Near space atmospheric profile construction method and device for optical payload radiation calibration

Abstract

The application provides a near space atmospheric profile construction method and device for optical load radiation calibration, and belongs to the technical field of remote sensing. The method comprises the following steps: constructing reference atmospheric profiles of different height layers based on at least one atmospheric reanalysis data and radio sounding data; constructing an atmospheric profile structure optimization model based on layered structure data of the reference atmospheric profiles and observed entrance pupil radiance observation values; and iteratively solving the atmospheric profile structure optimization model, taking the optimal solution of the atmospheric profile structure optimization model as the optimal layered structure of the atmospheric profile. By constructing the reference atmospheric profiles of different height layers and the atmospheric profile structure optimization model, local high-resolution atmospheric characteristics of different height layers can be determined, the accuracy of radiation transmission calculation is improved, and the reliability of radiation calibration results is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of remote sensing technology, and in particular to a method and apparatus for constructing near-space atmospheric profiles for optical payload radiometric calibration. Background Technology

[0002] In the field of remote sensing, radiometric calibration of optical payloads is a crucial step in ensuring the accuracy and consistency of remote sensing data. Its accuracy directly impacts the reliability of applications in various fields, including land surface feature retrieval, climate change monitoring, and environmental assessment. The core objective of radiometric calibration is to convert the digital signals (DN values) received by the remote sensing payload into physical quantities such as radiance values. This allows remote sensing images to reflect the true radiometric characteristics of target objects, ensuring the comparability and consistency of data across different times, spaces, and payloads, thus providing a reliable data foundation for subsequent remote sensing data analysis and scientific research. However, a lack of high-precision radiometric calibration significantly increases the uncertainty and error of optical remote sensing data in land cover classification, change detection, and parameter inversion, severely affecting the accuracy of applications and the scientific rigor of decision-making.

[0003] In spaceborne remote sensing, an atmosphere exists between the satellite payload and the ground target. The radiant energy from the surface target must pass through the atmosphere to reach the payload. The scattering, absorption, and re-radiation of radiant signals by the atmosphere exert complex modulation effects on the energy transfer process. Atmospheric state characteristics and their vertical distribution are crucial factors influencing radiative transfer, exhibiting significant differences across different geographical regions and seasons, especially in local environments with complex meteorological conditions. Atmospheric profiles describe the vertical variations of atmospheric parameters and play a central role in radiative transfer models. To accurately simulate atmospheric radiative transfer processes, high-precision atmospheric temperature, humidity, and pressure profiles are needed to accurately calculate the contribution of atmospheric radiative transfer, providing necessary reference data for atmospheric correction and radiometric calibration, and ensuring the reliable application of remote sensing data.

[0004] Currently, atmospheric radiative transfer simulation calculations in the radiometric calibration process of spaceborne optical remote sensing widely adopt the standard atmospheric model built into the atmospheric radiative transfer model. However, the standard atmospheric model is constructed based on long-term ground-based observation data of global regions. In the process of supporting the radiometric calibration of optical remote sensing payloads, it has insufficient applicability and a simple layered structure. It exhibits significant limitations when facing complex local environments, making it difficult to describe the atmospheric state and layered structure covering the entire atmospheric layer of adjacent space under local conditions. This, in turn, affects the radiometric calibration accuracy of optical remote sensing and the application value of remote sensing data. Summary of the Invention

[0005] This invention provides a method and apparatus for constructing near-space atmospheric profiles for optical payload radiometric calibration, which solves the defect in the prior art that atmospheric profiles under local conditions cannot meet the requirements for radiometric calibration accuracy.

[0006] This invention provides a method for constructing near-space atmospheric profiles for optical payload radiometric calibration, comprising:

[0007] Based on at least one atmospheric reanalysis data and radiosonde data, construct reference atmospheric profiles at different altitudes;

[0008] Based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values, an optimized atmospheric profile structure model is constructed.

[0009] The atmospheric profile structure optimization model is iteratively solved, and the optimal solution of the atmospheric profile structure optimization model is taken as the optimal layered structure of the atmospheric profile.

[0010] As one embodiment, the construction of an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values ​​includes:

[0011] Based on the layered structure data of the reference atmospheric profile, the simulated value of the entrance pupil radiance determined by the atmospheric radiative transfer model is obtained.

[0012] A cost function is constructed between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0013] As one embodiment, obtaining the simulated entrance pupil radiance value based on the atmospheric radiative transfer model according to the layered structure data of the reference atmospheric profile includes:

[0014] The layered structure data of the reference atmospheric profile are sampled to obtain atmospheric profile parameters;

[0015] Acquire observational geometric data, water vapor parameters, aerosol parameters, and surface reflectance spectral parameters;

[0016] Based on the atmospheric profile parameters, the observation geometric data, the water vapor parameters, the aerosol parameters, and the surface reflectance spectral parameters, the ultra-high spectral radiance at the entrance pupil of the spaceborne payload simulated by the atmospheric radiative transfer model is determined.

[0017] The simulated value of the entrance pupil radiance is determined by convolving the ultra-hyperspectral radiance and the spectral response function of the spaceborne hyperspectral remote sensor.

[0018] As one embodiment, the cost function for constructing the simulated entrance pupil radiance value and the observed entrance pupil radiance value includes:

[0019] Calculate the difference between the simulated value and the observed value of the entrance pupil radiance for any band of the spaceborne payload;

[0020] Based on the difference and the weighting function of any band of the satellite payload, a cost function is constructed between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0021] As one embodiment, it also includes:

[0022] If the optimal solution of the atmospheric profile structure optimization model is not obtained, the atmospheric profile structure optimization model is adjusted based on the sensitivity analysis method and the standard atmospheric profile until the optimal solution of the atmospheric profile structure optimization model is obtained.

[0023] As an example, the adjustment of the atmospheric profile structure optimization model based on sensitivity analysis and standard atmospheric profiles includes:

[0024] Based on the aforementioned sensitivity analysis method, a perturbation is applied to the standard atmospheric profile to determine the sensitivity coefficient of any band of the spaceborne payload.

[0025] Based on the ratio of the sensitivity coefficient of any band of the spaceborne payload to the sum of the sensitivity coefficients, the weighting function of any band of the spaceborne payload is determined to adjust the atmospheric profile structure optimization model.

[0026] As one embodiment, constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data includes:

[0027] Based on the radiosonde data, the observation field data and the observation field error covariance matrix at different altitudes were determined;

[0028] Based on the atmospheric reanalysis data, the background field data and background field error covariance matrix at different altitude levels are determined;

[0029] Based on the observation field error covariance matrix and the background field error covariance matrix of each height layer, a gain matrix for the corresponding height layer is constructed.

[0030] Based on the observation field data, background field data, and gain matrix of each altitude level, a reference atmospheric profile for the corresponding altitude level is constructed.

[0031] The present invention also provides a near-space atmospheric profile construction device for optical payload radiometric calibration, comprising:

[0032] The first construction module is used to construct reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data.

[0033] The second construction module is used to construct an atmospheric profile structure optimization model based on the hierarchical structure data of the reference atmospheric profile and the observed entrance pupil radiance values.

[0034] The determination module is used to iteratively solve the atmospheric profile structure optimization model and take the optimal solution of the atmospheric profile structure optimization model as the optimal layer structure of the atmospheric profile.

[0035] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a near-space atmospheric profile construction method for optical payload radiometric calibration as described above.

[0036] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a near-space atmospheric profile construction method for optical payload radiometric calibration as described above.

[0037] The near-space atmospheric profile construction method and apparatus for optical payload radiometric calibration provided by this invention facilitates the determination of local high-resolution atmospheric characteristics at different altitudes by constructing reference atmospheric profiles and atmospheric profile structure optimization models at different altitudes, thereby improving the accuracy of radiative transfer calculations and significantly enhancing the reliability of radiometric calibration results. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0039] Figure 1 This is one of the flowcharts illustrating the near-space atmospheric profile construction method for optical payload radiometric calibration provided by the present invention.

[0040] Figure 2 This is the second flowchart illustrating the near-space atmospheric profile construction method for optical payload radiometric calibration provided by this invention.

[0041] Figure 3 This is a schematic diagram of the near-space atmospheric profile construction device for optical payload radiometric calibration provided by the present invention.

[0042] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0044] Current atmospheric radiative transfer simulations in spaceborne optical remote sensing radiometric calibration processes widely employ standard atmospheric models built into existing atmospheric radiative transfer models. This approach has several drawbacks: First, standard atmospheric models based on ground-based observation data rely heavily on typical atmospheric state classifications, only capable of depicting atmospheric characteristics at large regional scales. While possessing some universality at large scales, they cannot accurately characterize atmospheric states under local conditions, failing to meet the high-precision requirements of radiometric calibration in such environments. Second, ground-based observation data exhibits single-point independent sampling characteristics, lacking good spatial representativeness and failing to characterize the spatial distribution of the atmosphere. Furthermore, the vertical stratification height of ground-based observation data is mostly within 1 kilometer, failing to cover the vertical height of adjacent space, thus making it impossible for ground-based observation data to effectively characterize the vertical distribution of the atmosphere across the entire atmospheric layer in the surrounding area. This can further affect the accuracy of radiative transfer simulations during the radiometric calibration of optical payloads in near-space. Furthermore, standard atmospheric models have relatively simple layered structures with stable vertical distributions. However, in actual local atmospheric environments, the layered structure differs significantly from standard atmospheric models due to multiple influences from geographical conditions and seasonal variations. This leads to significant deviations in the radiometric calibration results for optical payloads, thus affecting the accuracy of radiometric calibration and the reliability of remote sensing data. Finally, existing standard atmospheric models typically use fixed or preset layering standards, failing to flexibly adjust the layering structure according to actual local atmospheric conditions. The atmospheric structure at different atmospheric heights has varying impacts on the radiative transfer process. Ignoring the layering variations at key atmospheric heights (especially the boundary layer and troposphere) can result in irrational atmospheric profile layering structures, further reducing the accuracy of optical payload radiometric calibration.

[0045] To overcome the limitations of current standard atmospheric models used in atmospheric radiative transfer simulations during optical remote sensing payload radiometric calibration, which suffer from insufficient applicability and simple layered structures, and exhibit significant limitations in complex local environments, this invention proposes a method for constructing near-space atmospheric profiles for near-space local optical payload radiometric calibration. This method aims to accurately characterize the local atmospheric state characteristics and vertical layered structure used in atmospheric radiative transfer simulations during near-space optical payload radiometric calibration.

[0046] Figure 1 This is one of the flowcharts illustrating the near-space atmospheric profile construction method for optical payload radiometric calibration provided by the present invention, such as... Figure 1 As shown, the present invention provides a method for constructing near-space atmospheric profiles for optical payload radiometric calibration, which is used for optical payload radiometric calibration or atmospheric correction. The method includes the following steps.

[0047] Step S100: Based on at least one atmospheric reanalysis data and radiosonde data, construct reference atmospheric profiles for different altitude layers.

[0048] Atmospheric reanalysis data refers to atmospheric datasets generated by combining historical observational data with numerical models, resulting in high spatiotemporal resolution and physical consistency. These datasets can reproduce historical weather and climate conditions and serve as the data foundation for studying climate change and analyzing extreme events. They are widely used in meteorology, hydrology, ecology, and other fields. This invention does not limit the type of atmospheric reanalysis data; it uses ERA5 and MERRA-3 atmospheric reanalysis data as examples. ERA5 atmospheric reanalysis data is the fifth-generation global climate reanalysis product provided by the European Centre for Medium-Range Weather Forecasts (ECMWF). It covers a 31km grid on Earth and uses 137 levels to resolve the atmosphere from the surface to an altitude of 80km, including parameters such as relative humidity, boundary layer height, temperature, wind direction, and surface pressure. MERRA-3 (Modern-Era Retrospective Analysis for Research and Applications, Version 3) is a global atmospheric reanalysis dataset developed by NASA's Goddard Space Flight Center, providing global meteorological data from 1980 to the present, including various meteorological parameters such as temperature, humidity, wind speed, radiation, and precipitation.

[0049] Radiosonde data includes meteorological data acquired by radiosondes, which are instruments mounted on weather balloons. As the balloon ascends, the radiosonde measures and records atmospheric meteorological parameters at different altitudes.

[0050] Atmospheric reanalysis data includes atmospheric profile data at different altitudes. By correcting the atmospheric profile data at different altitudes using the background field error of the atmospheric reanalysis data and the observation field error of the radiosonde data, reference atmospheric profiles at different altitudes are obtained, resulting in atmospheric profiles with higher spatial resolution and more accurate physical characterization.

[0051] Step S200: Based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values, construct an optimized atmospheric profile structure model.

[0052] Atmospheric profile layering data primarily includes the physical properties of the atmosphere at different altitudes, such as temperature, humidity, air pressure, and gas composition. Entrance pupil radiance refers to the brightness reflected or radiated from the Earth's surface in the direction pointed to by the spacecraft's payload and detectors. It is usually represented by L and the unit is 1. (Watts per square meter per solid radian). The observed entrance pupil radiance value refers to the entrance pupil radiance value of the spaceborne payload. The spaceborne payload refers to the various instruments and equipment carried on spacecraft (such as satellites) to achieve their specific scientific, technological or application tasks, including cameras, sensors, radars, communication equipment, scientific instruments, etc.

[0053] The layered structure data of the reference atmospheric profile is used to determine the entrance pupil radiance value simulated by the atmospheric radiative transfer model, i.e., the simulated entrance pupil radiance value. The atmospheric profile structure optimization model is a cost function between the simulated entrance pupil radiance value and the observed entrance pupil radiance value. The atmospheric radiative transfer model is a physical model used to describe and calculate the absorption and scattering of radiative signals as they propagate through the atmosphere.

[0054] Step S300: Iteratively solve the atmospheric profile structure optimization model and take the optimal solution of the atmospheric profile structure optimization model as the optimal layered structure of the atmospheric profile.

[0055] The atmospheric profile structure optimization model is iteratively solved to minimize the error between the simulated and observed entrance pupil radiance values ​​or to converge to a certain value, thus obtaining the optimal solution of the atmospheric profile structure optimization model.

[0056] It is understood that by constructing reference atmospheric profiles and atmospheric profile structure optimization models at different altitudes, this invention facilitates the determination of local high-resolution atmospheric characteristics at different altitudes, improves the accuracy of radiative transfer calculations, and significantly enhances the reliability of radiometric calibration results.

[0057] Based on the above embodiments, as an optional embodiment, the step of constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data includes steps S110-S140.

[0058] Step S110: Based on the radiosonde data, determine the observation field data and the observation field error covariance matrix at different altitudes.

[0059] Step S120: Based on the atmospheric reanalysis data, determine the background field data and background field error covariance matrix at different altitude levels.

[0060] Step S130: Construct the gain matrix for the corresponding height layer based on the observation field error covariance matrix and the background field error covariance matrix for each height layer.

[0061] Step S140: Construct the reference atmospheric profile for the corresponding altitude layer based on the observation field data, the background field data, and the gain matrix for each altitude layer.

[0062] In step S110, the unified formula for calculating the observation field error covariance matrix is ​​as follows:

[0063] (1)

[0064] in, Represents the observation error covariance matrix. d This represents the distance between any two points; L This indicates the length of the relevant scale, usually set to 100 km; This represents the standard deviation of the observation field error, calculated from radiosonde data.

[0065] In step S120, the unified formula for calculating the background field error covariance matrix is ​​as follows:

[0066] (2)

[0067] in, Represents the background field error covariance matrix. The standard deviation representing the background field error is calculated from atmospheric reanalysis data.

[0068] In step S130, the unified calculation formula for the gain matrix is ​​as follows:

[0069] (3)

[0070] in, The gain matrix represents the magnitude of correction to the background field data. This represents the observation operator matrix, used to map background field data to the observation space; typically, the identity matrix is ​​chosen.

[0071] In step S140, the unified calculation formula for the reference atmospheric profile is as follows:

[0072] (4)

[0073] in, This represents the corrected atmospheric profile, i.e., the reference atmospheric profile; This represents background field data, i.e., reanalysis data with larger errors; This represents the observation field data, i.e., the reanalysis data with smaller errors.

[0074] Substitute the observation field data and observation field error covariance matrix at different altitude levels, and the background field data and background field error covariance matrix at different altitude levels into formulas (1) and (2) respectively, and then substitute formulas (1) and (2) into formulas (3) and (4) to obtain the optimal interpolation results of the reference atmospheric profiles with high precision and high vertical resolution at different altitude levels.

[0075] Understandably, this invention utilizes different reanalysis data and ground-based radiosonde data to evaluate the background error values ​​of different reanalysis data at different altitudes, constructs a covariance matrix of the background error, leverages the advantages of the hierarchical structure of different reanalysis data and the accuracy advantages exhibited by the reanalysis data in different altitude regions, and employs an optimal interpolation method to achieve dynamic adjustment and error minimization at different altitude levels, correcting and fusing the atmospheric profiles of the reanalysis data, thereby giving the atmospheric profiles higher spatial resolution and more accurate physical characterization.

[0076] Based on the above embodiments, as an optional embodiment, the step of constructing an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values ​​includes steps S210-S220.

[0077] Step S210: Based on the layered structure data of the reference atmospheric profile, obtain the simulated value of the entrance pupil radiance determined by the atmospheric radiative transfer model.

[0078] Optionally, obtaining the simulated entrance pupil radiance value based on the atmospheric radiative transfer model according to the layered structure data of the reference atmospheric profile includes steps S211-S214.

[0079] Step S211: Sample the layered structure data of the reference atmospheric profile to obtain atmospheric profile parameters.

[0080] The high-resolution vertical atmospheric profile obtained from formula (4) is used as a reference. Different atmospheric profiles (different structures, i.e., atmospheric profiles of different layers) are obtained by continuously adjusting the reference atmospheric profile. ,Right now This represents the layered structure of the atmospheric profile. The expressions for the atmospheric profile parameters are as follows:

[0081] (5)

[0082] in, Indicates in Atmospheric profile parameters obtained under atmospheric profile layering parameters; Represented as atmospheric profile stratification parameters. Indicates altitude, Indicates the number of atmospheric profile layers, relative to a reference atmospheric profile. Sampling Atmospheric profile of the layer, This represents the height of each layer. Given an initial... For reference atmospheric profile Initial atmospheric profile parameters can be obtained by sampling. .

[0083] Step S212: Obtain observation geometric data, water vapor parameters, aerosol parameters, and surface reflectance spectral parameters.

[0084] The observation geometry data includes the cosine of the observed zenith angle. Water vapor parameters include atmospheric transmittance between the sensor and the sun. The sensor refers to the Earth observation payload / camera mounted on a near-space platform or satellite platform. Aerosol parameters include atmospheric hemispherical albedo.

[0085] Step S213: Determine the ultra-high spectral radiance at the entrance pupil of the spaceborne payload simulated by the atmospheric radiative transfer model based on the atmospheric profile parameters, the observation geometric data, the water vapor parameters, the aerosol parameters, and the surface reflectance spectral parameters.

[0086] The atmospheric profile parameters, the observed geometric data, the water vapor parameters, the aerosol parameters, and the surface reflectance spectral parameters are input into the MODTRAN software. MODTRAN (MODerate Resolution Atmospheric Transmission) is a widely used meteorological and atmospheric radiative transfer model used to predict the transmission characteristics of light radiation in the atmosphere.

[0087] The formula for calculating ultra-high spectral radiance is shown below:

[0088] (6)

[0089] in, Indicates ultra-high spectral radiance; This indicates path radiation, or path radiation. This represents the total downward radiation from the sun; This represents the atmospheric transmittance between the sensor and the sun; ρ This represents the surface reflectance parameter. S Indicates atmospheric hemispheric albedo; It represents the cosine value of the observed zenith angle in observation geometry. , and S All are related to atmospheric profile parameters.

[0090] Step S214: Perform convolution calculation on the ultra-high spectral radiance and the spectral response function of the spaceborne hyperspectral remote sensor to determine the simulated value of the entrance pupil radiance.

[0091] A spaceborne hyperspectral remote sensor is a satellite sensor that uses hyperspectral imaging technology to acquire information about objects on the Earth's surface within a relatively narrow spectral range.

[0092] The formula for calculating the simulated value of entrance pupil radiance is as follows:

[0093] (7)

[0094] in, This is the simulated value of the entrance pupil radiance. To simulate the hyperspectral radiance (0.1 cm⁻¹) at the entrance pupil of a spaceborne payload using the MODTRAN atmospheric radiative transfer model. -1 ), Let λ1 and λ2 represent the starting and ending wavelengths of the spectral response function of the spaceborne hyperspectral remote sensor, respectively.

[0095] Atmospheric parameters, surface reflectance parameters, water vapor parameters, aerosol parameters, and observation geometric parameters are input into the MODTRAN software. The MODTRAN software then simulates and obtains high-spectral-resolution radiance data according to formula (6). According to formula (7), the radiance of high spectral resolution is... And the spectral response function (SRF) of the sensor in each band. Convolution is performed to obtain simulated entrance pupil radiance data for each channel of the sensor. .

[0096] Step S220: Construct a cost function between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0097] Optionally, the step of constructing the cost function between the simulated entrance pupil radiance value and the observed entrance pupil radiance value includes steps S221-S222.

[0098] Step S221: Calculate the difference between the simulated value and the observed value of the entrance pupil radiance for any band of the spaceborne payload.

[0099] Step S222: Based on the difference and the weighting function of any band of the spaceborne payload, construct a cost function between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0100] The expression for the atmospheric profile structure optimization model is shown below:

[0101] (8)

[0102] in, The number of bands for the payload; and The convolution of the spectral response functions of the spaceborne payloads with the first-order spectral response function is respectively the second-order spectral response function. Observed and simulated values ​​of entrance pupil radiance for the band; Representing the The weighting function for a band can be a fixed value or an adjustable value; Parameters representing the layered structure of the atmospheric profile.

[0103] Simulated entrance pupil radiance data were obtained by calculating the adjusted atmospheric profile using MODTRAN software (Equations (6) and (7)). Through continuous adjustment That is, continuously adjusting the layered structure of the atmospheric profile until the simulated entrance pupil radiance is achieved. and the measured entrance pupil radiance value Minimum, that is, until formula (8) reaches its minimum value. To achieve optimal atmospheric profile stratification, formula (8) also considers the different sensitivities of different bands to atmospheric parameters, i.e., some bands are sensitive to water vapor or other atmospheric parameters, and introduces weights accordingly. Characterize the sensitivity of atmospheric parameters in different wavebands.

[0104] It is understood that this invention employs a hierarchical optimization mechanism to construct an atmospheric profile structure optimization model, accurately characterizing the atmospheric profile structure at different altitude levels, enhancing the accuracy of radiative transfer calculations, and significantly improving the reliability of radiometric calibration results. This invention has broad applicability and scalability, providing a solid technical foundation for remote sensing applications across multiple platforms and scenarios, and providing reliable data support for the radiometric calibration of spaceborne optical payloads.

[0105] like Figure 2 As shown, based on the above embodiments, as an optional embodiment, the present invention further includes the following steps.

[0106] Step S400: If the optimal solution of the atmospheric profile structure optimization model is not obtained, the atmospheric profile structure optimization model is adjusted based on the sensitivity analysis method and the standard atmospheric profile until the optimal solution of the atmospheric profile structure optimization model is obtained.

[0107] Optionally, the adjustment of the atmospheric profile structure optimization model based on the sensitivity analysis method and the standard atmospheric profile includes steps S410-S420.

[0108] Step S410: Based on the sensitivity analysis method, apply a perturbation to the standard atmospheric profile to determine the sensitivity coefficient of any band of the spaceborne payload.

[0109] Step S420: Based on the ratio of the sensitivity coefficient of any band of the spaceborne payload to the sum of the sensitivity coefficients, determine the weighting function of any band of the spaceborne payload to adjust the atmospheric profile structure optimization model.

[0110] In step S410, a ±5% perturbation is applied to the temperature, humidity, and pressure parameters in the standard atmospheric profile using a sensitivity analysis method, and the radiance variation is simulated using MODTRAN. ).

[0111] The unified formula for calculating the sensitivity coefficient is as follows:

[0112] (9)

[0113] in, To simulate the change in radiance, This is the disturbance amount. By applying a disturbance to the standard atmospheric profile of different bands and substituting it into the above formula, the sensitivity coefficients of different bands can be obtained.

[0114] The formula for calculating the weighting function of any band of the spaceborne payload is as follows:

[0115] (10)

[0116] in, Let be the weighting function for the j-th band of the spaceborne payload. denoted as the sensitivity coefficient of the j-th band of the spaceborne payload.

[0117] Understandably, this invention introduces consistent local perturbations into different altitude ranges of the standard atmospheric profile, such as applying ±5% perturbations, and uses MODTRAN to simulate and calculate the radiance changes before and after the perturbation. It calculates the mean value of the radiance changes in each band, clarifies the sensitive altitude ranges of atmospheric parameters, and performs densification optimization on the atmospheric profile in the sensitive altitude ranges. This allows for adjustments to the atmospheric profile structure optimization model. Based on the actual acquired remote sensing data from the spaceborne payload and the surface reflectivity parameters and observation geometry at the time of transit, the Bayesian optimization algorithm is used to iteratively optimize and solve the structure optimization model. The densification optimization continuously optimizes the atmospheric profile layering in the sensitive altitude ranges until the model converges to a stable value. When the model satisfies the optimal solution, the corresponding result is the optimal atmospheric profile layering structure.

[0118] In summary, this invention constructs high-precision atmospheric profile data by fusing multi-source reanalysis data and ground-based radiosonde data, and combines this with an atmospheric profile structure optimization model to provide a more reliable and accurate atmospheric profile reference for the radiometric calibration of spaceborne optical payloads. Compared to traditional standard atmospheric profile models, the near-space atmospheric profile construction method for local optical payload radiometric calibration proposed in this invention not only considers the dynamic characteristics of the local environment but also achieves higher precision optimization in vertical structural stratification, constructing a high-resolution local atmospheric profile that provides a more accurate and optimized atmospheric profile reference for the radiometric calibration of optical payloads.

[0119] The near-space atmospheric profile construction apparatus for optical payload radiometric calibration provided by the present invention will be described below. The near-space atmospheric profile construction apparatus for optical payload radiometric calibration described below can be referred to in correspondence with the near-space atmospheric profile construction method for optical payload radiometric calibration described above.

[0120] Figure 3 This is a schematic diagram of the near-space atmospheric profile construction device for optical payload radiometric calibration provided by the present invention, as shown below. Figure 3 As shown, the present invention also provides a near-space atmospheric profile construction device for optical payload radiometric calibration, comprising the following modules.

[0121] The first construction module 310 is used to construct reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data.

[0122] The second construction module 320 is used to construct an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values.

[0123] The determination module 330 is used to iteratively solve the atmospheric profile structure optimization model and take the optimal solution of the atmospheric profile structure optimization model as the optimal layer structure of the atmospheric profile.

[0124] As an example, the first building module 310 is further configured to:

[0125] Based on the layered structure data of the reference atmospheric profile, the simulated value of the entrance pupil radiance determined by the atmospheric radiative transfer model is obtained.

[0126] A cost function is constructed between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0127] As an example, the first building module 310 is further configured to:

[0128] The layered structure data of the reference atmospheric profile are sampled to obtain atmospheric profile parameters;

[0129] Acquire observational geometric data, water vapor parameters, aerosol parameters, and surface reflectance spectral parameters;

[0130] Based on the atmospheric profile parameters, the observation geometric data, the water vapor parameters, the aerosol parameters, and the surface reflectance spectral parameters, the ultra-high spectral radiance at the entrance pupil of the spaceborne payload simulated by the atmospheric radiative transfer model is determined.

[0131] The simulated value of the entrance pupil radiance is determined by convolving the ultra-hyperspectral radiance and the spectral response function of the spaceborne hyperspectral remote sensor.

[0132] As an example, the first building module 310 is further configured to:

[0133] Calculate the difference between the simulated value and the observed value of the entrance pupil radiance for any band of the spaceborne payload;

[0134] Based on the difference and the weighting function of any band of the satellite payload, a cost function is constructed between the simulated entrance pupil radiance value and the observed entrance pupil radiance value to obtain the atmospheric profile structure optimization model.

[0135] As one embodiment, it also includes:

[0136] An adjustment module is used to adjust the atmospheric profile structure optimization model based on sensitivity analysis and a standard atmospheric profile if the optimal solution of the atmospheric profile structure optimization model is not obtained, until the optimal solution of the atmospheric profile structure optimization model is obtained.

[0137] As one embodiment, the adjustment module is further configured to:

[0138] Based on the aforementioned sensitivity analysis method, a perturbation is applied to the standard atmospheric profile to determine the sensitivity coefficient of any band of the spaceborne payload.

[0139] Based on the ratio of the sensitivity coefficient of any band of the spaceborne payload to the sum of the sensitivity coefficients, the weighting function of any band of the spaceborne payload is determined to adjust the atmospheric profile structure optimization model.

[0140] As one embodiment, constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data includes:

[0141] Based on the radiosonde data, the observation field data and the observation field error covariance matrix at different altitudes were determined;

[0142] Based on the atmospheric reanalysis data, the background field data and background field error covariance matrix at different altitude levels are determined;

[0143] Based on the observation field error covariance matrix and the background field error covariance matrix of each height layer, a gain matrix for the corresponding height layer is constructed.

[0144] Based on the observation field data, background field data, and gain matrix of each altitude level, a reference atmospheric profile for the corresponding altitude level is constructed.

[0145] The near-space atmospheric profile construction apparatus for optical payload radiometric calibration provided by the present invention is used to execute the near-space atmospheric profile construction method for optical payload radiometric calibration described in any of the above embodiments. It has the same technical effects as the near-space atmospheric profile construction method for optical payload radiometric calibration, and will not be described in detail here.

[0146] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 410, a communications interface 420, a memory 430, and a communication bus 440, wherein the processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute the near-space atmospheric profile construction method for optical payload radiometric calibration. The method includes: constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data; constructing an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profiles and observed entrance pupil radiance values; iteratively solving the atmospheric profile structure optimization model, and taking the optimal solution of the atmospheric profile structure optimization model as the optimal layered structure of the atmospheric profile.

[0147] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0148] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the near-space atmospheric profile construction method for optical payload radiometric calibration provided by the above methods. The method includes: constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data; constructing an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profiles and the observed entrance pupil radiance values; iteratively solving the atmospheric profile structure optimization model, and taking the optimal solution of the atmospheric profile structure optimization model as the optimal layered structure of the atmospheric profile.

[0149] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a near-space atmospheric profile construction method for optical payload radiometric calibration provided by the methods described above. The method includes: constructing reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data; constructing an atmospheric profile structure optimization model based on the layered structure data of the reference atmospheric profiles and observed entrance pupil radiance values; iteratively solving the atmospheric profile structure optimization model, and taking the optimal solution of the atmospheric profile structure optimization model as the optimal layered structure of the atmospheric profile.

[0150] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0151] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing near-space atmospheric profiles for optical payload radiometric calibration, characterized in that, include: Based on at least one atmospheric reanalysis data and radiosonde data, construct reference atmospheric profiles at different altitudes; Based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values, an optimized atmospheric profile structure model is constructed. The atmospheric profile structure optimization model is iteratively solved to minimize the error between the simulated and observed entrance pupil radiance values ​​or to converge to a certain value, thereby obtaining the optimal solution of the atmospheric profile structure optimization model. The optimal solution of the atmospheric profile structure optimization model is taken as the optimal layered structure of the atmospheric profile. The atmospheric profile structure optimization model is constructed based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values, including: Based on the layered structure data of the reference atmospheric profile, the simulated value of the entrance pupil radiance determined by the atmospheric radiative transfer model is obtained. Calculate the difference between the simulated value and the observed value of the entrance pupil radiance for any band of the spaceborne payload; Based on the difference and the weighting function of any band of the spaceborne payload, a cost function is constructed between the simulated value of the entrance pupil radiance and the observed value of the entrance pupil radiance, thus obtaining the atmospheric profile structure optimization model; The expression for the atmospheric profile structure optimization model is shown below: ； in, The number of bands for the payload; and The convolution of the spectral response functions of the spaceborne payloads with the first-order spectral response function is respectively the second-order spectral response function. Observed and simulated values ​​of entrance pupil radiance for the band; Representing the The weighting function for the band; Parameters representing the layered structure of the atmospheric profile.

2. The near-space atmospheric profile construction method for optical payload radiometric calibration according to claim 1, characterized in that, The step of obtaining the simulated entrance pupil radiance value based on the atmospheric radiative transfer model using the layered structure data of the reference atmospheric profile includes: The layered structure data of the reference atmospheric profile are sampled to obtain atmospheric profile parameters; Acquire observational geometric data, water vapor parameters, aerosol parameters, and surface reflectance spectral parameters; Based on the atmospheric profile parameters, the observation geometric data, the water vapor parameters, the aerosol parameters, and the surface reflectance spectral parameters, the ultra-high spectral radiance at the entrance pupil of the spaceborne payload simulated by the atmospheric radiative transfer model is determined. The simulated value of the entrance pupil radiance is determined by convolving the ultra-hyperspectral radiance and the spectral response function of the spaceborne hyperspectral remote sensor.

3. The near-space atmospheric profile construction method for optical payload radiometric calibration according to claim 2, characterized in that, Also includes: If the optimal solution of the atmospheric profile structure optimization model is not obtained, the atmospheric profile structure optimization model is adjusted based on the sensitivity analysis method and the standard atmospheric profile until the optimal solution of the atmospheric profile structure optimization model is obtained.

4. The near-space atmospheric profile construction method for optical payload radiometric calibration according to claim 3, characterized in that, The adjustment of the atmospheric profile structure optimization model based on sensitivity analysis and standard atmospheric profiles includes: Based on the aforementioned sensitivity analysis method, a perturbation is applied to the standard atmospheric profile to determine the sensitivity coefficient of any band of the spaceborne payload. Based on the ratio of the sensitivity coefficient of any band of the spaceborne payload to the sum of the sensitivity coefficients, the weighting function of any band of the spaceborne payload is determined to adjust the atmospheric profile structure optimization model.

5. The near-space atmospheric profile construction method for optical payload radiometric calibration according to claim 1, characterized in that, The construction of reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data includes: Based on the radiosonde data, the observation field data and the observation field error covariance matrix at different altitudes were determined; Based on the atmospheric reanalysis data, the background field data and background field error covariance matrix at different altitude levels are determined; Based on the observation field error covariance matrix and the background field error covariance matrix of each height layer, a gain matrix for the corresponding height layer is constructed. Based on the observation field data, background field data, and gain matrix of each altitude level, a reference atmospheric profile for the corresponding altitude level is constructed.

6. A near-space atmospheric profile construction device for optical payload radiometric calibration, characterized in that, include: The first construction module is used to construct reference atmospheric profiles at different altitudes based on at least one atmospheric reanalysis data and radiosonde data. The second construction module is used to construct an atmospheric profile structure optimization model based on the hierarchical structure data of the reference atmospheric profile and the observed entrance pupil radiance values. The determination module is used to iteratively solve the atmospheric profile structure optimization model so that the error between the simulated value of entrance pupil radiance and the observed value of entrance pupil radiance reaches the minimum value or converges to a certain value, thereby obtaining the optimal solution of the atmospheric profile structure optimization model. The optimal solution of the atmospheric profile structure optimization model is taken as the optimal layered structure of the atmospheric profile. The atmospheric profile structure optimization model is constructed based on the layered structure data of the reference atmospheric profile and the observed entrance pupil radiance values, including: Based on the layered structure data of the reference atmospheric profile, the simulated value of the entrance pupil radiance determined by the atmospheric radiative transfer model is obtained. Calculate the difference between the simulated value and the observed value of the entrance pupil radiance for any band of the spaceborne payload; Based on the difference and the weighting function of any band of the spaceborne payload, a cost function is constructed between the simulated value of the entrance pupil radiance and the observed value of the entrance pupil radiance, thus obtaining the atmospheric profile structure optimization model; The expression for the atmospheric profile structure optimization model is shown below: ； in, The number of bands for the payload; and The convolution of the spectral response functions of the spaceborne payloads with the first-order spectral response function is respectively the second-order spectral response function. Observed and simulated values ​​of entrance pupil radiance for the band; Representing the The weighting function for the band; Parameters representing the layered structure of the atmospheric profile.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the near-space atmospheric profile construction method for optical payload radiometric calibration as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the near-space atmospheric profile construction method for optical payload radiometric calibration as described in any one of claims 1-5.

Images