A hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution

Abstract

The application relates to a hyperspectral remote sensor cross-calibration method based on a dynamic Gaussian distribution, and solves the technical problem that the spectral matching factor obtained by the existing spectral matching factor calculation method is low in accuracy and precision, and it is difficult to improve the cross-calibration precision of the interference type hyperspectral remote sensor. The method comprises the following steps: 1, determining a cross-calibration test area; 2, calculating a simulated spectral radiance; 3, calculating the spectral matching factor of a plurality of spectral channels of a to-be-calibrated remote sensor B which have a greater influence on a to-be-calibrated remote sensor A based on a dynamic Gaussian distribution; 4, obtaining the final spectral matching factor of each spectral channel of the to-be-calibrated remote sensor A; 5, calculating the real spectral radiance of the plurality of spectral channels of the to-be-calibrated remote sensor B which have a greater influence on the to-be-calibrated remote sensor A; 6, calculating the real spectral radiance of the to-be-calibrated remote sensor A; 7, calculating the radiation calibration coefficient of each spectral channel of the to-be-calibrated remote sensor A, and completing the cross-calibration of the hyperspectral remote sensor.

Description

Technical Field

[0001] This invention specifically relates to a cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution. Background Technology

[0002] Interferometric hyperspectral remote sensors, with their unique advantages of multi-channel operation, high resolution, and high signal-to-noise ratio, have become an irreplaceable payload in the aerospace field. The hyperspectral data acquired by these sensors exhibits an equal wavenumber distribution during interferometric inversion, leading to uneven distribution of center wavelengths and inconsistent spectral resolutions across different spectral channels. Currently, on-orbit radiometric calibration often employs cross-calibration based on spectral channel matching, calculating the spectral matching factor between the channels of the sensor to be calibrated and the reference sensor to obtain the calibration coefficients for each spectral channel of the sensor to be calibrated.

[0003] Chinese patent CN102279393A discloses a cross-radiometric calibration method for hyperspectral sensors based on multispectral sensors. It adopts the principle of "closeness of center wavelength," utilizing a single channel of the multispectral sensor to provide calibration references for multiple channels of the hyperspectral sensor, establishing a one-to-many band mapping relationship between the hyperspectral and multispectral sensors. Based on calibration coefficients G and B, and using the relationship between the radiance L at the entrance pupil of the remote sensor to be calibrated and the digital count value DN, L = G * DN + B, the matching factor between channels of different bandwidths is calculated. Finally, the calibration coefficients are obtained through linear fitting.

[0004] The spectral matching factor, as a parameter quantifying the degree of matching between the spectral channels of two remote sensors, can be used to evaluate the accuracy and effectiveness of cross-calibration. However, current methods for calculating the spectral matching factor still have certain shortcomings, especially for interferometric hyperspectral remote sensors. When these sensors are used as the sensors to be calibrated, the uneven distribution of center wavelengths in each spectral channel and the inconsistent range of spectral response functions lead to significant errors in obtaining the spectral matching factor. Therefore, improving the accuracy and precision of the spectral matching factor, and thus enhancing the cross-calibration accuracy of interferometric hyperspectral remote sensors, is a pressing issue in the field of on-orbit radiometric calibration. Summary of the Invention

[0005] The purpose of this invention is to solve the technical problem that the accuracy and precision of the spectral matching factors obtained by existing spectral matching factor calculation methods are low, making it difficult to improve the cross-calibration accuracy of interferometric hyperspectral remote sensors. Instead, this invention provides a cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution, characterized by the following steps:

[0008] Step 1: Determine the cross-calibration test area and obtain the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) for the test area;

[0009] Step 2: Obtain the spectral channel characteristics of the remote sensor to be calibrated A and the reference remote sensor B; the spectral channel characteristics include the center wavelength, spectral channel, and spectral response function; based on the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the test area obtained in Step 1, simulate the apparent radiance of the remote sensor to be calibrated A and the reference remote sensor B respectively, and calculate the simulated spectral radiance of the remote sensor to be calibrated A and the reference remote sensor B after their respective spectral channel responses based on the apparent radiance and spectral channel characteristics;

[0010] Step 3: Based on the simulated spectral radiance obtained in Step 2, calculate the spectral matching factor SBAF for several spectral channels in the remote sensor to be calibrated (B) that have a significant impact on the remote sensor to be calibrated (A) based on the dynamic Gaussian distribution. i,g :

[0011]

[0012] In the formula, R A (λ i R represents the simulated spectral radiance of the remote sensor A to be calibrated after the response in the i-th spectral channel. B (λ1), R B (λ2), ..., R B (λ G () represent the simulated spectral radiance of the reference remote sensor B after responding to the spectral channels 1, 2, ..., G.

[0013] Step 4: Assign weight coefficients to the spectral matching factors obtained in Step 3, and weight them to obtain the final spectral matching factors for each spectral channel of the remote sensor A to be calibrated.

[0014] Step 5: Calculate the true spectral radiance of several spectral channels in remote sensor B that have a significant impact on remote sensor A.

[0015] Step 6: Calculate the true spectral radiance of each spectral channel of the remote sensor A to be calibrated based on the final spectral matching factor obtained in Step 4 and the true spectral radiance obtained in Step 5.

[0016] Step 7: Based on the true spectral radiance obtained in Step 6, calculate the radiometric calibration coefficients of each spectral channel of the remote sensor A to be calibrated, generate a spectral matching factor based on a dynamic Gaussian distribution, and complete the cross-calibration of the hyperspectral remote sensor.

[0017] Furthermore, step 2 specifically involves:

[0018] 2.1 Obtain the spectral channel characteristics of the remote sensor to be calibrated, A, and the reference remote sensor, B; the spectral channel characteristics include the center wavelength, spectral channel, and spectral response function;

[0019] 2.2 Input the atmospheric parameters, surface reflectance, and observation geometry parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) obtained in step 1 into the MODTRAN model; use the MODTRAN model to simulate the continuous apparent radiance of the remote sensor to be calibrated (A) and the reference remote sensor (B) at the top of the atmosphere, respectively, to obtain the apparent radiance f. A (λ) and f B (λ);

[0020] 2.3. Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance f obtained in step 2.2 A (λ), the simulated spectral radiance R of the remote sensor A to be calibrated after the response of the i-th spectral channel is calculated by the following formula. A (λ i ):

[0021]

[0022] In the formula, a and b are the values ​​of the given band λ. i The lower and upper bounds of the wavelength of the i-th spectral channel, i = 1, 2, 3, ..., I, where I is the number of spectral channels of the remote sensor A to be calibrated; S A (λ i ) is the spectral response function of the remote sensor A to be calibrated in the i-th spectral channel;

[0023] 2.4. Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance f obtained in step 2.2 B (λ), the simulated spectral radiance R of the reference remote sensor B after the response in the j-th spectral channel is calculated by the following formula. B (λ j ):

[0024]

[0025] In the formula, c and d are given band λ j The lower and upper bounds of the wavelength of the j-th spectral channel, j = 1, 2, 3, ..., J, where J is the number of spectral channels of the reference remote sensor B; S B (λ j ) is the spectral response function of the reference remote sensor B in the j-th spectral channel.

[0026] Furthermore, step 3 specifically involves:

[0027] 3.1. Based on the spectral resolution FWHM(λ) of the i-th spectral channel of the remote sensor A to be calibrated i ), calculate the Gaussian function (λ) of the i-th spectral channel. i ):

[0028]

[0029] gaussian(λ i )=exp[(x-μ) 2 / 2σ 2 ]

[0030] In the formula, σ is the Gaussian function (λ) of the i-th spectral channel. i The standard deviation of λ, where μ is the value at the center of the Gaussian function, is used as the center wavelength of the i-th spectral channel. i The value of ), where x is any wavelength within the response range of the remote sensor A to be calibrated;

[0031] 3.2. Based on the Gaussian function (λ) of the i-th spectral channel... i Plot a Gaussian function curve and normalize its area to obtain the g-th spectral channel of the reference remote sensor B that will have a significant impact on the i-th spectral channel of the remote sensor A to be calibrated during the spectral matching process, where g = 1, 2, ..., G, and G is the number of reference remote sensor B channels that will have a significant impact on the i-th spectral channel of the remote sensor A to be calibrated, G ∈ J; based on the simulated spectral radiance R obtained in step 2.4 B (λ j The spectral radiance of the G spectral channels of the reference remote sensor B is obtained accordingly.

[0032] 3.3. Based on the simulated spectral radiance R obtained in step 2.3 A (λ i Using the spectral radiance of the G spectral channels of the reference remote sensor B obtained in step 3.2, calculate the spectral matching factor SBAF of the G spectral channels of the reference remote sensor B that have a significant impact on the remote sensor A to be calibrated. i,g :

[0033]

[0034] Furthermore, step 4 specifically involves:

[0035] 4.1. The value of the Gaussian function at the center wavelength of the j-th spectral channel of the reference remote sensor B is used as the weight w of the spectral matching factor corresponding to the spectral radiance of the G channels of the reference remote sensor B. i,g ;

[0036] 4.2. The weighting coefficient w is calculated using the following formula.i,G Normalization is performed to obtain the weight coefficients w′ i,g :

[0037]

[0038] 4.3. Based on the weighting coefficient w′ obtained in step 4.2 i,g The spectral matching factor SBAF obtained in step 3.3 is calculated using the following formula. i,g Weighted summation yields the final spectral matching factor SBAF′ for each spectral channel of the remote sensor A to be calibrated. i,g :

[0039]

[0040] Furthermore, step 5 specifically includes:

[0041] 5.1 Acquire corresponding image pairs of the remote sensor to be calibrated (A) and the reference remote sensor (B) in the test area;

[0042] 5.2 Extract the DN values ​​of the remote sensor to be calibrated A in the i-th spectral channel and the reference remote sensor B in the 1st, 2nd...G-th spectral channels of the test area based on the image pairs with the same name;

[0043] 5.3. Based on the linear relationship between DN values ​​and spectral radiance, calculate the true spectral radiance L of the reference remote sensor B in G spectral channels. B,g :

[0044] L B,g =gain g *DN g +offset g

[0045] Where: DN g The reference remote sensor B has the DN value in the test region of the g-th spectral channel; gain g and offset g All of these are the radiometric calibration coefficients of the reference remote sensor B in the G spectral channels.

[0046] Furthermore, step 6 specifically includes:

[0047] The final spectral matching factor SBAF′ of each spectral channel of the remote sensor A to be calibrated, obtained from step 4.3. i,g Compared with the true spectral radiance L of the reference remote sensor B in G spectral channels obtained in step 5.3 B,g Calculate the true spectral radiance L of the remote sensor A to be calibrated in the test area in the i-th spectral channel. A,i :

[0048]

[0049] Furthermore, step 7 specifically includes:

[0050] Based on the true spectral radiance L obtained in step 6 A,i The calibration of the hyperspectral remote sensor A is achieved by fitting the calibration coefficients of each spectral channel of the sensor to be calibrated using the following formula, thus completing the cross-calibration of the hyperspectral remote sensor:

[0051] L A,i =gain i *DN i +offset i

[0052] In the formula, DN i Let DN be the test value of the remote sensor A to be calibrated in the i-th spectral channel; gain i and offset i All of these are the radiometric calibration coefficients of the remote sensor A to be calibrated in the i-th spectral channel.

[0053] The beneficial effects of this invention are:

[0054] 1. This invention provides a cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution. According to the data characteristics of interferometric hyperspectral remote sensors, the method uses a Gaussian function that conforms to the response characteristics of its spectral channels to constrain the selection of spectral channels in a reference remote sensor B. Furthermore, it uses a Gaussian distribution to assign weight coefficients to each spectral matching factor, thereby improving the accuracy of the spectral matching factor. It can also efficiently calculate the spectral matching factor of each spectral channel of the remote sensor A to be calibrated, simplifying the cross-calibration process and improving the cross-calibration accuracy.

[0055] 2. This invention provides a cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution. It fully considers the variability of the spectral channels of the interferometric remote sensor to be calibrated and utilizes a dynamic Gaussian function to achieve better utilization of the spectral channels of the reference remote sensor B. Considering that the spectral response function of the remote sensor to be calibrated A is significantly affected by the superposition effect of multiple spectral channels, the dynamic Gaussian distribution can more accurately capture and process these spectral features, thereby achieving more accurate spectral matching.

[0056] 3. The present invention provides a cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution, which can improve the accuracy of cross-calibration of interferometric hyperspectral remote sensors, provide ideas for optimizing the spectral matching factor of other hyperspectral remote sensors, and broaden the application scope of cross-calibration. Attached Figure Description

[0057] Figure 1 This is a flowchart of an embodiment of a hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to the present invention;

[0058] Figure 2This is a comparison chart of processing hyperspectral remote sensor data using traditional methods (blue) and the hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution of the present invention (red). Detailed Implementation

[0059] like Figure 1 As shown, a hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution is presented. The method simulates the radiance of each spectral channel of two remote sensors using the MODTRAN model, and then, combining the spectral channel characteristics of the two remote sensors, calculates the spectral matching factor set using the dynamic Gaussian distribution method. Next, the DN value of the cross region of the two remote sensor image pairs is extracted, and the calibration of the remote sensor A to be calibrated is achieved based on the linear relationship between the DN value and the radiance. The method specifically includes the following steps:

[0060] Step 1: Determine the cross-calibration test area and obtain the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) for the test area.

[0061] Step 2: Obtain the spectral channel characteristics of the remote sensor to be calibrated A and the reference remote sensor B; the spectral channel characteristics include the center wavelength, spectral channel, and spectral response function; based on the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the test area obtained in Step 1, simulate the apparent radiance of the remote sensor to be calibrated A and the reference remote sensor B respectively, and calculate the simulated spectral radiance of the remote sensor to be calibrated A and the reference remote sensor B after their respective spectral channel responses based on the apparent radiance and spectral channel characteristics;

[0062] 2.1 Obtain the spectral channel characteristics of the remote sensor to be calibrated, A, and the reference remote sensor, B; the spectral channel characteristics include the center wavelength, spectral channel, and spectral response function;

[0063] 2.2 Input the atmospheric parameters, surface reflectance, and observation geometry parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) obtained in step 1 into the MODTRAN model; use the MODTRAN model to simulate the continuous apparent radiance of the remote sensor to be calibrated (A) and the reference remote sensor (B) at the top of the atmosphere, respectively, to obtain the apparent radiance f. A (λ) and f B (λ);

[0064] 2.3. Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance f obtained in step 2.2 A (λ), the simulated spectral radiance R of the remote sensor A to be calibrated after the response of the i-th spectral channel is calculated by the following formula. A (λ i ):

[0065]

[0066] In the formula, a and b are the values ​​of the given band λ. i The lower and upper bounds of the wavelength of the i-th spectral channel, i = 1, 2, 3, ..., I, where I is the number of spectral channels of the remote sensor A to be calibrated; S A (λ i ) is the spectral response function of the remote sensor A to be calibrated in the i-th spectral channel;

[0067] 2.4. Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance f obtained in step 2.2 B (λ), the simulated spectral radiance R of the reference remote sensor B after the response in the j-th spectral channel is calculated by the following formula. B (λ j ):

[0068]

[0069] In the formula, c and d are given band λ j The lower and upper bounds of the wavelength of the j-th spectral channel, j = 1, 2, 3, ..., J, where J is the number of spectral channels of the reference remote sensor B; S B (λ j ) is the spectral response function of the reference remote sensor B in the j-th spectral channel;

[0070] Step 3, based on the simulated spectral radiance R A (λ i ) and simulated spectral radiance R B (λ j Based on the dynamic Gaussian distribution, the spectral matching factors of multiple spectral channels in remote sensor B that have a significant impact on remote sensor A are calculated.

[0071] The remote sensor to be calibrated, A, is an interferometric hyperspectral remote sensor with a Gaussian spectral response function. Its spectral channel bandwidth and spectral resolution are constantly changing. Using the dynamic Gaussian distribution method, the spectral channel of the reference remote sensor B is retrieved for the i-th spectral channel (any spectral channel) of the remote sensor to be calibrated, A.

[0072] 3.1. Based on the spectral resolution FWHM(λ) of the i-th spectral channel of the remote sensor A to be calibrated i ), calculate the Gaussian function (λ) of the i-th spectral channel. i ):

[0073]

[0074] gaussian(λ i )=exp[(x-μ) 2 / 2σ 2 ]

[0075] In the formula, σ is the Gaussian function (λ) of the i-th spectral channel. i The standard deviation of λ, where μ is the value at the center of the Gaussian function, is used as the center wavelength of the i-th spectral channel. i The value of ), where x is any wavelength within the response range of the remote sensor A to be calibrated;

[0076] 3.2. Based on the Gaussian function (λ) of the i-th spectral channel of the remote sensor A to be calibrated i Plot the Gaussian function curve and normalize the area of ​​the function.

[0077] When |x-μ|=2σ, the area enclosed by the Gaussian function accounts for 95% of the total area, which can completely cover the influential spectral channels, thus yielding the search range [μ-2σ,μ+2σ], i.e., [center_wave(λ i )-2σ,center_wave(λ i The center wavelength of the reference remote sensor B within [)+2σ] is used to obtain the g-th spectral channel of the reference remote sensor B that has a significant impact on the i-th spectral channel of the remote sensor A to be calibrated during the spectral matching process, where g = 1, 2, ..., G, and G is the number of reference remote sensor B channels that have a significant impact on the i-th spectral channel of the remote sensor A to be calibrated, G ∈ J. Based on the simulated spectral radiance R obtained in step 2.4... B (λ j The spectral radiance of the G spectral channels of the reference remote sensor B is obtained accordingly.

[0078] 3.3. Based on the simulated spectral radiance R obtained in step 2.3 A (λ i Using the spectral radiance of the G spectral channels of the reference remote sensor B obtained in step 3.2, calculate the spectral matching factor SBAF of the G spectral channels of the reference remote sensor B that have a significant impact on the remote sensor A to be calibrated. i,g :

[0079]

[0080] Step 4, SBAF is the spectral matching factor for the G spectral channels. i,g Assign weight coefficients and weight them to obtain the final spectral matching factors for each spectral channel of the remote sensor A to be calibrated;

[0081] 4.1. The value of the Gaussian function at the center wavelength of the j-th spectral channel of the reference remote sensor B is used as the weight w of the spectral matching factor corresponding to the spectral radiance of the G channels of the reference remote sensor B. i,g ;

[0082] 4.2. The weighting coefficient w is calculated using the following formula.i,G Normalization is performed to obtain the weight coefficients w′ i,g :

[0083]

[0084] 4.3. Based on the weighting coefficient w′ obtained in step 4.2 i,g The spectral matching factor SBAF obtained in step 3.3 is calculated using the following formula. i,g Weighted summation yields the final spectral matching factor SBAF′ for each spectral channel of the remote sensor A to be calibrated. i,g :

[0085]

[0086] Step 5: Calculate the true spectral radiance of the reference remote sensor B in G spectral channels;

[0087] 5.1 Acquire corresponding image pairs of the remote sensor to be calibrated (A) and the reference remote sensor (B) in the test area;

[0088] 5.2 Extract the DN (Digital Number) values ​​of the remote sensor to be calibrated A in the i-th spectral channel and the reference remote sensor B in the 1st, 2nd...G spectral channels of the test area based on the image pairs with the same name.

[0089] 5.3. Based on the linear relationship between DN values ​​and spectral radiance, calculate the true spectral radiance L of the reference remote sensor B in G spectral channels. B,g :

[0090] L B,g =gain g *DN g +offset g

[0091] Where: DN g The reference remote sensor B has the DN value in the test region of the g-th spectral channel; gain g and offset g All are radiometric calibration coefficients of the reference remote sensor B in the G spectral channels;

[0092] Step 6: The final spectral matching factor SBAF′ of each spectral channel of the remote sensor A to be calibrated, obtained from Step 4.3. i,g Compared with the true spectral radiance L of the reference remote sensor B in G spectral channels obtained in step 5.3 B,g Calculate the true spectral radiance L of the remote sensor A to be calibrated in the test area in the i-th spectral channel. A,i :

[0093]

[0094] Step 7, based on the true spectral radiance L obtained in Step 6 A,i DN combined with cross-image i The calibration values ​​are obtained by fitting the calibration coefficients of each spectral channel of the remote sensor A to be calibrated using the following formula:

[0095] L A,i =gain i *DN i +offset i .

[0096] This invention calculates spectral matching factors based on the data characteristics and spectral channel features of an interferometric hyperspectral remote sensor, using multiple spectral channels of a Gaussian-distributed reference remote sensor B. Dynamic weight allocation is then performed according to the distribution characteristics of the spectral channels to improve the accuracy of the spectral matching factors, thereby enhancing the accuracy of cross-calibration. This method optimizes the spectral matching factor generation method, improves the reliability of cross-calibration results, and solves the problem of low accuracy in spectral matching factors caused by uneven distribution of spectral channel center wavelengths and inconsistent spectral resolution in the cross-calibration of interferometric hyperspectral remote sensors.

[0097] like Figure 2 As shown, in the visible and near-infrared (VNIR) bands, the spectral response function of the latter half of the spectral channels is significantly affected by the superposition effect of multiple spectral channels. Compared with traditional methods, the hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution of this invention can scientifically allocate weights according to the distribution characteristics of spectral channels, thereby achieving more accurate spectral matching and obtaining higher precision cross-calibration results. In this embodiment, the remote sensor to be calibrated, A, is the hyperspectral remote sensor (interferometric type) of the Environment-2 satellite, and the reference remote sensor, B, is the hyperspectral remote sensor of the Gaofen-5 satellite.

Claims

1. A cross-calibration method for hyperspectral remote sensors based on dynamic Gaussian distribution, used for cross-calibration of interferometric hyperspectral remote sensors, characterized in that, Includes the following steps: Step 1: Determine the cross-calibration test area and obtain the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) for the test area; Step 2: Obtain the spectral channel characteristics of the remote sensor to be calibrated, A, and the reference remote sensor, B; The spectral channel characteristics include center wavelength, spectral channel, and spectral response function; based on the atmospheric parameters, surface reflectance, and observation geometric condition parameters of the test area obtained in step 1, the apparent radiance of the remote sensor to be calibrated A and the reference remote sensor B are simulated respectively, and based on the apparent radiance and spectral channel characteristics, the simulated spectral radiance of the remote sensor to be calibrated A and the reference remote sensor B after their respective spectral channel responses are calculated; Step 3: Based on the simulated spectral radiance obtained in Step 2, calculate the spectral matching factors of several spectral channels in the reference remote sensor B that have a significant impact on the remote sensor A to be calibrated, based on the dynamic Gaussian distribution. : ； In the formula, For remote sensor A to be calibrated in the first Simulated spectral radiance after spectral channel response , … These represent the simulated spectral radiance of reference sensor B after responding to the spectral channels 1, 2, ..., G; where the search range of the center wavelength of reference sensor B is... ,in, It is the first Center wavelength value of spectral channel It is the first Spectral channel Gaussian function The standard deviation of is calculated using the following formula: ； In the formula, Let be the spectral resolution of the i-th spectral channel; The first Spectral channel Gaussian function The calculation formula is as follows: ； In the formula, It is the value at the center of the Gaussian function. For any wavelength within the response range of the remote sensor A to be calibrated; Step 4: Assign weight coefficients to the spectral matching factors obtained in Step 3, and weight them to obtain the final spectral matching factors for each spectral channel of the remote sensor A to be calibrated. Step 5: Calculate the true spectral radiance of several spectral channels in remote sensor B that have a significant impact on remote sensor A. Step 6: Calculate the true spectral radiance of each spectral channel of the remote sensor A to be calibrated based on the final spectral matching factor obtained in Step 4 and the true spectral radiance obtained in Step 5. Step 7: Based on the true spectral radiance obtained in Step 6, calculate the radiometric calibration coefficients of each spectral channel of the remote sensor A to be calibrated, generate a spectral matching factor based on a dynamic Gaussian distribution, and complete the cross-calibration of the hyperspectral remote sensor.

2. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 1, characterized in that, Step 2 is as follows: 2.1 Obtain the spectral channel characteristics of the remote sensor to be calibrated, A, and the reference remote sensor, B; The spectral channel characteristics include the center wavelength, spectral channel, and spectral response function; 2.2 Input the atmospheric parameters, surface reflectance, and observation geometry parameters of the remote sensor to be calibrated (A) and the reference remote sensor (B) obtained in step 1 into the MODTRAN model; use the MODTRAN model to simulate the continuous apparent radiance of the remote sensor to be calibrated (A) and the reference remote sensor (B) at the top of the atmosphere, respectively, to obtain the apparent radiance. and ; 2.3 Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance obtained in step 2.2 The following formula is used to calculate the value of the remote sensor A to be calibrated in the first... Simulated spectral radiance after spectral channel response : ； In the formula, and These are the given bands In the The lower and upper bounds of the wavelength of the spectral channel. =1,2,3,…,I,I is the number of spectral channels of the remote sensor A to be calibrated; The remote sensor A to be calibrated is in the first Spectral response function of the spectral channel; 2.

4. Based on the spectral channel characteristics obtained in step 2.1 and the apparent radiance obtained in step 2.2 The reference remote sensor B is calculated using the following formula at the 1st... Simulated spectral radiance after spectral channel response : ； In the formula, and It is a given band In the The lower and upper bounds of the wavelength of the spectral channel. =1,2,3,…,J, where J is the number of spectral channels of the reference remote sensor B; Reference remote sensor B is in the 1st Spectral response function of the spectral channel.

3. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 2, characterized in that, Step 3 specifically involves: 3.1 According to the first Spectral channel Gaussian function Plot the Gaussian function curve and normalize its area to obtain the first Gaussian function curve for the remote sensor A to be calibrated during spectral matching. The spectral channels will have a significant impact on the reference remote sensor B. Spectral channels G is the first digit of the remote sensor A to be calibrated. The number of channels in the reference remote sensor B, G∈, will have a significant impact on the spectral channels. Based on the simulated spectral radiance obtained in step 2.4 The spectral radiance of the corresponding G spectral channels of the reference remote sensor B is obtained; 3.

2. Based on the simulated spectral radiance obtained in step 2.3 Using the spectral radiance of the G spectral channels of the reference remote sensor B obtained in step 3.2, calculate the spectral matching factor of the G spectral channels of the reference remote sensor B that have a significant impact on the remote sensor A to be calibrated. : 。 4. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 3, characterized in that, Step 4 is as follows: 4.

1. The value of the Gaussian function at the center wavelength of the j-th spectral channel of the reference remote sensor B is used as the weight of the spectral matching factor corresponding to the spectral radiance of the G channels of the reference remote sensor B. ; 4.

2. Weighting coefficients are calculated using the following formula. Normalization is performed to obtain the weight coefficients. : ； 4.

3. Based on the weighting coefficients obtained in step 4.2 The spectral matching factor obtained in step 3.3 is calculated using the following formula. Weighted summation yields the final spectral matching factor for each spectral channel of the remote sensor A to be calibrated. : 。 5. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 4, characterized in that, Step 5 specifically involves: 5.1 Acquire corresponding image pairs of the remote sensor to be calibrated (A) and the reference remote sensor (B) in the test area; 5.

2. Based on the images with the same name, extract the values ​​of the remote sensor A to be calibrated in the [missing information]. DN values ​​of spectral channels and reference remote sensor B in the experimental regions of spectral channels 1, 2, ..., G; 5.3 Based on the linear relationship between DN values ​​and spectral radiance, calculate the true spectral radiance of reference remote sensor B in G spectral channels. : ； In the formula: For reference, remote sensor B in the 1st DN value of the test area in the spectral channel; and All of these are the radiometric calibration coefficients of the reference remote sensor B in the G spectral channels.

6. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 5, characterized in that, Step 6 specifically involves: The final spectral matching factors for each spectral channel of the remote sensor A to be calibrated, obtained from step 4.

3. The true spectral radiance of the reference remote sensor B in G spectral channels obtained in step 5.

3. The calculation of the remote sensor A to be calibrated in the test area in the first... True spectral radiance of spectral channels : 。 7. The hyperspectral remote sensor cross-calibration method based on dynamic Gaussian distribution according to claim 6, characterized in that, Step 7 specifically includes: Based on the true spectral radiance obtained in step 6 The calibration of the hyperspectral remote sensor A is achieved by fitting the calibration coefficients of each spectral channel of the sensor to be calibrated using the following formula, thus completing the cross-calibration of the hyperspectral remote sensor: ； In the formula, For remote sensor A to be calibrated in the first DN value of the test area in the spectral channel; and All are remote sensors A to be calibrated in the first Radiometric calibration coefficients of spectral channels.

Images