2n-1) pixels is selected, where n takes any integer from 2, 3, or 4. The uniformity of the effective pixels in the set of pixels is calculated, and the data whose uniformity meets the accuracy requirements of satellite radiometric products are used as the satellite observation data in the satellite-ground matching dataset.

[0010] The spectral data measured in the field of the global radiation field network are integrated according to the band response function of the satellite to be recalibrated, and the resulting data is used as the ground observation data of the satellite-ground matching dataset;

[0011] Using ground observation data as the vertical axis and satellite observation data as the horizontal axis, the linear correlation coefficient is calculated; the observation data of the satellite to be recalibrated is then recalibrated based on the linear correlation coefficient.

[0012] Preferably, spatiotemporal matching is performed between in-situ observation data from the global radiation field network and observation data from satellites to be recalibrated, including:

[0013] Perform time matching: Filter data whose transit time of the satellite to be recalibrated differs from the actual data acquisition time of the Global Radiation Field Network by within ±30 minutes;

[0014] Spatial matching: Calculate the imaging pixels of the satellite to be recalibrated based on the latitude and longitude data collected on-site from the global radiation field network.

[0015] lon j-left <lon i <lon j-right

[0016] lat j-bottom <lat i <lat j-top

[0017] In the formula, lon i The longitude of the data collected on-site from the global radiation field network is lat. i The latitude of the data collected on-site from the global radiation field network, lon j-left To recalibrate the left boundary longitude of the satellite's pixels, lon j-right To recalibrate the right boundary longitude of the satellite pixels, lat j-bottom To recalibrate the lower boundary latitude of the satellite pixels, lat j-top The upper boundary latitude of the pixels of the satellite to be recalibrated;

[0018] Centered on the satellite imaging pixels corresponding to the measured data of the global radiation field network, the average value of (2n-1)×(2n-1) pixels is calculated, and a satellite-to-ground matching dataset is established with the corresponding measured data of the global radiation field network.

[0019] Preferably, the spectral data measured in-situ from the global radiation field network are integrated according to the band response function of the satellite to be recalibrated:

[0020]

[0021] Among them, R rs (λ) represents the reflectivity value of a single band in the field, λ0 and λ1 represent the range of the band, and R k (λ) is the response coefficient of the corresponding single band.

[0022] Preferably, one optical satellite is selected from the multiple optical satellites to be compared as the optical reference satellite. The selection principle is to select a sun-synchronous satellite with good satellite payload status and stable satellite product quality.

[0023] A method for evaluating the radiometric consistency of multiple optical satellites based on a global radiation field network, which, based on the recalibrated optical reference satellite and each optical comparison satellite obtained in claim 1, pairs the optical reference satellite and each optical comparison satellite together, and performs the following operations on each pair: performing spatiotemporal matching and spectral matching factor matching to obtain a star matching dataset; calculating the consistency evaluation coefficient of the star matching dataset, and evaluating the results of the optical reference satellite and the optical comparison satellite.

[0024] Preferably, the method for combining the recalibrated optical reference satellite and an optical comparison satellite for spatiotemporal matching is as follows:

[0025] First, time matching is performed: the difference in transit time between the optical reference satellite and the optical comparison satellite is within ±15 minutes, and the corresponding data of the optical comparison satellite is obtained by screening.

[0026] Next, spatial matching is performed: the optical reference satellite and the optical comparison satellite are gridded at the same resolution, and the latitude and longitude of each pixel of the optical reference satellite and the optical comparison satellite are calculated to obtain the gridded pixel corresponding to that pixel;

[0027] The data from the optical reference satellite and the optical contrast satellite are matched one by one according to the grid order. When multiple pixels correspond to one grid, the average value of the multiple pixels is used for matching. The gridded data from the optical reference satellite and the optical contrast satellite are matched grid by grid to obtain the star matching dataset.

[0028] Preferably, the method for performing spectral matching factor matching by combining the recalibrated optical reference satellite and an optical contrast satellite is as follows:

[0029] The bands where the center wavelength difference between the optical reference satellite and the optical comparison satellite is less than 5 nm are used as comparison bands. The spectral matching factor of the star crossover is calculated according to the band response functions of the optical reference satellite and the optical comparison satellite.

[0030]

[0031] In the formula, R band (λ) is the spectral matching factor between the optical reference satellite band and the corresponding band of the optical comparison satellite, R rs (λ) represents the reflectivity value for a typical field application, λ0 and λ1 represent the band range of the optical reference satellite band response function, and R k1 (λ) represents the response coefficient of a single band corresponding to the optical reference satellite band, λ3 and λ4 represent the band range of the optical comparison satellite band response function, and R k2 (λ) is the response coefficient of a single band corresponding to the optical contrast satellite band;

[0032] The optical contrast satellite band radiance values ​​in the star-matching data are multiplied by the corresponding band spectral matching factor of the optical reference satellite band to the optical contrast satellite to obtain the corrected optical contrast satellite band radiance values ​​in the star-matching data, which are used for consistency assessment.

[0033] Preferably, the consistency evaluation coefficient of the star matching dataset is calculated to evaluate the results of the optical benchmark satellite and the optical comparison satellite, including:

[0034] Calculate the covariance of the star matching dataset; if the covariance is positive, it is positively correlated, and if the covariance is negative, it is negatively correlated. The larger the value, the stronger the correlation and the better the consistency.

[0035] Preferably, the consistency evaluation coefficient of the star matching dataset is calculated to evaluate the results of the optical benchmark satellite and the optical comparison satellite, including:

[0036] Calculate the covariance of the star-matching dataset;

[0037] Analyze the correlation coefficients of the dataset, including univariate regression, to solve for the intercept b1 and slope b0;

[0038] The correlation coefficient in the covariance consistent dataset, based on positive correlation, is used to judge the systematic differences between the optical reference satellite and the optical comparison satellite. The closer the slope is to 1 and the smaller the intercept, the higher the consistency between the two.

[0039] The advantages of this invention compared to the prior art are:

[0040] (1) This invention uses global radiation field network data to recalibrate optical reference satellites and optical comparison satellites, which can reduce the differences in satellite radiation products caused by differences in satellite payload attenuation and satellite processing algorithms. After correcting these errors, a consistency analysis is performed on the optical reference satellites and optical comparison satellites. The comparison process designed in this invention can better compare the radiation consistency of optical reference satellites and optical comparison satellites.

[0041] (2) This invention, through rational spatiotemporal matching combined with refined band correction factors, more meticulously compares the radiation consistency of optical reference satellites and optical comparison satellites, effectively overcoming the problem of low accuracy in traditional consistency assessment methods. Attached Figure Description

[0042] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0043] Figure 1 This is a flowchart of the method for evaluating the radiometric consistency of multiple optical satellites according to an embodiment of the present invention. Detailed Implementation

[0044] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0045] To address the requirements for the radiometric consistency assessment of multiple optical satellites, this invention proposes a radiometric benchmark correction and consistency assessment method based on a global radiation field network. Starting with the recalibration of optical benchmark satellites and optical comparison satellites based on global radiation field network data, this invention constructs a radiometric consistency assessment method for multiple optical satellites, along with related data processing methods and procedures.

[0046] This method is as follows Figure 1 As shown, it specifically includes:

[0047] (S1) Select optical reference satellites and optical comparison satellites

[0048] Following the principles below, a unique optical reference satellite and optical comparison satellite are identified among the multiple optical satellites to be compared.

[0049] a) Optical reference satellites are selected from sun-synchronous satellites with good payload status, stable satellite product quality, large time span, and high spatial coverage. Generally, there is only one reference satellite.

[0050] b) There can be multiple optical comparison satellites. Each satellite is compared with a reference satellite, and then multi-satellite consistency analysis can be performed using the reference satellite.

[0051] (S2) Recalibrate the radiation products of the optical reference satellite using field observation data from the global radiation field network.

[0052] a) Spatiotemporal matching is performed on the field observation data of the global radiation field network and the observation data of the optical reference satellite to obtain the satellite-ground matching dataset. Spatiotemporal matching is divided into time matching and space matching.

[0053] Spatiotemporal matching: The difference between the transit time of the optical reference satellite and the actual data acquisition time of the global radiation field network is within ±30 minutes.

[0054] Spatial matching: The imaging pixels of the optical reference satellite are calculated based on the latitude and longitude data collected on-site from the global radiation field network. The calculation formula is as follows:

[0055] lon j-left <lon i <lon j-right

[0056] lat j-bottom <lat i <lat j-top

[0057] Among them, lon i The longitude of the data collected on-site from the global radiation field network is lat. i The latitude of the data collected on-site from the global radiation field network, lon j-left The left boundary longitude of the pixels of the optical reference / comparison satellite, lon j-right lat represents the right boundary longitude of the pixels of the optical reference satellite. j-bottom lat represents the lower boundary latitude of the pixels of the optical reference satellite. j-top The latitude of the upper boundary of the pixels of the optical reference satellite.

[0058] Centered on the satellite imaging pixels corresponding to the measured data of the global radiation field network, the average value of 5×5 pixels is calculated, and a satellite-to-ground matching dataset is established with the corresponding measured data of the global radiation field network.

[0059] b) Perform quality control on the matching dataset. Centering on the satellite imaging pixels corresponding to the latitude and longitude of the field data acquisition, calculate that the number of effective pixels in a 5×5 pixel array is greater than 50%, and calculate the uniformity of the effective pixels in the 5×5 pixel array. If the uniformity is less than 15% (the accuracy of general satellite radiometric products), then the average value of the effective pixels in the 5×5 pixel array is used as the corresponding value. The formula for calculating uniformity is as follows:

[0060]

[0061] Where, x j The radiation value of a single pixel. It is the average value of the effective pixels in a 5×5 pixel grid.

[0062] Data that meets the homogeneity requirement will be used as satellite observation data in the satellite-to-ground matching dataset.

[0063] c) Integrate the spectral data measured in the field from the global radiation field network according to the band response function of the optical reference satellite. The integration formula for the band response function is as follows:

[0064]

[0065] Among them, R rs (λ) represents the reflectivity value of a single band in the field, λ0 and λ1 represent the range of the band, and R k (λ) is the response coefficient of the corresponding single band.

[0066] The data after band response integration is used as ground observation data for the satellite-to-ground matching dataset.

[0067] d) For the satellite-ground matching dataset, with ground observation data as the vertical axis and satellite observation data as the horizontal axis, calculate the linear correlation coefficient, such as intercept and slope, using big data analysis.

[0068] The observation data of the optical reference satellite and the optical comparison satellite are recalibrated based on the intercept and slope obtained from the analysis, so that the radiation reference of the optical reference satellite and the optical comparison satellite is calibrated to the same level.

[0069] (S3) Recalibrate the radiation products of the optical comparison satellite using field observation data from the global radiation field network. This step is the same as step (S2).

[0070] (S4) After performing data quality control such as identifying anomalies at level 2 and removing observation anomalies on the recalibrated optical reference satellite and optical comparison satellite, the two are matched for spatiotemporal matching and spectral matching factors.

[0071] a) Spatiotemporal matching of observation data from optical reference satellites and comparison satellites is performed to obtain a satellite-to-ground matching dataset. Spatiotemporal matching is divided into time matching and spatial matching.

[0072] Time matching: The difference in transit time between the optical reference satellite and the optical comparison satellite is within ±15 minutes;

[0073] Spatial matching: The optical reference satellite and the optical comparison satellite are gridded at the same resolution. The latitude and longitude of each pixel of the optical reference / comparison satellite are calculated according to the following formula to obtain the gridded pixel corresponding to that pixel.

[0074]

[0075]

[0076] In the formula, i lon For pixel values ​​in the longitude direction (horizontal direction) of a two-dimensional grid, lon j For optical reference / comparison, the longitude of each pixel on the satellite is lon left The minimum longitude (positive for east, negative for west) corresponds to the gridded region. step This represents the longitude resolution of the two-dimensional grid, i.e., the longitude difference between two adjacent grids. `int` represents integer rounding (all decimal places are discarded). lat For the pixel values ​​in the 2D grid along the lat direction (vertical direction), lat j For optical reference / comparison, the latitude of each pixel in the satellite, lat top The lat value represents the maximum latitude corresponding to the gridded region (positive for North latitude, negative for South latitude). step This represents the latitudinal resolution of the two-dimensional grid, i.e., the latitudinal difference between two adjacent grids. `int` represents integer (all decimal places are discarded).

[0077] The data from the optical reference satellite and the optical contrast satellite are matched one by one according to the grid order. When multiple pixels correspond to one grid, the average value of the multiple pixels is used for matching. The gridded data from the optical reference satellite and the optical contrast satellite are matched grid by grid to obtain the star matching dataset.

[0078] c) Calculation of spectral matching factor: Bands with a center wavelength difference of less than 5 nm between the optical reference satellite and the optical comparison satellite are used as comparison bands (when the center wavelength difference is greater than 5 nm, the two bands are not suitable for comparative analysis due to differences in band observation values). Typical ground spectra of the selected matching area are used, and the spectral matching factor of the star intersection is calculated according to the band response functions of the optical reference satellite and the optical comparison satellite. The specific formula is as follows:

[0079]

[0080] Among them, R band(λ) is the spectral matching factor between the optical reference satellite band and the corresponding band of the optical comparison satellite, R rs (λ) represents the reflectivity value for a typical field application, λ0 and λ1 represent the band range of the optical reference satellite band response function, and R k1 (λ) represents the response coefficient of a single band corresponding to the optical reference satellite band, λ3 and λ4 represent the band range of the optical comparison satellite band response function, and R k1 (λ) is the response coefficient of a single band corresponding to the optical contrast satellite band.

[0081] Multiply the radiance value of the optical comparison satellite band in the star matching data by the spectral matching factor of the optical reference satellite band to the corresponding band of the optical comparison satellite to obtain the radiance value of the optical comparison satellite band in the star matching data.

[0082] (S5) Calculation of the consistency evaluation coefficient between optical reference satellites and optical comparison satellites

[0083] For the star matching dataset, use big data analysis to determine the correlation between the two datasets.

[0084] a) Analyze the covariance of the dataset, using the following formula:

[0085]

[0086] Where x j and y represents the band radiance value and band average value of pixel j of the optical reference satellite. j and This represents the radiance value and average radiance value of pixel j in the optical contrast satellite.

[0087] b) Analyze the correlation coefficients of the dataset, including using univariate regression to solve for the intercept b1 and slope b0. The specific formulas are as follows:

[0088]

[0089]

[0090] (S6) Consistency Assessment Analysis of Optical Reference Satellites and Optical Comparison Satellites

[0091] The results of the optical reference satellite and the optical comparison satellite are evaluated based on the consistency evaluation coefficients calculated above.

[0092] a) The covariance of the dataset: if it is positive, it is positively correlated; if it is negative, it is negatively correlated. The larger the value, the stronger the correlation and the better the consistency.

[0093] b) The correlation coefficient in the covariance consistent dataset is used to judge the systematic differences between the optical reference satellite and the optical comparison satellite based on positive correlation. The closer the slope is to 1 and the smaller the intercept, the higher the consistency between the two.

[0094] The criteria for determining covariance and correlation coefficients shall be determined by the personnel who subsequently apply satellite radiation products.

[0095] The embodiments described above are merely preferred embodiments of the present invention. Ordinary variations and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for multi-satellite radiation reference calibration of optical satellites based on a global radiation field network, characterized in that, include: One optical satellite is selected from the multiple optical satellites to be compared as the optical reference satellite, and the remaining optical satellites are used as optical comparison satellites. The optical reference satellite and each optical comparison satellite were recalibrated using in-situ observation data from the global radiation field network, so that the radiation reference of the optical reference satellite and each optical comparison satellite was calibrated to the same level. The optical reference satellite and each optical comparison satellite were recalibrated using in-situ observation data from the global radiation field network. The recalibration method was the same, specifically as follows: Spatiotemporal matching was performed on the field observation data of the global radiation field network and the observation data of the satellites to be recalibrated to obtain the satellite-ground matching dataset; Quality control is performed on the satellite-ground matching dataset. Centered on the satellite imaging pixels corresponding to the latitude and longitude of the field data acquisition, a set of pixels with more than 50% of the effective pixels of (2n-1)×(2n-1) pixels is selected, where n takes any integer from 2, 3, or 4. The uniformity of the effective pixels in the set of pixels is calculated, and the data whose uniformity meets the accuracy requirements of satellite radiometric products are used as the satellite observation data in the satellite-ground matching dataset. The spectral data measured in the field of the global radiation field network are integrated according to the band response function of the satellite to be recalibrated, and the resulting data is used as the ground observation data of the satellite-ground matching dataset; Using ground-based observation data as the vertical axis and satellite observation data as the horizontal axis, calculate the linear correlation coefficient. The observation data of the satellite to be recalibrated are recalibrated based on the linear correlation coefficient.

2. The optical satellite multi-satellite radiation reference correction method based on a global radiation field network according to claim 1, characterized in that, Spatiotemporal matching was performed on in-situ observation data from the global radiation field network and observation data from satellites requiring recalibration, including: Perform time matching: Filter data whose transit time of the satellite to be recalibrated differs from the actual data acquisition time of the Global Radiation Field Network by within ±30 minutes; Spatial matching: Calculate the imaging pixels of the satellite to be recalibrated based on the latitude and longitude data collected on-site from the global radiation field network. lon j-left <lon i <lon j-right years j-bottom <lat i <lat j-top In the formula, lon i The longitude of the data collected on-site from the global radiation field network is lat. i The latitude of the data collected on-site from the global radiation field network, lon j-left To recalibrate the left boundary longitude of the satellite's pixels, lon j-right To recalibrate the right boundary longitude of the satellite pixels, lat j-bottom To recalibrate the lower boundary latitude of the satellite pixels, lat j-top The upper boundary latitude of the pixels of the satellite to be recalibrated; Centered on the satellite imaging pixels corresponding to the measured data of the global radiation field network, the average value of (2n-1)×(2n-1) pixels is calculated, and a satellite-to-ground matching dataset is established with the corresponding measured data of the global radiation field network.

3. The optical satellite multi-satellite radiation reference correction method based on a global radiation field network according to claim 1, characterized in that, The spectral data measured in the field from the global radiation field network are integrated according to the band response function of the satellite to be recalibrated: Among them, R rs (λ) represents the reflectivity value of a single band in the field, λ0 and λ1 represent the range of the band, and R k (λ) is the response coefficient of the corresponding single band.

4. The optical satellite multi-satellite radiation reference correction method based on a global radiation field network according to claim 1, characterized in that, One optical satellite was selected from among the multiple optical satellites to be compared as the optical reference satellite. The selection principle was to choose a sun-synchronous satellite with good satellite payload status and stable satellite product quality.

5. A method for assessing the radiometric consistency of multiple optical satellites based on a global radiation field network, characterized in that, Based on the recalibrated optical reference satellite and each optical comparison satellite obtained in claim 1, the optical reference satellite and each optical comparison satellite are paired together, and the following operations are performed on each pair: spatiotemporal matching and spectral matching factor matching are performed to obtain a star matching dataset. Calculate the consistency evaluation coefficient of the star matching dataset to evaluate the results of the optical benchmark satellite and the optical comparison satellite.

6. The method for evaluating the radiometric consistency of multiple optical satellites based on a global radiation field network according to claim 5, characterized in that, The method for combining the recalibrated optical reference satellite and an optical comparison satellite for spatiotemporal matching is as follows: Time matching was performed: the difference in transit time between the optical reference satellite and the optical comparison satellite was within ±15 minutes, and the corresponding data of the optical comparison satellite were obtained by screening. Spatial matching is performed: the optical reference satellite and the optical contrast satellite are gridded at the same resolution, and the latitude and longitude of each pixel in the optical reference satellite and the optical contrast satellite are calculated to obtain the gridded pixel corresponding to that pixel. In the formula, i lon j lat These represent the pixel values ​​along the longitude and latitude axes of a two-dimensional grid, respectively. j lat j These represent the longitude and latitude of each pixel of the satellite currently being calculated. left The minimum longitude corresponding to the gridded region, lon step This represents the longitude resolution of a two-dimensional grid, specifically the longitude difference between two adjacent grid cells; `int` represents the integer value. top lat represents the maximum latitude corresponding to the gridded region. step The latitudinal resolution of a two-dimensional grid is the latitudinal difference between two adjacent grids. The data from the optical reference satellite and the optical contrast satellite are matched one by one according to the grid order. When multiple pixels correspond to one grid, the average value of the multiple pixels is used for matching. The gridded data from the optical reference satellite and the optical contrast satellite are matched grid by grid to obtain the star matching dataset.

7. The method for evaluating the radiometric consistency of multiple optical satellites based on a global radiation field network according to claim 5, characterized in that, The method for performing spectral matching factor matching by combining the recalibrated optical reference satellite and an optical comparison satellite is as follows: The bands where the center wavelength difference between the optical reference satellite and the optical comparison satellite is less than 5 nm are used as comparison bands. The spectral matching factor of the star crossover is calculated according to the band response functions of the optical reference satellite and the optical comparison satellite. In the formula, R band (λ) is the spectral matching factor between the optical reference satellite band and the corresponding band of the optical comparison satellite, R rs (λ) represents the reflectivity value for a typical field application, λ0 and λ1 represent the band range of the optical reference satellite band response function, and R k1 (λ) represents the response coefficient of a single band corresponding to the optical reference satellite band, λ3 and λ4 represent the band range of the optical comparison satellite band response function, and R k2 (λ) is the response coefficient of a single band corresponding to the optical contrast satellite band; The optical contrast satellite band radiance values ​​in the star-matching data are multiplied by the corresponding band spectral matching factor of the optical reference satellite band to the optical contrast satellite to obtain the corrected optical contrast satellite band radiance values ​​in the star-matching data, which are used for consistency assessment.

8. The method for evaluating the radiometric consistency of multiple optical satellites based on a global radiation field network according to claim 5, characterized in that, The consistency evaluation coefficient of the star matching dataset is calculated, and the results of the optical benchmark satellite and the optical comparison satellite are evaluated, including: Calculate the covariance of the star matching dataset; if the covariance is positive, it is positively correlated, and if the covariance is negative, it is negatively correlated. The larger the value, the stronger the correlation and the better the consistency.

9. The method for evaluating the radiometric consistency of multiple optical satellites based on a global radiation field network according to claim 5, characterized in that, The consistency evaluation coefficient of the star matching dataset is calculated, and the results of the optical benchmark satellite and the optical comparison satellite are evaluated, including: Calculate the covariance of the star-matching dataset; Analyze the correlation coefficients of the dataset, including univariate regression, to solve for the intercept b1 and slope b0; The correlation coefficient in the covariance consistent dataset, based on positive correlation, is used to judge the systematic differences between the optical reference satellite and the optical comparison satellite. The closer the slope is to 1 and the smaller the intercept, the higher the consistency between the two.

Images