A method for establishing a dataset for multi-scale remote sensing calibration and authenticity verification site screening

Abstract

The present invention discloses a data set establishment method for multi-scale remote sensing calibration and authenticity verification site screening, and belongs to the field of data set establishment technology. The present invention includes collecting atmospheric data of the target area in the past five years; dividing the atmospheric data into a number of block areas after pre-processing; calculating the pixel values ​​in the block areas, screening out areas with stable atmospheric conditions as preparation areas; downloading satellite image data of the preparation areas, and resampling the satellite image data according to a five-layer resampling method; processing the resampled satellite images in blocks according to specific needs, and calculating specific indicators of each block, screening out specific types of ground objects and sites according to set thresholds, and organizing site data to establish a multi-scale calibration and authenticity verification verification data set; selecting areas with stable surface underlying surfaces from the data set as alternative references for radiation calibration sites.

Description

Technical Field

[0001] The present invention relates to the technical field of data set establishment, and in particular to a data set establishment method for multi-scale remote sensing calibration and authenticity verification site screening. Background Art

[0002] With the development of China's space industry and the maturity of commercial satellite technology, the market demand for high-precision satellite data products has increased significantly. However, satellites are affected by the harsh environment of space for a long time, and the payloads will inevitably age and be damaged. In this context, in order to ensure the accuracy of satellites and the quality of products, it is necessary to regularly perform radiation calibration on the payloads and conduct authenticity inspections on data products to ensure the accuracy of satellites and the quality of products.

[0003] The site calibration method is the most mature and widely used calibration method. The selection of an ideal radiation calibration site plays a decisive role in the calibration accuracy of satellite payloads. However, there are some problems in the current selection of calibration sites: the traditional manual search for radiation calibration sites costs a lot of human resources, the site selection indicators are vague, and there are no clear quantitative standards and operating procedures. These problems have greatly affected the rationality of site selection.

[0004] Authenticity verification is an important part of verifying the quality of satellite data. In order to ensure the accuracy of the data, long-term verification of remote sensing products is required. Using the cross-verification method to cross-verify with data from different satellites is a way to achieve long-term series verification of remote sensing data products, which can save a lot of time and labor costs. However, this method requires a large amount of long-term, multi-type ground image data for reference verification, which takes a lot of time to collect and screen the verification data, and the process is cumbersome and complicated, which greatly affects the progress of the experiment and the reliability of the verification results. Summary of the invention

[0005] The purpose of the present invention is to provide a method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening to solve the problems raised in the above-mentioned background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening, the method comprising the following steps:

[0008] S100. Collect atmospheric data of the target area for m years, where m is a positive integer; pre-process the atmospheric data and divide it into several sub-areas; the atmospheric data includes cloud cover, aerosol and water vapor data;

[0009] S200. Calculate the pixel values ​​in the sub-block area, and determine whether the atmospheric conditions in each sub-block area are suitable as calibration and authenticity test sites according to the set atmospheric threshold index, select areas with stable atmospheric conditions, and use them as preparation areas;

[0010] S300. Download satellite image data of various resolution scales within a time span of one year in the preparation area, and perform data preprocessing on all downloaded image data; and resample the satellite image data of various resolution scales according to a five-layer resampling method;

[0011] S400. After resampling, the satellite images of different resolution scales are divided into blocks according to specific needs to obtain a number of sub-block areas; and the specific indicators of each sub-block area are calculated, and the specific type of ground objects and sites are screened out according to the set threshold, and the site data are sorted to establish a multi-scale calibration and authenticity verification data set;

[0012] S500. Select an area with stable underlying surface from the data set as an alternative reference for radiation calibration site.

[0013] Furthermore, step S100 includes:

[0014] S101. Download m years of atmospheric data by using data products of MODIS satellite, and obtain atmospheric data of fixed date data of each month in the target area as a period; the data products of MODIS satellite include Mod04_12 (aerosol thickness), Mod05_12 (water vapor), and Mod06_12 (cloud amount); the atmospheric data include cloud amount, aerosol optical thickness and water vapor data; the target area is selected from 90°E-110°E, 30°N-45°N, and the range includes the three provinces of Inner Mongolia, Gansu and Qinghai;

[0015] S102. Use IDL and MCTK to perform batch projection conversion and geometric correction on the downloaded atmospheric data; IDL (Interactive Data Language) is a programming language specifically used for scientific data analysis and visualization, providing a rich library of functions and tools suitable for processing, analyzing and visualizing various types of scientific data; MCTK (MODIS Conversion Toolkit) is an IDL toolkit for processing MODIS satellite data, providing functions such as projection conversion, geometric correction, cloud mask, data clipping, etc., to facilitate users to pre-process and analyze MODIS data;

[0016] S103. Splice and crop the data after projection conversion and geometric correction to complete data preprocessing; after preprocessing, divide the data into several sub-areas. If the size of the sub-area is 20km*20km, since the resolution of Mod04_l2 is 10km*10km, the image is divided into 2*2 small areas according to the pixel; the resolutions of Mod05_l2 and Mod06_l2 are both 1km*1km, so the image is divided into 20*20 small areas according to the pixel.

[0017] Further, step S200 includes:

[0018] S201. Extract the values ​​of all pixels in the block area to form a pixel value set DN, where DN = {DN1, DN2, ..., DN n}, where DN1, DN2, ..., DN n Represents the value of all pixels in a block area, n is the number of all pixels in a block area; and the value set of pixels in each block area is processed by averaging to obtain the mean data Based on this data, the annual average data of the block area is calculated according to the year;

[0019] S202. Set the annual average atmospheric threshold standard to Q, where the threshold standard Q is cloud cover <40%, aerosol optical thickness <0.2, and water vapor content <1.5. If the data of block area A for five consecutive years meet the threshold standard, it means that the atmospheric conditions of block area A are stable, and block area A is used as a reserve area. The atmospheric threshold standard is not static. The atmospheric thresholds of calibration and authenticity test sites in different regions should be adjusted according to the climatic conditions of the region. For example, the Taklimakan calibration site is located in the northwest desert area of ​​my country. The atmosphere is greatly affected by the desert climate, resulting in a higher aerosol content than that in non-desert areas.

[0020] Furthermore, step S300 includes:

[0021] S301. Download satellite image data of various resolution scales within a one-year time span of the preparation area, and perform data preprocessing on all downloaded image data;

[0022] S302. Resample satellite image data of various resolution scales according to a five-layer resampling method; the five-layer resampling method is:

[0023] like Figure 2 As shown, firstly, a meter level, a*10 meter level, a*10 2 Meter level, a*10 3 Meter level, a*10 4Meter level is the five-layer standard; determine the original resolution of each satellite image, such as 1m, 10m, 100m, 1000m, 10000m. Determine the target resolution of each satellite image, such as 2m, 20m, 200m, 2000m, 20000m. At the same level, large pixels can correspond to small pixels; for example, 2*2 pixels with a resolution of 1m correspond to a pixel with a resolution of 2m, 4*4 pixels with a resolution of 1m correspond to a pixel with a resolution of 4m, and so on. In the first two levels, the first two resolution pixels can arbitrarily correspond to all resolutions of the upper level; for example, the 10m and 20m pixel resolutions at the second level can arbitrarily correspond to the pixel resolutions of all levels from the third to the fifth level, and so on. At the last two levels, the last two resolution pixels can arbitrarily correspond to all resolutions within the range of the resolution oblique line and the straight line box of the level, except for the fifth level; the fifth level is specially treated. For example, a 40m resolution can correspond to the straight line connecting 1000m and 8m and the straight line connecting 10m and 80m in the second layer, and all resolutions within the range of the two straight lines (except the fifth layer). If the maximum resolution of the fifth layer is 10km, it can correspond to all resolutions except 8000m and 800m.

[0024] Furthermore, step S400 includes:

[0025] S401. Divide the images of different resolutions into blocks of a km*a km in size to obtain several sub-block areas;

[0026] S402. Calculate the coefficient of variation CV, normalized vegetation index NDVI, and normalized water index NDWI of each sub-block area. The calculation formula is as follows:

[0027]

[0028] Among them, DN i is the pixel value of the i-th pixel in the sub-block area, i = 1, 2…n; n is the number of pixels in the sub-block area; It is the mean value data of the pixel values ​​in the sub-block area;

[0029]

[0030] Among them, NIR is the reflectivity of the pixel in the near-infrared band, and RED is the reflectivity of the pixel in the infrared band;

[0031]

[0032] Among them, GREEN is the reflectance of the green band of the pixel, and NIR is the reflectance of the near-infrared band of the pixel;

[0033] S403. Determine the calibration and authenticity test sites of different types of land objects according to the set threshold conditions; if the threshold condition P1 is met, it is recorded as the verification data of the surface type being desert and Gobi; if the threshold condition P2 is met, it is recorded as the verification data of the surface type being vegetation; if the threshold condition P3 is met, it is recorded as the verification data of the surface type being water body; and the threshold indicators for distinguishing different types of land objects are not fixed, and need to be adjusted in combination with specific needs and actual conditions. At the same time, specific parameters can also be added or replaced to meet more different division needs;

[0034] S404. Organize the above verification data and establish a multi-scale calibration and authenticity verification data set with a time span of one year.

[0035] Furthermore, step S500 includes:

[0036] Since the verification data set for multi-scale calibration and authenticity verification is divided into three types of verification data of different surface types, the areas with stable surface underlying surfaces, Gobi, desert and water bodies are selected as alternative references for radiation calibration sites. Vegetation is easily affected by seasons and growth cycles, and the surface radiation characteristics are prone to change, so they will not be selected as alternative references; it is judged whether the number of images of the screened underlying surface area exceeds two-thirds of the total number of images. If so, the area is determined as an alternative calibration site. If not, the area is removed.

[0037] Compared with the prior art, the beneficial effects achieved by the present invention are: the constructed multi-scale verification data set provides rich multi-resolution verification samples for the verification of satellite calibration coefficients and the authenticity inspection of products, which can more comprehensively and comprehensively verify the accuracy of the calibration results of the satellite to be calibrated, and provide a large amount of data reference for the cross-calibration verification of satellites. To a certain extent, it solves the problem of insufficient reference data for long-time series cross-calibration, and provides a certain number of data samples for satellite long-time series calibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0039] Figure 1 It is a flow chart of a method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening of the present invention;

[0040] Figure 2 It is a schematic diagram of a five-layer resampling method of a data set establishment method for multi-scale remote sensing calibration and authenticity verification site screening of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0043] A method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening, the method comprising the following steps:

[0044] S100. Collect atmospheric data of the target area for m years, where m is a positive integer; pre-process the atmospheric data and divide it into several sub-areas; the atmospheric data includes cloud cover, aerosol and water vapor data;

[0045] S200. Calculate the pixel values ​​in the sub-block area, and determine whether the atmospheric conditions in each sub-block area are suitable as calibration and authenticity test sites according to the set atmospheric threshold index, select areas with stable atmospheric conditions, and use them as preparation areas;

[0046] S300. Download satellite image data of various resolution scales within a time span of one year in the preparation area, and perform data preprocessing on all downloaded image data; and resample the satellite image data of various resolution scales according to a five-layer resampling method;

[0047] S400. After resampling, the satellite images of different resolution scales are divided into blocks according to specific needs to obtain a number of sub-block areas; and the specific indicators of each sub-block area are calculated, and the specific type of ground objects and sites are screened out according to the set threshold, and the site data are sorted to establish a multi-scale calibration and authenticity verification data set;

[0048] S500. Select an area with stable underlying surface from the data set as an alternative reference for radiation calibration site.

[0049] Step S100 includes:

[0050] S101. Download m years of atmospheric data by using data products of MODIS satellite, and obtain atmospheric data of fixed date data of each month in the target area as a period; the data products of MODIS satellite include Mod04_12 (aerosol thickness), Mod05_12 (water vapor), and Mod06_12 (cloud amount); the atmospheric data include cloud amount, aerosol optical thickness and water vapor data; the target area is selected from 90°E-110°E, 30°N-45°N, and the range includes the three provinces of Inner Mongolia, Gansu and Qinghai;

[0051] S102. Use IDL and MCTK to perform batch projection conversion and geometric correction on the downloaded atmospheric data; IDL (Interactive Data Language) is a programming language specifically used for scientific data analysis and visualization, providing a rich library of functions and tools suitable for processing, analyzing and visualizing various types of scientific data; MCTK (MODIS Conversion Toolkit) is an IDL toolkit for processing MODIS satellite data, providing functions such as projection conversion, geometric correction, cloud mask, data clipping, etc., to facilitate users to pre-process and analyze MODIS data;

[0052] S103. Splice and crop the data after projection conversion and geometric correction to complete data preprocessing; after preprocessing, divide the data into several sub-areas. If the size of the sub-area is 20km*20km, since the resolution of Mod04_l2 is 10km*10km, the image is divided into 2*2 small areas according to the pixel; the resolutions of Mod05_l2 and Mod06_l2 are both 1km*1km, so the image is divided into 20*20 small areas according to the pixel.

[0053] Step S200 includes:

[0054] S201. Extract the values ​​of all pixels in the block area to form a pixel value set DN, where DN = {DN1, DN2, ..., DN n}, where DN1, DN2, ..., DN n Represents the value of all pixels in a block area, n is the number of all pixels in a block area; and the value set of pixels in each block area is processed by averaging to obtain the mean data Based on this data, the annual average data of the block area is calculated according to the year;

[0055] S202. Set the annual average atmospheric threshold standard to Q, where the threshold standard Q is cloud cover <40%, aerosol optical thickness <0.2, and water vapor content <1.5. If the data of block area A for five consecutive years meet the threshold standard, it means that the atmospheric conditions of block area A are stable, and block area A is used as a reserve area. The atmospheric threshold standard is not static. The atmospheric thresholds of calibration and authenticity test sites in different regions should be adjusted according to the climatic conditions of the region. For example, the Taklimakan calibration site is located in the northwest desert area of ​​my country. The atmosphere is greatly affected by the desert climate, resulting in a higher aerosol content than that in non-desert areas.

[0056] Step S300 includes:

[0057] S301. Download satellite image data of various resolution scales within a one-year time span of the preparation area, and perform data preprocessing on all downloaded image data;

[0058] S302. Resample satellite image data of various resolution scales according to a five-layer resampling method; the five-layer resampling method is:

[0059] like Figure 2 As shown, firstly, a meter level, a*10 meter level, a*10 2 Meter level, a*10 3 Meter level, a*10 4 Meter level is the five-layer standard; determine the original resolution of each satellite image, such as 1m, 10m, 100m, 1000m, 10000m. Determine the target resolution of each satellite image, such as 2m, 20m, 200m, 2000m, 20000m. At the same level, large pixels can correspond to small pixels; for example, 2*2 pixels with a resolution of 1m correspond to a pixel with a resolution of 2m, 4*4 pixels with a resolution of 1m correspond to a pixel with a resolution of 4m, and so on. In the first two levels, the first two resolution pixels can arbitrarily correspond to all resolutions of the upper level; for example, the 10m and 20m pixel resolutions at the second level can arbitrarily correspond to the pixel resolutions of all levels from the third to the fifth level, and so on. At the last two levels, the last two resolution pixels can arbitrarily correspond to all resolutions within the range of the resolution oblique line and the straight line box of the level, except for the fifth level; the fifth level is specially treated. For example, a 40m resolution can correspond to the straight line connecting 1000m and 8m and the straight line connecting 10m and 80m in the second layer, and all resolutions within the range of the two straight lines (except the fifth layer). If the maximum resolution of the fifth layer is 10km, it can correspond to all resolutions except 8000m and 800m.

[0060] Step S400 includes:

[0061] S401. Divide the images of different resolutions into blocks of a km*a km in size to obtain several sub-block areas;

[0062] S402. Calculate the coefficient of variation CV, normalized vegetation index NDVI, and normalized water index NDWI of each sub-block area. The calculation formula is as follows:

[0063]

[0064] Among them, DN i is the pixel value of the i-th pixel in the sub-block area, i = 1, 2…n; n is the number of pixels in the sub-block area; It is the mean value data of the pixel values ​​in the sub-block area;

[0065]

[0066] Among them, NIR is the reflectivity of the pixel in the near-infrared band, and RED is the reflectivity of the pixel in the infrared band;

[0067]

[0068] Among them, GREEN is the reflectance of the green band of the pixel, and NIR is the reflectance of the near-infrared band of the pixel;

[0069] S403. Determine the calibration and authenticity test sites of different types of land objects according to the set threshold conditions; if the threshold condition P1 is met, it is recorded as the verification data of the surface type being desert and Gobi; if the threshold condition P2 is met, it is recorded as the verification data of the surface type being vegetation; if the threshold condition P3 is met, it is recorded as the verification data of the surface type being water body; and the threshold indicators for distinguishing different types of land objects are not fixed, and need to be adjusted in combination with specific needs and actual conditions. At the same time, specific parameters can also be added or replaced to meet more different division needs;

[0070] S404. Organize the above verification data and establish a multi-scale calibration and authenticity verification data set with a time span of one year.

[0071] Step S500 includes:

[0072] Since the verification data set for multi-scale calibration and authenticity verification is divided into three types of verification data of different surface types, the areas with stable surface underlying surfaces, Gobi, desert and water bodies are selected as alternative references for radiation calibration sites. Vegetation is easily affected by seasons and growth cycles, and the surface radiation characteristics are prone to change, so they will not be selected as alternative references; it is judged whether the number of images of the screened underlying surface area exceeds two-thirds of the total number of images. If so, the area is determined as an alternative calibration site. If not, the area is removed.

[0073] In this embodiment:

[0074] The coefficient of variation (CV), normalized difference vegetation index (NDVI), and normalized water index (NDWI) of each small area are calculated, and the calibration and authenticity test sites of different land object types are determined according to the set threshold conditions, taking the MOD02HKM apparent reflectance data as an example. When the block area meets the threshold condition P1, NDWI < 0, NDVI < 0.2, and CV < 0.03 are recorded as verification data of the surface type of desert and Gobi; when the block area meets the threshold condition P2, 0 <NDWI<0.2、NDVI> When the threshold condition P3 is met, NDWI>0.2, NDVI<0.2, and CV<0.03, it is recorded as the verification data that the surface type is vegetation; when the block area meets the threshold condition P3, NDWI>0.2, NDVI<0.2, and CV<0.02, it is recorded as the verification data that the surface type is water.

[0075] Moreover, the threshold indicators for distinguishing different types of land features are not static and need to be adjusted based on specific needs and actual conditions. At the same time, specific parameters can also be added or replaced to meet more different classification needs.

[0076] Determine whether the number of underlying surface area images screened out exceeds two-thirds of the total number of images. If so, determine that the area is an alternative calibration site. If not, remove the area. For example, there are 300 MOD02HKM images downloaded for one year. When screening areas with Gobi as the underlying surface type, there are 200 block images of the same area in Golmud that meet the threshold index. Then the area (Golmud) is selected as an alternative radiation calibration site.

[0077] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0078] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening, characterized in that: The method comprises the following steps: S100. Collect atmospheric data of the target area for m years, where m is a positive integer; pre-process the atmospheric data and divide it into several sub-areas; the atmospheric data includes cloud cover, aerosol and water vapor data; S200. Calculate the pixel values ​​in the sub-block area, and determine whether the atmospheric conditions in each sub-block area are suitable as calibration and authenticity test sites according to the set atmospheric threshold index, select areas with stable atmospheric conditions, and use them as preparation areas; S300. Download satellite image data of various resolution scales within a time span of one year in the preparation area, and perform data preprocessing on all downloaded image data; and resample the satellite image data of various resolution scales according to a five-layer resampling method; S400. After resampling, the satellite images of different resolution scales are divided into blocks according to specific needs to obtain a number of sub-block areas; and the specific indicators of each sub-block area are calculated, and the specific type of ground objects and sites are screened out according to the set threshold, and the site data are sorted to establish a multi-scale calibration and authenticity verification data set; S500. Selecting a region with a stable underlying surface from the data set as a candidate reference for radiation calibration site; The step S400 includes: S401. Divide the images of different resolutions into blocks of a km*a km in size to obtain several sub-block areas; S402. Calculate the coefficient of variation CV, normalized vegetation index NDVI, and normalized water index NDWI of each sub-block area. The calculation formula is as follows: Among them, DN i is the pixel value of the i-th pixel in the sub-block area, i = 1, 2…n; n is the number of pixels in the sub-block area; It is the mean value data of the pixel values ​​in the sub-block area; Among them, NIR is the reflectivity of the pixel in the near-infrared band, and RED is the reflectivity of the pixel in the infrared band; Among them, GREEN is the reflectance of the green band of the pixel, and NIR is the reflectance of the near-infrared band of the pixel; S403. According to the set threshold conditions, the calibration and authenticity test sites of different types of land objects are determined; if the threshold condition P1 is met, the verification data of the surface type is desert and Gobi is recorded; if the threshold condition P2 is met, the verification data of the surface type is vegetation is recorded; if the threshold condition P3 is met, the verification data of the surface type is water is recorded; S404. Organize the above verification data and establish a multi-scale calibration and authenticity verification data set with a time span of one year.

2. The method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening according to claim 1, characterized in that: The step S100 includes: S101. Download m years of atmospheric data by using data products from MODIS satellites, and obtain atmospheric data of the target area with fixed monthly date data as a period; the atmospheric data includes cloud cover, aerosol optical thickness and water vapor data; S102. Use IDL and MCTK to perform batch projection conversion and geometric correction on the downloaded atmospheric data; S103. Splice and crop the data after projection conversion and geometric correction to complete data preprocessing; after the preprocessing is completed, divide the data into several block areas.

3. The method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening according to claim 2, characterized in that: The step S200 includes: S201. Extract the values ​​of all pixels in the block area to form a pixel value set DN, where DN = {DN1, DN2, ..., DN n }, where DN1, DN2, ..., DN n Represents the value of all pixels in a block area, n is the number of all pixels in a block area; and the value set of pixels in each block area is processed by averaging to obtain the mean data Based on this data, the annual average data of the block area is calculated according to the year; S202. Set the annual average atmospheric threshold standard to Q. If the data of block area A for five consecutive years meet the threshold standard, it means that the atmospheric conditions in block area A are stable, and block area A is used as a reserve area.

4. The method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening according to claim 3, characterized in that: The step S300 includes: S301. Download satellite image data of various resolution scales within a one-year time span of the preparation area, and perform data preprocessing on all downloaded image data; S302. Resample satellite image data of various resolution scales according to a five-layer resampling method; the five-layer resampling method is: First, we use a meter level, a*10 meter level, a*10 2 Meter level, a*10 3 Meter level, a*10 4 The meter level is the five-layer standard; determine the original resolution of each satellite image, and determine the target resolution of each satellite image; at the same level, large pixels can correspond to small pixels; in the first two levels, the first two pixel resolutions can correspond to any resolution of the upper level; in the last two levels, the last two pixel resolutions can correspond to any resolution within the range of the resolution's oblique line and the level's straight line selection except the fifth level; the fifth level is specially treated.

5. The method for establishing a data set for multi-scale remote sensing calibration and authenticity verification site screening according to claim 4, characterized in that: The step S500 includes: Since the verification data set for multi-scale calibration and authenticity verification includes verification data of three different surface types, the Gobi, desert and water bodies with stable surface underlying surfaces are selected as alternative references for radiation calibration sites. It is judged whether the number of images of the screened underlying surface area exceeds two-thirds of the total number of images. If so, the area is determined as an alternative calibration site. If not, the area is removed.

Images