A calibration method for a satellite optical remote sensing detector

Abstract

This invention belongs to the field of satellite remote sensing technology and discloses a calibration method for satellite optical remote sensing detectors. This invention utilizes a ground-based calibration light source system capable of generating monochromatic light of multiple wavelengths and adjusting the light source's intensity and wavelength. Based on the calibration parameters, it obtains total test information after multiple rounds of measurements, and calibrates the calibration parameters of the satellite optical remote sensing detector based on this total test information. This invention enables comprehensive calibration of satellite optical remote sensing detectors and calibrates and verifies the quality of inversion products produced by satellite optical remote sensing detectors.

Description

Technical Field

[0001] This invention belongs to the field of satellite remote sensing technology, and more specifically, relates to a calibration method for a satellite optical remote sensing detector. Background Technology

[0002] The effectiveness of satellite remote sensing applications largely depends on the level of quantification of remote sensing data. Although satellite optical remote sensing detectors undergo rigorous laboratory calibration before launch, after launch, under space operating conditions, the glass materials, optical films, and optical detectors of the optical system are affected by factors such as radiation from cosmic rays, vacuum environment, and temperature changes. Under the radiation of high-energy rays, windows and lenses gradually become opaque. Radiation damage can lead to the degradation of the sensitive parameters of the optical detectors, and can also induce phenomena such as reduced charge transfer efficiency, increased dark current, and drift of flat band voltage and threshold voltage. This affects performance such as spectral sensitivity, responsivity, linearity, noise equivalent power, and dynamic range, resulting in a decrease in the level of quantification of remote sensing data.

[0003] Existing satellite ground calibration light sources have a single emission spectrum and uncontrollable wavelength, making them unsuitable for calibrating the spectral response of satellite optical sensors. In other words, because existing ground calibration light source systems cannot change the emission wavelength, they can only calibrate parameters such as the noise equivalent power and dynamic range of satellite optical remote sensing detectors, but lack the ability to calibrate the spectral sensitivity characteristics of satellite optical remote sensing detectors (the relationship between the detector's responsivity and the incident wavelength is called spectral sensitivity characteristics). Therefore, there is an urgent need to design and implement a scheme capable of comprehensively calibrating the performance parameters of satellite optical remote sensing detectors. Summary of the Invention

[0004] The purpose of this invention is to provide a calibration method for satellite optical remote sensing detectors to comprehensively calibrate the performance parameters of satellite optical remote sensing detectors.

[0005] This invention provides a calibration method for a satellite optical remote sensing detector, comprising the following steps:

[0006] Step 1: Adjust the light source direction of the ground calibration light source system to align it with the satellite's overhead position; turn on the atmospheric transmittance measurement equipment;

[0007] Step 2: According to the calibration index, control the ground calibration light source system to emit light according to the specified emission wavelength and emission intensity, and perform the i-th round of measurement within the time range of the satellite passing overhead, and obtain the output data of the satellite optical remote sensing detector and atmospheric transmittance data corresponding to the i-th round of measurement;

[0008] Step 3: After the satellite overpass time ends, determine whether the total test information meets the requirements for calibrating the indicators to be calibrated. The total test information includes the output data of the satellite optical remote sensing detector, atmospheric transmittance data, and ground calibration light source data corresponding to all rounds of testing obtained so far. If it meets the requirements, proceed to step 4. If it does not meet the requirements, return to step 2, change at least one parameter of the emission wavelength and emission intensity, and then perform the next round of measurement.

[0009] Step 4: Based on the total test information, calibrate the calibration parameters of the satellite optical remote sensing detector.

[0010] Preferably, during the measurement process, the ground calibration light source system is controlled to maintain the operating temperature, and the luminous intensity is monitored, measured, and compensated in real time.

[0011] Preferably, the ground calibration light source system includes n light source units, where n is an integer greater than or equal to 3; each light source unit is used to generate monochromatic light of a specific wavelength, and the monochromatic light generated by different light source units corresponds to different wavelengths, with multiple wavelengths respectively corresponding to multiple wavelengths in the response spectrum of the satellite optical remote sensing detector.

[0012] Preferably, the indicators to be calibrated are divided into two categories: one is the indicators related to the power of the light source, and the other is the parameters that change the spectral sensitivity and spectral response curve.

[0013] Regarding the indicators related to light source power, the conversion relationship between the luminous power of the ground calibration light source system and the radiance value received by the satellite optical remote sensing detector is as follows:

[0014]

[0015] In the formula, P is the luminous power of the ground calibration light source system, L is the radiance value received by the satellite optical remote sensing detector, and η atm Atmospheric transmittance, PSF is the detector point spread function, ISRF is the value of the detector spectral response curve at the selected emission wavelength, θ is the divergence angle of the ground calibration source, and A pixel Let Δλ be the ground area corresponding to a single pixel of a satellite optical remote sensing detector, and let Δλ be the bandwidth of the satellite optical remote sensing detector.

[0016] When performing quantitative analysis of the spectral response, multiple light sources with different specific wavelengths are used. In this case, the luminous power of the ground calibration light source system is the integral of the luminous power spectral density, expressed as:

[0017]

[0018] In the formula, λ k For the k-th specific wavelength used, Let B be the luminous power of the k-th light source unit in the ground calibration light source system when it emits light, and let Δλ be the luminous power spectral density. k Let be the spectral width of the k-th specific wavelength light source;

[0019] By combining the two formulas above, the spectral response of the satellite optical remote sensing detector to a specific wavelength light source can be obtained.

[0020] Preferably, the indicators related to light source power include responsivity, linearity, noise equivalent power, and dynamic range. When the indicator to be calibrated is one of the indicators related to light source power, calibration is performed in the following manner:

[0021] Step a: Control multiple light source units in the ground calibration light source system to emit light simultaneously; set i=1, and when setting the first measurement window of the i-th round of measurement, the light source emitting power P of the ground calibration light source system is... i1 Maximum optical power P max Measure the output value I of the satellite optical remote sensing detector. i1 ;

[0022] Step b: As the number of measurement windows increases, the light source power is decreased sequentially; the light source power under multiple measurement windows is recorded as P. i2 ... P ij The output value I of the corresponding satellite optical remote sensing detector was measured respectively. i2 ... I ij Where j is an integer greater than or equal to 4;

[0023] Step c, determine I ij Is it less than the root mean square value of the noise? If not, let i = i + 1, and set P. i1 =P (i-1)j Then return to step b; if so, calibrate the index related to the light source power based on the light source luminous power and the corresponding output value of the satellite optical remote sensing detector.

[0024] Preferably, in step b, j is set to 4, and the following settings are made:

[0025] Preferably, when calibrating linearity, a curve of measurement points is plotted with the luminous power of the light source as the abscissa and the output value of the satellite optical remote sensing detector as the ordinate, and a fitting straight line is selected for fitting; the linearity is represented by the relative error σ, as follows:

[0026]

[0027] In the formula, ΔY maxY represents the maximum deviation between the measured point curve and the fitted straight line, and Y is the full-scale output value of the satellite optical remote sensing detector.

[0028] When calibrating the noise equivalent power, find the light source luminous power in the measurement point data where the output value of the satellite optical remote sensing detector equals the root mean square value of the noise, and denote this light source luminous power as P. min Calculate P min The corresponding minimum radiance value L received by the satellite optical remote sensing detector min This is used to represent the noise equivalent power of the detector;

[0029] When the relative error σ is greater than 30%, the satellite optical remote sensing detector is considered to be in a saturated state. When calibrating the dynamic range, the light source luminous power corresponding to when the output value of the satellite optical remote sensing detector reaches saturation is found in the measurement point data, and this light source luminous power is recorded as P. full Calculate P full The corresponding maximum radiance value L received by the satellite optical remote sensing detector full The dynamic range of the satellite optical remote sensing detector is expressed as:

[0030] When calibrating the responsivity, if the relative error σ is less than 30%, the slope of the curve at the measurement point is used as the responsivity of the detector.

[0031] Preferably, the spectral response of the satellite optical remote sensing detector at the time of manufacture is set to R(λ), and the spectral response of the satellite optical remote sensing detector in orbit is R′(λ)=AR(a(λ-λ)). c The parameters that change the detector's spectral response curve include the intensity coefficient A and the peak wavelength shift λ of the spectral response curve in orbit. c And the spectral bandwidth coefficient α; when the parameter to be calibrated is a parameter that changes the spectral response curve of the detector, calibration is performed in the following manner:

[0032] Step a: Select a light source unit with a specific wavelength of λ1 to emit light, set the total luminous power, and measure the output value I(λ1) of the satellite optical remote sensing detector;

[0033] Step b: Keeping the total luminous power constant, change the wavelength of the light source, sequentially selecting specific wavelengths λ2, ..., λ... n The light source unit emits light, and the corresponding output values ​​I(λ2), ..., I(λ) of the satellite optical remote sensing detector are measured respectively. n );

[0034] Step c: Plot (λ1, I(λ1)), (λ2, I(λ2)), ..., (λ1) on the detector spectral response curve. n ,I(λn Multiple measurement points are used to invert the spectral response curve changes of the satellite optical remote sensing detector based on the measurement points.

[0035] Preferably, the implementation methods for retrieving changes in the spectral response curve based on the measurement point include:

[0036] Based on experience, the intensity coefficient A and peak wavelength shift λ of the spectral response curve set in orbit are... c And the range of values ​​for the spectral bandwidth coefficient α;

[0037] Select a set of parameters A from the above range of values. i a i and λ ci The spectral response R′ under this set of parameters is obtained. i (λ):

[0038] R′ i (λ)=A i R(a i (λ-λ ci ))

[0039] For multiple specific wavelengths λ1, ..., λ2 selected n The corresponding response R′ after the spectral change is obtained. i (λ1), ..., R′ i (λ n ):

[0040] R′ i (λ1)=A i R(a i (λ1-λ ci ))

[0041] ...

[0042] R′ i (λ n ) = A i R(a i (λ n -λ ci ))

[0043] Define an evaluation function ξ i To measure the selected A i a i , λ ci Relationship with actual spectral response changes:

[0044]

[0045] In A, a, λ c Within the range of values ​​for ξ, select parameters to find the evaluation function ξ. i A reaches its minimum valuef a f , λ cf The calibrated spectral response curve R′(λ)=A was obtained. f R(a f (λ-λ cf )).

[0046] Preferably, the ground calibration light source system is capable of independently emitting multiple specific wavelengths λ1, ..., λ2. n Monochromatic light, and λ1 < λ2 < ... < λ n The optimal selection of the aforementioned specific wavelengths is achieved using the following method:

[0047] Based on the spectral response of the satellite optical remote sensing detector at the time of manufacture, the value range for each specific wavelength is set;

[0048] Based on experience, the intensity coefficient, peak wavelength shift, and spectral bandwidth coefficient of the spectral response curve set in orbit are A0, a0, and λ, respectively. c0 ;

[0049] From λ1, ..., λ n Select a set of parameters λ from the range of values. 1i 、…、λ ni The corresponding parameters that change in the spectral response curve are A0, a0, and λ. c0 The response R′ at that time i (λ 1i ), ..., R′ i (λ ni ):

[0050] R′ i (λ 1i )=A0R(a0(λ 1i -λ c0 ))

[0051] ...

[0052] R′ i (λ ni )=A0R(a0(λ ni -λ c0 ))

[0053] Using the aforementioned method of retrieving the spectral response curve change based on the measurement point, R′ i (λ 1i ), ..., R′ i (λ ni Perform an inversion operation to obtain A. i a i and λ ci Define its error function for:

[0054]

[0055] In the formula, λ peak This represents the peak wavelength of the satellite optical remote sensing detector's response;

[0056] In λ1, ..., λ n Within the range of values, select wavelength combinations to find the error function. λ at the minimum value 1f , …, λ nf This allows for the optimal selection of a specific wavelength.

[0057] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0058] (1) This invention utilizes a ground calibration light source system that can generate monochromatic light of multiple wavelengths and adjust the light source intensity and wavelength. It can not only calibrate the power-related indicators of the satellite optical remote sensing detector, but also calibrate the parameters of the satellite optical remote sensing detector such as spectral sensitivity and spectral response curve. In other words, this invention can meet the requirements for calibrating parameters such as spectral sensitivity, responsivity, linearity, noise equivalent power, and dynamic range of the satellite optical remote sensing detector, thereby enabling comprehensive calibration of the satellite optical remote sensing detector and calibration and verification of the quality of the inversion products produced by the satellite optical remote sensing detector.

[0059] (2) In the process of performing the measurement, the present invention controls the ground calibration light source system to maintain the working temperature and performs real-time monitoring, measurement and feedback compensation of the luminous intensity, which can ensure the long-term stable operation of the ground calibration light source system and provide a guarantee for achieving accurate calibration.

[0060] (3) Existing technologies lack the steps for calibrating satellite optical remote sensing detectors using ground-based calibration light source systems, and also lack methods for inverting changes in parameters such as spectral sensitivity, responsivity, linearity, noise equivalent power, and dynamic range of satellite detectors using measurement data. This invention fills the above gaps and provides specific solutions, including proposing a method for inverting changes in spectral response curves, constructing an expression model for the on-orbit spectral response of satellite optical remote sensing detectors, and proposing to incorporate the intensity coefficient A and peak wavelength drift λ of the on-orbit spectral response curve. c By using the spectral bandwidth coefficient α as a parameter to change the spectral response curve of the detector, this invention enables comprehensive calibration of satellite optical remote sensing detectors.

[0061] (4) To calibrate the spectral response curve of a satellite optical remote sensing detector, the ground calibration light source system needs to generate three or more monochromatic lights with defined wavelengths and intensities. Different emission spectra can be obtained by arranging and combining monochromatic lights of different wavelengths, thereby calibrating the spectral response of the satellite optical remote sensing detector. Existing technologies lack methods for selecting three or more single optimal wavelengths within the response wavelength range of a satellite optical remote sensing detector. This invention fills this gap by proposing a wavelength optimization selection method, thereby enabling optimal calibration of the spectral response of the satellite optical remote sensing detector. Attached Figure Description

[0062] Figure 1 This is an overall flowchart of a calibration method for a satellite optical remote sensing detector provided in Embodiment 1 of the present invention;

[0063] Figure 2 This is a schematic diagram of the process for calibrating indicators related to light source power in a calibration method for a satellite optical remote sensing detector provided in Embodiment 2 of the present invention.

[0064] Figure 3 This is a schematic diagram illustrating the process of calibrating the spectral response curve of a satellite optical remote sensing detector according to a calibration method provided in Embodiment 3 of the present invention.

[0065] Figure 4 The distribution of the first wavelength when the error function is 0;

[0066] Figure 5 The distribution of the second wavelength when the error function is 0;

[0067] Figure 6 This represents the distribution of the third wavelength when the error function is 0. Detailed Implementation

[0068] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0069] Example 1:

[0070] Example 1 provides a calibration method for a satellite optical remote sensing detector, which mainly includes the following steps:

[0071] Step 1: Adjust the light source direction of the ground calibration light source system to align it with the satellite's overhead position; turn on the atmospheric transmittance measurement equipment;

[0072] Step 2: According to the calibration index, control the ground calibration light source system to emit light according to the specified emission wavelength and emission intensity, and perform the i-th round of measurement within the time range of the satellite passing overhead, and obtain the output data of the satellite optical remote sensing detector and atmospheric transmittance data corresponding to the i-th round of measurement;

[0073] Step 3: After the satellite overpass time ends, determine whether the total test information meets the requirements for calibrating the indicators to be calibrated. The total test information includes the output data of the satellite optical remote sensing detector, atmospheric transmittance data, and ground calibration light source data corresponding to all rounds of testing obtained so far. If it meets the requirements, proceed to step 4. If it does not meet the requirements, return to step 2, change at least one parameter of the emission wavelength and emission intensity, and then perform the next round of measurement.

[0074] Step 4: Based on the total test information, calibrate the calibration parameters of the satellite optical remote sensing detector.

[0075] During the measurement process, the ground calibration light source system is maintained at its operating temperature, and the luminous intensity is monitored, measured, and compensated in real time. For example, the temperature of the ground calibration light source system is adjusted and controlled by a temperature control unit to keep the temperature of the ground calibration light source system constant.

[0076] The ground calibration light source system comprises n light source units, where n is an integer greater than or equal to 3; each light source unit is used to generate monochromatic light of a specific wavelength, and the monochromatic light generated by different light source units corresponds to different wavelengths, with multiple wavelengths corresponding to multiple wavelengths in the response spectrum of the satellite optical remote sensing detector.

[0077] The ground calibration light source system can emit light that meets specified conditions (including spectral and intensity conditions) as needed after receiving control information. For example, it can emit monochromatic light of one wavelength for a certain period of time according to a certain calibration requirement, emit monochromatic light of two wavelengths for another period of time under another calibration requirement, emit multiple monochromatic lights simultaneously under a third calibration requirement, and so on.

[0078] The monochromatic light generated by each of the light source units is a point light source, and the point light source maintains a constant luminous intensity at all angles within the field of view of the satellite optical remote sensing detector.

[0079] Each light source unit includes multiple fixed-power modules and multiple variable-power modules; each variable-power module is connected to a variable-power switch, which controls the on / off state of the connected variable-power module; the variable-power module emits light in steps of 1 / m of the light power of the fixed-power module in the light source unit, where m is an adjustment parameter. The light intensity information of the emitted light source can be monitored in real time using a light intensity detection unit, and the light power of the variable-power module can be adjusted and controlled based on this light intensity information. Each fixed-power module is connected to a fixed-power switch, which controls the on / off state of the connected fixed-power module; the ground calibration light source system emits light in steps of the light power of the fixed-power modules.

[0080] When calibrating various parameters of a satellite optical remote sensing detector, the light intensity conditions under different calibration targets can be met by sequentially decreasing or increasing the number of activated fixed optical power modules. A variable module switch can control the activation or deactivation of the connected variable optical power modules. Under the control of the corresponding light source driver, the variable optical power modules emit light in steps of 1 / m of the optical power of the fixed optical power modules in that light source unit. The variable optical power modules can compensate for the decrease in optical power of the fixed optical power modules due to aging and other issues, enabling quantification and precise control of light intensity and ensuring the long-term stable operation of the ground calibration light source system. The stability of the ground calibration light source system can be further improved by using a light intensity detection unit and a temperature control unit.

[0081] The atmospheric transmittance measurement equipment includes, but is not limited to, atmospheric lidar, solar photometer, weather balloon, and other equipment.

[0082] See Figure 1 The calibration method for the satellite optical remote sensing detector provided in Example 1 requires multiple satellite overpass measurements for ground calibration of the light source. For a single satellite overpass measurement S180, the following workflow S110-S140 can be used:

[0083] S110: Adjust the direction of the ground calibration light source in advance according to the satellite's predetermined orbit, aligning it with the satellite's overhead position. Preheat the ground calibration light source system and maintain its operating temperature. Turn on the atmospheric transmittance measurement equipment to monitor the atmospheric environment in real time, measuring the transmittance of the atmosphere at different wavelengths.

[0084] S120: Based on different calibration parameters, the ground calibration light source emits light according to the specified emission wavelength and intensity, and performs real-time monitoring, measurement and feedback compensation of the emission intensity of the ground calibration light source.

[0085] S130 and S140: After the satellite's overhead transit time ends, the ground calibration light source stops working, and real-time data such as ground calibration light source luminous intensity and atmospheric transmittance are processed, along with the output data of the satellite's optical remote sensing detector.

[0086] S150: Determine whether the measured data meets the requirements of the target indicator.

[0087] S160: If the measurement data does not meet the requirements of the calibration index of the satellite optical remote sensing detector, wait for the next satellite overhead, change the wavelength and intensity of the light source, and then perform a single measurement process S180.

[0088] S170: If multiple single measurements have been performed and the measured data has met the requirements for calibrating the satellite optical remote sensing detector, then the data output data of the satellite optical remote sensing detector is processed, and the calibration index of the satellite optical remote sensing detector is calibrated according to the test results.

[0089] The calibration parameters of the satellite optical remote sensing detector are divided into two categories: one is the parameters related to the power of the light source, and the other is the parameters that change the spectral sensitivity and spectral response curve. The two types of parameters need to be calibrated separately.

[0090] Regarding the indicators related to light source power, the conversion relationship between the luminous power of the ground calibration light source system and the radiance value received by the satellite optical remote sensing detector is as follows:

[0091]

[0092] In the formula, P is the luminous power of the ground calibration light source system, and L is the radiance value received by the satellite optical remote sensing detector (the radiance value received by the detector after launch is used as the spectral response of the satellite optical remote sensing detector to a specific wavelength light source); η atm Atmospheric transmittance is measured in real time using an atmospheric transmittance measuring device during detector calibration experiments; PSF is the detector's point spread function; ISRF is the value of the detector's spectral response curve at the selected emission wavelength (ISRF is the detector's response to each wavelength before launch); θ is the divergence angle of the ground calibration light source; A pixel Let Δλ be the ground area corresponding to one pixel of a satellite optical remote sensing detector, and let Δλ be the bandwidth of the satellite optical remote sensing detector.

[0093] For example, for the low-light channel of the medium-resolution spectral imager on the FY-3E satellite, θ is set to 100 mrad, A pixel The value is 10 6 m, Δλ is 400nm.

[0094] When performing quantitative analysis of the spectral response, multiple light sources with different specific wavelengths are used. In this case, the luminous power of the ground calibration light source system is the integral of the luminous power spectral density, expressed as:

[0095]

[0096] In the formula, λ k For the k-th specific wavelength used, Let B be the luminous power of the k-th light source unit in the ground calibration light source system when it emits light, and let Δλ be the luminous power spectral density. k Let be the spectral width of the k-th specific wavelength light source.

[0097] By combining the two formulas above, the spectral response of the satellite optical remote sensing detector to a specific wavelength light source can be obtained.

[0098] Based on Example 1, the present invention provides a calibration method for the responsivity, linearity, noise equivalent power and dynamic range of a satellite optical remote sensing detector, as detailed in Example 2.

[0099] Example 2:

[0100] Indicators related to light source power include responsivity, linearity, noise equivalent power, and dynamic range. When the indicator to be calibrated is one of the indicators related to light source power, calibration is performed in the following manner, namely, the calibration method provided in Example 2 includes the following steps:

[0101] Step a, corresponding to Figure 2 S210: Controls multiple light source units in the ground calibration light source system to emit light simultaneously, i.e., emit multiple specific wavelengths; sets i=1, and when setting the first measurement window of the i-th round of measurement, the light source emitting power P of the ground calibration light source system is... i1 Maximum optical power P max Measure the output value I of the satellite optical remote sensing detector. i1 Waiting for the next measurement window.

[0102] Step b, corresponding to Figure 2 S220 in the diagram: As the number of measurement windows increases, the luminous power of the light source is decreased sequentially; the luminous power of the light source under multiple measurement windows is denoted as P. i2 ... P ij The output value I of the corresponding satellite optical remote sensing detector was measured respectively. i2 ... I ij ; where j is an integer greater than or equal to 4.

[0103] For example, set j=4, set The output value corresponding to the satellite optical remote sensing detector is I.i2 I i3 I i4 j is set to 4, and measurement points are selected according to the relationship of 1, 2, 5, 10. This facilitates conversion to logarithmic coordinates and can most accurately obtain the equivalent power of noise.

[0104] Step c: Determine I ij (For example, I) i4 Is it less than the root mean square value of the noise? Figure 2 S230 in the middle;

[0105] If not, then let i = i + 1, and set P. i1 =P (i-1)j Then return to step b (for example, when j is 4, P) i1 =P (i-1)4 The next round of light source luminous power P i1 For the previous round P i4 ),correspond Figure 2 S240 in;

[0106] If so, then based on the light source's luminous power and its corresponding output value from the satellite's optical remote sensing detector, the indicators related to the light source's power are calibrated. Figure 2 The S250 in the middle.

[0107] The calibration response, linearity, noise equivalent power, and dynamic range are explained below.

[0108] (1) Linearity refers to the degree to which the detector output value I is proportional to the incident light power P.

[0109] When calibrating linearity, a curve of measurement points is plotted with the luminous power of the light source as the abscissa and the output value of the satellite optical remote sensing detector as the ordinate. A fitting straight line is then selected for fitting. Linearity is an index characterizing the degree of agreement or deviation between the measurement point curve and the selected fitting straight line. Linearity is represented by the relative error σ, as shown below:

[0110]

[0111] In the formula, ΔY max Y represents the maximum deviation between the measured point curve and the fitted straight line, and Y is the full-scale output value of the satellite optical remote sensing detector.

[0112] (2) Noise equivalent power is defined as the incident radiation flux when the signal-to-noise ratio is 1, that is, when the effective value of the detector output signal is equal to the root mean square value of the noise. It is used to characterize the detector's ability to detect weak signals.

[0113] When calibrating the noise equivalent power, find the light source luminous power in the measurement point data where the output value of the satellite optical remote sensing detector equals the root mean square value of the noise, and denote this light source luminous power as P. min Calculate P min The corresponding minimum radiance value L received by the satellite optical remote sensing detector min This is used to represent the noise equivalent power of the detector.

[0114] (3) Dynamic range is defined as the ratio of the maximum incident light power that the detector can respond to to the minimum incident light power.

[0115] When the relative error σ is greater than 30%, the satellite optical remote sensing detector is considered to be in a saturated state. When calibrating the dynamic range, the light source luminous power corresponding to when the output value of the satellite optical remote sensing detector reaches saturation is found in the measurement point data, and this light source luminous power is recorded as P. full Calculate P full The corresponding maximum radiance value L received by the satellite optical remote sensing detector full The dynamic range of the satellite optical remote sensing detector is expressed as:

[0116] (4) The responsivity is defined as the relationship between the detector output value I and the incident light power P when the relative error σ characterizing the linearity is less than 30%, which is used to characterize the photoelectric conversion characteristics of the detector.

[0117] When calibrating the responsivity, if the relative error σ is less than 30%, the slope ΔI / ΔP of the curve at the measurement point is used as the responsivity of the detector.

[0118] Based on Example 1, the present invention provides a calibration method for the parameters of spectral sensitivity and spectral response curve change of satellite optical remote sensing detectors, as detailed in Example 3.

[0119] Example 3:

[0120] Assuming the detector's spectral response is known to be R(λ) at the time of manufacture, the distortion of the detector's spectral response curve after satellite launch is mainly manifested in the intensity coefficient (λ) and peak wavelength shift (λ). c Given the spectral bandwidth coefficient (a), the on-orbit spectral response of the satellite optical remote sensing detector is:

[0121] R′(λ)=AR(a(λ-λ c ))

[0122] The ground-based calibration light source system used in this invention can emit multiple (three or more) monochromatic light wavelengths, with wavelengths of λ1, λ2, λ3, ..., λ... nThe parameters that change in the spectral response curve include the intensity coefficient (A) and the peak wavelength shift (λ). c ) and spectral bandwidth coefficient (a).

[0123] The calibration method provided in Example 3 includes the following steps:

[0124] Step a, corresponding to Figure 3 S310 and S320: Select a light source unit with a specific wavelength of λ1 to emit light, set the total luminous power, measure the output value I(λ1) of the satellite optical remote sensing detector, and wait for the next measurement window.

[0125] Among these, when setting the total luminous power, the satellite optical remote sensing detector's response corresponds to the upper limit of the detector's dynamic range.

[0126] Step b, corresponding to Figure 3 S330 and S340: Keeping the total luminous power constant, change the wavelength of the light source, and sequentially select specific wavelengths as λ2, ..., λ n The light source unit emits light, and the corresponding output values ​​I(λ2), ..., I(λ) of the satellite optical remote sensing detector are measured respectively. n ).

[0127] Step c, corresponding to Figure 3 S350 in the diagram: Plot (λ1,I(λ1)), (λ2,I(λ2)), ..., (λ2) on the detector's spectral response curve. n ,I(λ n Multiple measurement points are used to invert the spectral response curve changes of the satellite optical remote sensing detector based on the measurement points.

[0128] In addition, step d, corresponding to S360 and S370, may be included: if the measurement data meets the requirements for calibrating the spectral response curve of the satellite optical remote sensing instrument, the average value of multiple measurement results shall be taken as the calibration result.

[0129] The methods for retrieving changes in the spectral response curve based on the measurement points include:

[0130] (i) Based on experience, the intensity coefficient A and peak wavelength shift λ of the spectral response curve set on track are determined. c And the range of values ​​for the spectral bandwidth coefficient α.

[0131] For example, for the low-light channel of the low-light type of the medium-resolution spectral imager carried on the FY-3E satellite, A, a, and λ can be set. c The ranges of values ​​for λ are: A∈[0.8,1.2], a∈[0.8,1.2], λ c ∈[-50,50]nm.

[0132] (ii) Let the spectral response of the satellite optical remote sensing detector at the time of manufacture be denoted as R(λ), and select a set of parameters A from the above range of values. i a i and λ ci The spectral response R′ under this set of parameters is obtained. i (λ): R′ i (λ)=A i R(a i (λ-λ ci )).

[0133] (iii) For the selected multiple specific wavelengths λ1, ..., λ n The corresponding response R′ after the spectral change is obtained. i (λ1), ..., R′ i (λ n ):

[0134] R′ i (λ1)=A i R(a i (λ1-λ ci ))

[0135] ...

[0136] R′ i (λ n ) = A i R(a i (λ n -λ ci ))

[0137] For example, considering both calibration effectiveness and cost, the ground calibration light source system for the low-light channel of the medium-resolution spectral imager carried on the FY-3E satellite is designed as a three-wavelength light source for the selected three wavelengths λ1, λ2, and λ3. The response R′ after spectral changes can be obtained. i (λ1), R′ i (λ2), R′ i (λ3): R′ i (λ1)=A i R(a i (λ1-λ ci )), R′ i (λ2)=A i R(a i (λ2-λ ci )), R′ i (λ3)=A i R(a i (λ3-λ ci )).

[0138] (iv) Define an evaluation function ξ i To measure the selected A i a i , λ ci Relationship with actual spectral response changes:

[0139]

[0140] For example, for three selected wavelengths λ1, λ2, and λ3, the evaluation function ξ i as follows:

[0141]

[0142] (v) In A, a, λ c Within the range of values ​​for ξ, select parameters to find the evaluation function ξ. i A reaches its minimum value f a f , λ cf The calibrated spectral response curve R′(λ)=A was obtained. f R(a f (λ-λ cf )).

[0143] Furthermore, in order to obtain the optimal calibration result of the spectral response of the satellite optical remote sensing detector, this invention also provides a method for optimizing the selection of wavelength of the ground calibration light source system of the satellite optical remote sensing detector using the aforementioned inversion steps and methods.

[0144] The ground calibration light source system can independently emit multiple specific wavelengths, namely λ1, ..., λ2. n Monochromatic light, and λ1 < λ2 < ... < λ n The optimal selection of the aforementioned specific wavelengths is achieved using the following method:

[0145] (i) Set the value range for each specific wavelength based on the spectral response of the satellite optical remote sensing detector at the time of manufacture.

[0146] For example, assume the ground calibration light source system is designed as a three-wavelength light source with wavelengths λ1, λ2, and λ3, where λ1 < λ2 < λ3. The wavelength range can be set according to the spectral response of the satellite optical remote sensing detector to be calibrated at the time of manufacture: λ1 ∈ [λ2, λ3, λ4, λ5]. min ,λ peak ]nm, λ2∈(λ1,λ3)nm, λ3∈(λ peak ,λ max ]nm. Wherein, λ min λ is the minimum wavelength of the detector response. max λ is the maximum wavelength of the detector response. peakλ represents the peak wavelength of the detector response. For example, for the low-light channel of the low-light type of the medium-resolution spectral imager carried on the FY-3E satellite, λ can be set... min =500nm, λ peak =654nm,λ max =930nm.

[0147] (II) Based on experience, the intensity coefficient, peak wavelength drift, and spectral bandwidth coefficient of the spectral response curve set in orbit are A0, a0, and λ, respectively. c0 Among them, A0, a0 and λ c0 This represents the status of a satellite optical remote sensing detector at a certain stage.

[0148] (III) From λ1, ..., λ n Select a set of parameters λ from the range of values. 1i 、…、λ ni The corresponding parameters that change in the spectral response curve are A0, a0, and λ. c0 The response R′ at that time i (λ 1i ), ..., R′ i (λ ni ):

[0149] R′ i (λ 1i )=A0R(a0(λ 1i -λ c0 ))

[0150] ...

[0151] R′ i (λ ni )=A0R(a0(λ ni -λ c0 ))

[0152] (iv) Using the aforementioned method of retrieving the spectral response curve change based on the measurement point, R′ i (λ 1i ), ..., R′ i (λ ni Perform an inversion operation to obtain A. i a i and λ ci Define its error function for:

[0153]

[0154] In the formula, λ peak This represents the peak wavelength of the satellite optical remote sensing detector's response.

[0155] (v) In λ1, ..., λn Within the range of values, select wavelength combinations to find the error function. λ at the minimum value 1f , …, λ nf This allows for the optimal selection of a specific wavelength. If multiple wavelength combinations exist... If multiple local minima exist, the global minimum is taken as the optimal result.

[0156] For example, for the low-light channel of the low-light type of the medium-resolution spectral imager carried on the FY-3E satellite, A0 = 0.9, a0 = 0.9, λ c0 =-10. In other words, after the low-light channel of the medium-resolution spectral imager on the FY-3E satellite has been operating in orbit for a period of time, and the response intensity has attenuated by 10%, the peak wavelength has shifted 10 nm to the left, and the spectral bandwidth of the response curve has contracted by 10%, the wavelength of the ground calibration light source is optimized. 100,000 combinations are randomly selected within the range of λ1, λ2, and λ3. Figure 4 The distribution of the first wavelength λ1 when the error function is 0 shows that λ1 is concentrated at 535nm, so λ1 is chosen to be 535nm. Figure 5 The distribution of the second wavelength λ2 when the error function is 0 shows that λ2 is concentrated at 535nm, so λ2 is chosen to be 670nm. Figure 6 The distribution of the third wavelength λ3 when the error function is 0 shows that λ3 is concentrated at 875nm, so λ3 is chosen to be 875nm.

[0157] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A calibration method of a satellite optical remote sensing detector, characterized in that, Includes the following steps: Step 1: Adjust the light source direction of the ground calibration light source system so that it is aligned with the position where the satellite passes overhead; Turn on the atmospheric transmittance measuring device; Step 2: According to the calibration index, control the ground calibration light source system to emit light according to the specified emission wavelength and emission intensity, and perform the i-th round of measurement within the time range of the satellite passing overhead, and obtain the output data of the satellite optical remote sensing detector and atmospheric transmittance data corresponding to the i-th round of measurement; Step 3: After the satellite overpass time ends, determine whether the total test information meets the requirements for calibrating the indicators to be calibrated. The total test information includes the output data of the satellite optical remote sensing detector, atmospheric transmittance data, and ground calibration light source data corresponding to all rounds of testing obtained so far. If it meets the requirements, proceed to step 4. If the conditions are not met, return to step 2, change at least one of the emission wavelength and emission intensity, and then perform the next round of measurement. Step 4: Calibrate the calibration parameters of the satellite optical remote sensing detector based on the total test information; Among them, the spectral response of the satellite optical remote sensing detector at the time of manufacture is set as follows: Spectral response of satellite optical remote sensing detectors in orbit The parameters that change the detector's spectral response curve include the intensity coefficient of the spectral response curve in orbit. Peak wavelength drift and spectral bandwidth coefficient When the parameter to be calibrated is a parameter that changes the spectral response curve of the detector, calibration shall be performed in the following manner: Step a, selecting a specific wavelength as light source unit emits light, setting the total light emitting power, measuring the output value of the satellite optical remote sensing detector ; Step b, keeping the total luminous power unchanged, replacing the light source wavelength, sequentially selecting specific wavelength as the light source unit to emit light, respectively measuring the output value of the corresponding satellite optical remote sensing detector . ; Step c. plotting on a probe spectral response curve 、 a plurality of measurement points from which to invert a change in the spectral response curve of the satellite optical remote sensing probe.

2. The calibration method of a satellite optical remote-sensing detector according to claim 1, characterized in that, During the measurement process, the ground calibration light source system is controlled to maintain the operating temperature, and the luminous intensity is monitored, measured, and compensated in real time.

3. The calibration method of a satellite optical remote sensor according to claim 1, characterized in that, The ground calibration light source system comprises n light source units, where n is an integer greater than or equal to 3; each light source unit is used to generate monochromatic light of a specific wavelength, and the monochromatic light generated by different light source units corresponds to different wavelengths, with multiple wavelengths corresponding to multiple wavelengths in the response spectrum of the satellite optical remote sensing detector.

4. The calibration method for a satellite optical remote sensing detector according to claim 3, characterized in that, The indicators to be calibrated are divided into two categories: one is the indicators related to the power of the light source, and the other is the parameters that change the spectral sensitivity and spectral response curve. Regarding the indicators related to light source power, the conversion relationship between the luminous power of the ground calibration light source system and the radiance value received by the satellite optical remote sensing detector is as follows: In the formula, To calibrate the luminous power of the ground-based light source system, This refers to the radiance value received by the satellite's optical remote sensing detector. Atmospheric transmittance, Let be the detector point spread function. The value of the detector's spectral response curve at the selected emission wavelength. To calibrate the divergence angle of the light source on the ground, This represents the ground area corresponding to one pixel of a satellite optical remote sensing detector. For the bandwidth of satellite optical remote sensing detectors; When performing quantitative analysis of the spectral response, multiple light sources with different specific wavelengths are used. In this case, the luminous power of the ground calibration light source system is the integral of the luminous power spectral density, expressed as: In the formula, For the k-th specific wavelength used, The luminous power of the k-th light source unit in the ground calibration light source system when it emits light. The luminous power spectral density, Let be the spectral width of the k-th specific wavelength light source; By combining the two formulas above, the spectral response of the satellite optical remote sensing detector to a specific wavelength light source can be obtained.

5. The calibration method of a satellite optical remote sensor according to claim 3, characterized in that, Indicators related to light source power include responsivity, linearity, noise equivalent power, and dynamic range. When the indicator to be calibrated is one of these light source power-related indicators, calibration is performed using the following method: Step a: Control multiple light source units in the ground calibration light source system to emit light simultaneously; When i=1 and the first measurement window of the i-th round of measurement is set, the luminous power of the ground calibration light source system is... Maximum optical power Measure the output value of the satellite optical remote sensing detector ; Step b: As the number of measurement windows increases, the light source power is decreased sequentially; the light source power under multiple measurement windows is recorded as follows. ... The output values ​​of the corresponding satellite optical remote sensing detectors were measured respectively. ... Where j is an integer greater than or equal to 4; Step c, Judgment Is it less than the root mean square value of the noise? If not, then let... and set Then return to step b; if so, calibrate the index related to the light source power based on the light source luminous power and the corresponding output value of the satellite optical remote sensing detector.

6. The calibration method for a satellite optical remote sensing detector according to claim 5, characterized in that, In step b, set j = 4, set , , .

7. The calibration method of a satellite optical remote sensor according to claim 5, characterized in that, When calibrating linearity, the luminous power of the light source is used as the abscissa and the output value of the satellite optical remote sensing detector is used as the ordinate to plot the measurement point curve, and a fitting straight line is selected for fitting. The linearity is expressed in relative error is expressed as follows: In the formula, Y is the full-scale output value of the satellite optical remote sensor. When calibrating the noise equivalent power, find the light source luminous power in the measurement point data where the output value of the satellite optical remote sensing detector equals the root mean square value of the noise, and record this light source luminous power as _____. ,calculate The corresponding minimum radiance value received by the satellite optical remote sensing detector This is used to represent the noise equivalent power of the detector; When the relative error When the output value of the satellite optical remote sensing detector exceeds 30%, it is considered to be in a saturated state. During dynamic range calibration, the luminous power of the light source corresponding to the saturation point when the output value of the satellite optical remote sensing detector reaches saturation is found in the measurement point data, and this luminous power is recorded as... ,calculate The corresponding maximum radiance value received by the satellite optical remote sensing detector The dynamic range of the satellite optical remote sensing detector is expressed as: ; When calibrating the responsivity, the relative error is less than 30%, the slope of the measurement point curve is taken as the responsivity of the detector.

8. The method of calibrating a satellite optical remote sensor according to claim 1, wherein, The methods for realizing the change in the spectral response curve based on the measurement point include: Empirically setting the value range of intensity coefficients of spectral response curves of the track , peak wavelength drift , and spectral bandwidth coefficients ​ A set of parameters is selected from the above value range , and the spectral response under the set of parameters is obtained : For a selected plurality of specific wavelengths Corresponding resulting response after spectral change : …… A merit function is defined to measure the chosen relationship to the actual spectral response variations: exist Select parameters within the range of values ​​to find the evaluation function. When the minimum value is reached ; Obtain the calibrated spectral response curve .

9. The calibration method of a satellite optical remote sensor according to claim 8, characterized in that, The ground calibration light source system is capable of independently emitting multiple specific wavelengths. Monochromatic light, and The optimal selection of the aforementioned specific wavelengths is achieved using the following method: Based on the spectral response of the satellite optical remote sensing detector at the time of manufacture, the value range for each specific wavelength is set; The intensity coefficient, peak wavelength drift and spectral bandwidth coefficient of the spectral response curve of the track are set according to experience , and ; From a set of parameters is selected from the range of values of corresponding to the parameter of the spectral response curve variation is the response at : …… With the implementation manner of inversing the spectral response curve change according to the measuring point, the inverse operation is performed to obtain , and the error function of is defined as . ​ In the formula, is the peak wavelength of the response of the satellite optical remote-sensing detector; In the wavelength combination is selected within the value range of the wavelength, and the error function is searched for the minimum value of , and the optimal selection of the specific wavelength is realized.

Images