A method for radiometric calibration of satellite optical remote sensors based on the moon

Abstract

The present invention provides a method for radiometric calibration of satellite optical remote sensors based on the moon, including the steps of: performing preprocessing such as dark current subtraction and flat field correction on lunar images, and extracting the total gray value and total number of pixels of the moon; calculating the equivalent disk reflectance of corresponding lunar observation conditions using the ROLO lunar radiation model; calculating the average reflectance of the lunar front area of the M<supgt;3< / supgt> reflectance image and performing smoothing and resampling processing; using the average reflectance of the M<supgt;3< / supgt> lunar image to smooth the ROLO model values; performing interpolation and resampling processing on the smoothed ROLO model values, establishing a fitting relationship between the lunar observation image and the smoothed model values, and obtaining the absolute radiometric calibration coefficient. The present invention makes full use of the good characteristics of the moon as a radiometric calibration source without atmospheric interference and stability, and greatly saves the manpower and material resources for on-orbit radiometric calibration of satellite optical remote sensors.

Description

Technical Field

[0001] The present invention relates to an on-orbit radiation calibration method for a satellite optical remote sensor, belonging to the technical field of remote sensing satellite calibration. Background Art

[0002] During the on-orbit operation of an optical remote sensing satellite, it will be affected by various radiations and electromagnetic interferences, and the satellite payload devices will age. Quantitative remote sensing applications have put forward strict requirements for the high precision and stability of remote sensing data products, and it is necessary to calibrate the radiation response of the satellite payload with high frequency and high precision through on-orbit radiation calibration. Currently, the commonly used on-orbit absolute radiation calibration mainly includes on-board calibration and field calibration. Both of them have some drawbacks: 1) On-board calibration includes in-satellite calibration and out-of-satellite calibration; among them, in-satellite calibration uses a tungsten lamp as the calibration source, but the calibration lamp spectrum is quite different from the solar spectrum, resulting in a large uncertainty in spectral matching; out-of-satellite calibration uses the sun as the radiation source, but the diffuser plate will age after being exposed to the strong radiation of outer space for a long time, affecting the calibration accuracy; 2) Field calibration synchronously measures the reflectivity of the ground calibration field when the satellite passes by. This method is affected by the atmosphere, and for long-term high-frequency calibration, this calibration method has high human and material costs.

[0003] The moon, as a radiation source for satellite on-orbit calibration, has many advantages compared with other calibration methods: ① The moon has no atmosphere, providing a calibration source for the satellite without atmospheric interference; ② The lunar surface has a long-term stable and smooth reflectivity spectrum; ③ The radiation energy of the moon is within the dynamic response range of most payloads; ④ The background around the moon is a black sky. The lunar image has clear brightness boundaries and high contrast, and can well reflect calibration parameters such as dark current, flat field, and readout noise in radiation calibration, which are often unavailable in earth observation; ⑤ The moon calibration has low cost, and there are similar observation geometries every month, which is suitable for carrying out calibration regularly with high frequency.

[0004] In 1996, the United States Geological Survey (USGS) launched the Robotic Lunar Observatory (ROLO) program. Based on ground-based telescopes, lunar radiance images from near lunar eclipse to full moon were collected for 6 years (1997 - 2003), and a lunar radiation model was established, which can provide the lunar spectrum of 32 bands from 350 nm to 2500 nm. In 2008, the Moon Mineralogy Mapper (M ,

[0005] , 3 , ) carried by India's Chandrayaan-1 conducted multi-cycle observations of the moon, covering 85 bands, without being affected by atmospheric attenuation, and being more able to reflect the true entire lunar spectrum. These long-term and large-scale observations provide model and data support for lunar calibration. Summary of the Invention

[0005] The present invention aims to solve the problems in the prior art and provide a lunar-based satellite optical remote sensor radiation calibration method. The method uses the Moon to calibrate the satellite optical remote sensor without using onboard calibration equipment, without requiring field calibration, and without considering weather factors and atmospheric influences. 3 The smooth ROLO model value of the average reflectivity of the lunar image provides high-frequency and high-precision absolute radiometric calibration for the satellite.

[0006] To achieve the above object, the technical solution provided by the present invention is:

[0007] A lunar-based satellite optical remote sensor radiation calibration method, characterized by comprising the following steps:

[0008] Step 1: Select the lunar image in the range of [-90°, 90°] for dark current subtraction and flat field correction preprocessing, and extract the total grayscale value T of the moon. DN and the total number of pixels T pixel ;

[0009] Step 2: Based on the lunar observation geometry, use the ROLO lunar radiation model to calculate the equivalent disk reflectivity A at the standard distance. ROLO ;

[0010] Step 3, calculate the lunar mineralogy mapper M 3 The reflectivity image of the lunar front area is averaged, and the cubic spline interpolation method is used to smooth it to obtain M 3 Average reflectivity of the lunar front image

[0011] Step 4, for M 3 Average reflectivity of the lunar front image Resampling is performed to obtain the wavelength λ of the ROLO lunar radiation model k M in the corresponding position 3 Average reflectivity of the lunar front image Then use The equivalent disk reflectivity A obtained in step 2 ROLO The linear relationship obtained by fitting is the equivalent disk reflectivity A ROLO Smoothing is performed to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k ;

[0012] Step 5: Smooth the ROLO lunar radiation model value A ROLO,k Perform interpolation and resampling to obtain the satellite optical remote sensor wavelength λ i ROLO lunar radiation model value A at the corresponding position i , the least squares fitting method is used to establish the lunar image and ROLO lunar radiation model value A iThe relationship between them is used to obtain the absolute radiometric calibration coefficient gains and bias.

[0013] Further, step 3 includes the following steps:

[0014] S3.1, traverse the M 3 reflectance image pixels. If the lunar longitude of a certain pixel is outside the range of [-90°, 90°], then mask this pixel to obtain the lunar near-side reflectance image img1 with mask marks;

[0015] S3.2, traverse the pixels in the lunar near-side reflectance image img1 without mask marks. If the pixel value is 0, then mask this pixel to obtain the lunar reflectance image img2 with mask marks;

[0016] S3.3, traverse the pixels in the lunar reflectance image img2 with mask marks that do not have mask marks, calculate the mean value of the unmarked pixels in each band reflectance image, and smooth it using the cubic spline interpolation method to obtain the M 3 average reflectance of the lunar near-side image

[0017] Further, step 4 includes the following steps:

[0018] S4.1, resample the M 3 average reflectance of the lunar plane to the wavelength position λ of the ROLO model k , to obtain the M k corresponding to the wavelength λ of the ROLO lunar radiation model 3 average reflectance of the lunar near-side image Specifically, it is given by the formula:

[0019]

[0020] where R k,s (λ) is the spectral response function of the ROLO lunar radiation model, k represents the spectral band of the ROLO lunar radiation model, s represents the wavelength in the spectral response function of the ROLO lunar radiation model, λ1 is the lower limit corresponding to the wavelength of the ROLO model spectral response function, and λ2 is the upper limit corresponding to the wavelength of the ROLO model spectral response function;

[0021] S4.2, perform a linear fit on the equivalent disk reflectance A ROLO of the ROLO lunar radiation model and the M k corresponding to the wavelength λ of the ROLO lunar radiation model 3 average reflectance of the lunar near-side image and apply the least squares method to obtain the smoothed value A of the ROLO lunar radiation modelROLO,k , the fitting formula is as follows:

[0022]

[0023] Among them, a and b are coefficients obtained by linear fitting.

[0024] Furthermore, step 5 includes the following steps:

[0025] S5.1, Use the cubic spline interpolation method to interpolate the smoothed value A of the ROLO model ROLO,k to obtain the ROLO model value A corresponding to the wavelength of the spectral response function of the satellite remote sensor to be absolutely radiometrically calibrated. ROLO,j , resample it using the resampling formula, and then obtain the ROLO lunar radiation model value A at the corresponding position of the central wavelength λ of the satellite optical remote sensor i , and the resampling formula is as follows: i , the resampling formula is as follows:

[0026]

[0027] Among them, R i,j (λ) is the spectral response function of the satellite remote sensor, i represents the spectral band of the satellite remote sensor, j represents the wavelength in the spectral response function of the satellite remote sensor, and λ m is the lower limit corresponding to the wavelength of the spectral response function of the satellite remote sensor, and λ n is the upper limit corresponding to the wavelength of the spectral response function of the satellite remote sensor;

[0028] S5.2, Use the resampling formula to perform resampling calculation on the solar irradiance to obtain the solar irradiance E at the corresponding position of the central wavelength λ of the satellite optical remote sensor i , and the calculation formula is; i , the calculation formula is;

[0029]

[0030] Among them, E(λ) is the solar irradiance;

[0031] S5.3, Calculate the lunar equivalent disk irradiance model value I at the standard distance i , and the calculation formula is:

[0032] Among them, Ω M is the lunar solid angle, and Ω M = 6.4177×10 5 sr;

[0033] S5.4, Calculate the lunar equivalent disk irradiance model value I' at the measured distance i , and the calculation formula is:

[0034]

[0035] Among them, D S-M is the distance between the sun and the moon, D V-M is the distance between the satellite optical sensor and the moon, ME is the average distance between the moon and the earth, AU is the astronomical unit, ME = 3.844×10 8 m, 1AU = 1.49598×10 11 m;

[0036] S5.5. Apply the least squares fitting method to construct the linear relationship between the measured lunar image of each band and the ROLO lunar radiation model value A i at different lunar phase angles, and obtain the absolute radiometric calibration coefficient gain gains and bias bias; for the lunar observation image of the satellite optical remote sensor, the formula for calculating the lunar equivalent disk irradiance I' i,view at the measured distance is:

[0037]

[0038] Among them, L is the radiance, DN is the gray value, where Ω p is the solid angle of a single pixel of the satellite optical image;

[0039] Theoretically, the lunar equivalent disk irradiance model value I' i at the measured distance is equal to the lunar equivalent disk irradiance I' i,view measured by the satellite, so the fitting formula is:

[0040]

[0041] The present invention also protects a satellite optical remote sensor radiometric calibration system based on the moon, including:

[0042] An extraction module for performing dark current subtraction and flat field correction preprocessing on the lunar image in the lunar phase angle range of [-90°, 90°], and extracting the total lunar gray value T DN and the total number of pixels T pixel ;

[0043] An equivalent disk reflectance A ROLO acquisition module: for calculating the equivalent disk reflectance A ROLO at the standard distance by using the ROLO lunar radiation model according to the lunar observation geometric conditions;

[0044] ]A processing module for calculating the mean value of the lunar front area of the reflectance image of M 3 and performing smoothing processing by using the cubic spline interpolation method to obtain the average reflectance of the lunar front image M 3 ; For M 3 Average reflectivity of the lunar near - side image Perform resampling processing to obtain the average reflectivity of the lunar near - side image at the corresponding position of the wavelength λ of the ROLO lunar radiation model k of M 3 Average reflectivity of the lunar near - side image Furthermore, utilize and the linear fitting relationship with the equivalent disk reflectivity A ROLO to smooth the equivalent disk reflectivity A ROLO to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k ; For interpolating and resampling the smoothed value A of the ROLO lunar radiation model ROLO,k to obtain the value A of the ROLO lunar radiation model at the corresponding position of the central wavelength λ of the satellite remote sensor i ; i ;

[0045] Radiometric calibration module, which is used to establish the relationship between the lunar image and the value A of the ROLO lunar radiation model by using the least - squares fitting method, and obtain the absolute radiometric calibration coefficient gains and bias i between them

[0046] The present invention also protects an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above - mentioned satellite radiometric calibration method based on the moon is implemented

[0047] The present invention also protects a computer - readable storage medium storing a computer program, and the computer program enables a computer to execute the satellite optical remote sensor radiometric calibration method based on the moon as described in the above claims

[0048] A satellite optical remote sensor radiometric calibration method based on the moon provided by the present invention smooths the ROLO model through the average reflectivity of the M 3 image, and uses the least - squares fitting method to establish the relationship between the lunar observation image of the satellite optical remote sensor and the ROLO model value, and calibrates the absolute radiometric calibration coefficient of each band of the satellite optical remote sensor

[0049] The present invention solves the deficiencies of the traditional calibration method in the technical background, such as high cost, complex operation, and easy decay of satellite devices. Specifically, compared with the existing methods, the present invention has the following beneficial effects

[0050] (1) It does not require on - satellite calibration equipment and does not require field calibration, and can perform high - frequency and high - precision on - orbit absolute radiometric calibration of satellite optical remote sensors in the Earth orbit for a long time

[0051] (2) Considering that the ROLO model value cannot fully obtain the smooth spectral shape of the moon, the average reflectance of the lunar near side region of the M 3 image is used to smooth the ROLO model value, which can obtain a more scientific and smooth lunar spectrum and improve the reliability of the lunar calibration result.

[0052] Therefore, the present invention makes full use of the good characteristics of the moon as a radiation calibration source without atmospheric interference and stability, and greatly saves the manpower and material resources for on-orbit radiation calibration of satellite optical remote sensors. Description of the Drawings

[0053] Figure 1 is a flowchart of an embodiment of the present invention.

[0054] Figure 2 is the average reflectance of the smoothed M 3 lunar near side image.

[0055] Figure 3 is the ROLO model value smoothed in an embodiment of the present invention. Detailed Description of the Invention

[0056] The present invention will now be further described in detail with reference to the accompanying drawings.

[0057] The present invention provides a method for radiometric calibration of satellite optical remote sensors based on the moon, as Figure 1 shown, mainly including the following steps:

[0058] Step 1: Select lunar images with a lunar phase angle of [-90°, 90°] for preprocessing such as dark current subtraction and flat field correction, and extract the total gray value T DN and the total number of pixels T pixel ;

[0059] Step 2: Calculate the ROLO lunar radiation model value according to the lunar observation geometric conditions to obtain the equivalent disk reflectance A ROLO at the standard distance;

[0060] Step 3: Calculate the mean value of the reflectance image of the lunar near side region of M 3 , where M 3 is the lunar mineral mapper, and further smooth it by cubic spline interpolation to obtain the average reflectance 3 of the lunar near side image of M Specifically, it includes:

[0061] Step 3.1: Traverse M 3Reflectivity image pixels, masking the pixels outside the range of lunar longitude [-90°, 90°]. Since the payload of the lunar orbiting spacecraft can photograph the near and far sides of the moon, while the satellite can only observe the near side of the moon, a near-side lunar reflectivity image img1 with masking marks is obtained;

[0062] Step 3.2, traverse the pixels in the near-side lunar reflectivity image img1 that do not have masking marks. If the pixel value is 0, then mask the pixel. Since the lunar images obtained by the payload of the lunar orbiting spacecraft may have incomplete coverage, a lunar reflectivity image img2 with masking marks is obtained;

[0063] Step 3.3, traverse the pixels in the lunar reflectivity image img2 with masking marks that do not have masking marks, calculate the mean value of the pixels without masking marks in each band reflectivity image, and smooth it using cubic spline interpolation to obtain M 3 Average reflectivity of the near-side lunar image

[0064] Through the above processing, M as shown in Figure 2 is obtained 3 Average reflectivity A of the near-side lunar image M3 .

[0065] Step 4, resample M 3 Average reflectivity of the near-side lunar image and smooth the equivalent disk reflectivity A of the ROLO lunar radiation model obtained in Step 2 with this, ROLO to obtain the smoothed value A of the ROLO lunar radiation model. ROLO,k . Specifically, it includes:

[0066] Step 4.1, resample M 3 Average reflectivity of the near-side lunar image to the wavelength position λ of the ROLO lunar radiation model k , and obtain the M at the corresponding position of the wavelength λ of the ROLO lunar radiation model k . The resampling calculation formula is as follows: 3 Average reflectivity of the near-side lunar image The resampling calculation formula is as follows:

[0067]

[0068] where R k,s (λ) is the spectral response function of the ROLO lunar radiation model, k represents the spectral band of the ROLO lunar radiation model, s represents the wavelength in the spectral response function of the ROLO lunar radiation model, λ1 is the lower limit corresponding to the wavelength of the ROLO model spectral response function, and λ2 is the upper limit corresponding to the wavelength of the ROLO model spectral response function.

[0069] Step 4.2, for the equivalent disk reflectance A of the ROLO lunar radiation model ROLO and the wavelength λ of the ROLO lunar radiation model k at the corresponding position of M 3 the average reflectance of the lunar near side image perform a linear fit, and use the least squares fitting method to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k , and the fitting formula is as follows:

[0070]

[0071] where a and b are both coefficients obtained from the linear fit.

[0072] Through the above processing, the smoothed value A of the ROLO lunar radiation model as shown in Figure 3 is obtained. ROLO,k .

[0073] Step 5, perform interpolation and resampling on the smoothed value A of the ROLO lunar radiation model ROLO,k to obtain the value A of the ROLO lunar radiation model at the corresponding position of the central wavelength λ of the satellite optical remote sensor i , and use the least squares fitting method to establish the relationship between the lunar image and the value A of the ROLO lunar radiation model i , and obtain the absolute radiometric calibration coefficient gain gains and bias bias. Specifically, it includes: i

[0074] Step 5.1, use the cubic spline interpolation method to perform interpolation on the smoothed value A of the ROLO lunar radiation model ROLO,k to obtain the ROLO model value corresponding to the wavelength of the spectral response function of the satellite optical remote sensor to be absolutely radiometrically calibrated, and perform resampling on it using the resampling formula, and then obtain the value A of the ROLO lunar radiation model at the corresponding position of the central wavelength λ of the satellite optical remote sensor i , and the resampling formula is given by i , where R

[0075]

[0076] i,j i,j (λ) is the spectral response function of the satellite remote sensor, i represents the spectral band of the satellite remote sensor, j represents the wavelength in the spectral response function of the satellite remote sensor, λ m is the lower limit corresponding to the wavelength of the spectral response function of the satellite remote sensor, and λ n is the upper limit corresponding to the wavelength of the spectral response function of the satellite remote sensor;

[0077] Step 5.2, resample the solar irradiance using the resampling formula to obtain the solar irradiance E at the central wavelength λ of the satellite remote sensor. i at the corresponding position, and the calculation formula is: i , the calculation formula is;

[0078]

[0079] where E(λ) is the solar irradiance;

[0080] Step 5.3, calculate the lunar equivalent disk irradiance model value I at the standard distance i , and the calculation formula is:

[0081] where Ω M is the lunar solid angle, Ω M = 6.4177×10 5 sr;

[0082] Step 5.4, calculate the lunar equivalent disk irradiance model value I' at the measured distance i , and the calculation formula is:

[0083]

[0084] where D S-M is the distance between the sun and the moon, D V-M is the distance between the satellite optical sensor and the moon, ME is the average distance between the moon and the earth, AU is the astronomical unit, ME = 3.844×10 8 m, 1AU = 1.49598×10 11 m;

[0085] Step 5.5, apply the least squares fitting method to construct the linear relationship between the measured lunar image of each band of the satellite optical remote sensor at different lunar phase angles and the ROLO lunar radiation model value A i to obtain the absolute radiometric calibration coefficient gain gains and bias bias. For the lunar observation image of the satellite optical remote sensor, the calculation formula of the lunar equivalent disk irradiance I' at the measured distance is: i,view The calculation formula is:

[0086]

[0087] where L is the radiance and DN is the gray value, where Ω p is the solid angle of a single pixel of the satellite image. Theoretically, the lunar equivalent disk irradiance model value I' at the measured distance i is equal to the lunar equivalent disk irradiance I' measured by the satellite, so the fitting formula is: i,view equal, so the fitting formula is:

[0088]

[0089] In another embodiment, the present invention provides a lunar-based satellite optical remote sensor radiometric calibration system, comprising:

[0090] An extraction module, configured to perform dark current subtraction and flat-field correction preprocessing on lunar images within the [-90°, 90°] lunar phase angle range, and extract the total lunar gray value T DN and the total number of pixels T pixel ;

[0091] An equivalent disk reflectance A ROLO An acquisition module: configured to calculate the equivalent disk reflectance A at a standard distance using the ROLO lunar radiation model according to the lunar observation geometric conditions ROLO ;

[0092] A processing module, configured to calculate the mean value of the lunar front area of the reflectance image of M 3 and perform smoothing processing using the cubic spline interpolation method to obtain the average reflectance of the lunar front image of M 3 The average reflectance of the lunar front image For resampling the average reflectance of the lunar front image of M 3 The average reflectance of the lunar front image to obtain the average reflectance of M k at the corresponding position of the wavelength λ of the ROLO lunar radiation model 3 The average reflectance of the lunar front image Furthermore, using and the linear fitting relationship of the equivalent disk reflectance A ROLO to smooth the equivalent disk reflectance A ROLO to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k ; for interpolating and resampling the smoothed value A of the ROLO lunar radiation model ROLO , k to obtain the ROLO lunar radiation model value A i at the corresponding position of the wavelength λ of the satellite optical remote sensor i ;

[0093] A radiometric calibration module, configured to establish a relationship between the lunar image and the ROLO lunar radiation model value A i using the least squares fitting method, and obtain the absolute radiometric calibration coefficient gain gains and bias bias.

[0094] In another embodiment, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned lunar-based satellite radiometric calibration method is implemented.

[0095] In another embodiment, the present invention provides a computer-readable storage medium storing a computer program, which causes a computer to execute the above-described lunar satellite radiometric calibration method.

[0096] As described above, the above are only the preferred embodiments of the present invention, and do not impose any formal limitations on the present invention. Any person skilled in the relevant art, without departing from the scope of the technical solution of the present invention and based on the technical essence of the present invention, any simple modifications, equivalent replacements, and improvements made to the above embodiments shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A radiation calibration method for a satellite optical remote sensor based on the moon, characterized in that It includes the following steps: Step 1: Select lunar images within the lunar phase angle range of [-90°, 90°] for preprocessing of dark current subtraction and flat field correction, and extract the total lunar grayscale value T DN and the total number of pixels T pixel ; Step 2: According to the lunar observation geometric conditions, use the ROLO lunar radiation model to calculate the equivalent disk reflectivity A at the standard distance ROLO ; Step 3, calculate the mean value of the reflectivity image of the lunar mineral mapper M 3 for the frontal area of the Moon, and obtain M through smoothing processing using the cubic spline interpolation method 3 The average reflectivity of the frontal image of the Moon Step 4, for M 3 Average reflectivity of the lunar near - side image Perform resampling to obtain the M at the corresponding position of the wavelength λ of the ROLO lunar radiation model k Average reflectivity of the lunar near - side image 3 Average reflectivity of the lunar near - side image Furthermore, use The linear relationship fitted with the equivalent disk reflectivity A obtained in Step 2 ROLO To smooth the equivalent disk reflectivity A ROLO To obtain the smoothed value A of the ROLO lunar radiation model ROLO,k ; Step 5: Interpolate and resample the smoothed value A of the ROLO lunar radiation model ROLO,k to obtain the value A of the ROLO lunar radiation model corresponding to the central wavelength λ i of the satellite optical remote sensor. Use the least squares fitting method to establish the relationship i between the lunar image and the value A of the ROLO lunar radiation model i to obtain the absolute radiometric calibration coefficient gains and bias.

2. The method for radiometric calibration of a satellite optical remote sensor based on the moon according to claim 1, characterized in that: Step 3 includes the following steps: S3.1, traverse M 3 For the reflectivity image pixels, if the lunar longitude of a certain pixel is outside the range of [-90°, 90°], then mask this pixel to obtain the lunar near-side reflectivity image img1 with masking marks; S3.2, traverse the pixels in the lunar front reflectance image img1 without a shielding mark. If the pixel value is 0, shield the pixel to obtain the lunar reflectance image img2 with a shielding mark; S3.3, traverse the pixels without masking marks in the lunar reflectivity image img2 with masking marks, calculate the mean value of the unmarked pixels in each band's reflectivity image, and smooth it using the cubic spline interpolation method to obtain M 3 Average reflectivity of the lunar near side image 3. A method for radiometric calibration of a satellite optical remote sensor based on the moon, characterized in that: Step 4 includes the following steps: S4.1, resample M 3 Average reflectivity of the lunar plane to the wavelength position λ of the ROLO model k to obtain the M corresponding to the wavelength λ of the ROLO lunar radiation model k at the corresponding position 3 Average reflectivity of the lunar near - side image Specifically, by the formula: Given, where R k,s (λ) is the spectral response function of the ROLO lunar radiation model, λ1 is the lower limit corresponding to the wavelength of the ROLO model spectral response function, and λ2 is the upper limit corresponding to the wavelength of the ROLO model spectral response function; S4.2, for the equivalent disk reflectivity A of the ROLO lunar radiation model ROLO and the wavelength λ of the ROLO lunar radiation model k corresponding to the M 3 average reflectivity of the lunar near-side image perform a linear fit, and use the least squares method to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k , and the fitting formula is as follows: Among them, a and b are coefficients obtained by linear fitting.

4. A method for radiometric calibration of a satellite optical remote sensor based on the moon according to claim 1, characterized in that: Step 5 includes the following steps: S5.

1. Interpolate the smoothed value A of the ROLO model using cubic spline interpolation method ROLO,k to obtain the ROLO model value A corresponding to the wavelength of the spectral response function of the satellite optical remote sensor to be absolutely radiometrically calibrated ROLO,j . Resample it using the resampling calculation formula to further obtain the ROLO lunar radiation model value A at the corresponding position of the central wavelength λ i of the satellite optical remote sensor. The resampling formula is: i ​ Among them, R i,j (λ) is the spectral response function of the satellite remote sensor, and λ m is the lower limit corresponding to the wavelength of the spectral response function of the satellite remote sensor, and λ n is the upper limit corresponding to the wavelength of the spectral response function of the satellite remote sensor; S5.2, resample the solar irradiance using the resampling formula to obtain the solar irradiance E at the corresponding position of the central wavelength λ of the satellite optical remote sensor. The calculation formula is as follows: i The solar irradiance E at the corresponding position i , and the calculation formula is: Among them, E(λ) is the solar irradiance; S5.3, calculate the lunar equivalent disk irradiance model value I at the standard distance i , and the calculation formula is as follows: where Ω M is the solid angle of the moon, and Ω M = 6.4177×10 5 sr; S5.4, calculate the lunar equivalent disk irradiance model value I' at the measured distance i , and the calculation formula is as follows: Among them, D S-M is the distance between the sun and the moon, D V-M is the distance between the satellite optical sensor and the moon, ME is the average distance between the moon and the earth, AU is the astronomical unit, ME = 3.844×10 8 m, 1AU = 1.49598×10 11 m; S5.5, applying the least squares fitting method to construct the linear relationship between the measured lunar images of each band at different lunar phase angles and the ROLO lunar radiation model values A, and obtaining the absolute radiometric calibration coefficient gains and biases; for the lunar observation images of satellite optical remote sensors, the calculation formula for the equivalent disk irradiance I' of the moon at the measured distance is: i Between, obtaining the absolute radiometric calibration coefficient gains and biases; for the lunar observation images of satellite optical remote sensors, the equivalent disk irradiance I' of the moon at the measured distance i,view The calculation formula is: where L is the radiance, DN is the gray value, and Ω p is the solid angle of a single pixel of the satellite image; Model value I' of the lunar equivalent disk irradiance under the measured distance i is equal to the lunar equivalent disk irradiance I' measured by the satellite i,view and the fitting formula is:

5. A satellite optical remote sensor radiometric calibration system based on the moon, characterized in that, It includes: An extraction module is used to perform preprocessing such as dark current subtraction and flat field correction on lunar images within the lunar phase angle range of [-90°, 90°], and extract the total lunar gray value T DN and the total number of pixels T pixel ; Equivalent disk reflectance A ROLO Obtaining module: configured to calculate the equivalent disk reflectance A at a standard distance by using the ROLO lunar radiation model according to the lunar observation geometric conditions ROLO ; Processing module, used to calculate M 3 The reflectivity image of the moon's front area is averaged, and the cubic spline interpolation method is used to smooth it to obtain M 3 Average reflectivity of the lunar front image For M 3 Average reflectivity of the lunar front image Resampling is performed to obtain the wavelength λ of the ROLO lunar radiation model k M in the corresponding position 3 Average reflectivity of the lunar front image Then use and the equivalent disk reflectivity A ROLO The linear fitting relationship of the equivalent disk reflectivity A ROLO Smoothing is performed to obtain the smoothed value A of the ROLO lunar radiation model ROLO,k ; Used to smooth the value A of the ROLO lunar radiation model ROLO,k Perform interpolation and resampling to obtain the central wavelength λ of the satellite optical remote sensor i ROLO lunar radiation model value A at the corresponding position i ; Radiometric calibration module, which is used to establish the relationship between the lunar image and the ROLO lunar radiometric model value A by using the least squares fitting method, and obtain the absolute radiometric calibration coefficient gains and bias. i between them, and obtain the absolute radiometric calibration coefficient gains and bias.

6. An electronic device, characterized in that, It includes: A memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the lunar-based satellite optical remote sensor radiometric calibration method according to any one of claims 1-4.

7. A computer-readable storage medium storing a computer program, the computer program causing a computer to execute the lunar-based satellite optical remote sensor radiometric calibration method according to any one of claims 1-4.

Images