Universal simulation method for satellite remote sensing of top-of-atmosphere radiation

Abstract

This invention proposes a general method for simulating atmospheric top radiation from satellite remote sensing, belonging to the field of remote sensing image data processing technology. The method includes: (1) calculating the latitude, longitude, and observation geometry of the transit area; (2) constructing global environmental parameter background field data; (3) performing radiation simulations for the ocean and land regions within the transit area; (4) calculating the atmospheric top radiance of the ocean and land regions respectively; and (5) stitching together the atmospheric top radiance data of the ocean and land regions to obtain the atmospheric top radiance data for the entire image. This general method for simulating atmospheric top radiation from satellite remote sensing combines the reliability of physical models with the modeling capabilities of neural networks. It addresses the low reliability of existing satellite remote sensing atmospheric top radiation simulations while also being less time-consuming compared to full-data neural network simulations.

Description

Technical Field

[0001] This invention relates to the field of remote sensing image data processing technology, specifically to a general method for simulating satellite remote sensing atmospheric top radiation data. Background Technology

[0002] High-precision zone top radiation simulation is a key fundamental technology in the field of remote sensing. Its core value lies in providing data support for the lifecycle of satellite missions, mainly including the following two aspects:

[0003] (1) Satellite mission planning and design before launch: Before the satellite launch, the signals that the remote sensor will receive under different observation times and different ground and atmospheric conditions will be simulated to optimize the satellite orbit design, determine the optimal operating band, dynamic range and key performance indicators such as signal-to-noise ratio of the remote sensor.

[0004] (2) Remote sensing algorithm verification and development: Simulated atmospheric top radiation data can be used as "standard input" to test and verify the accuracy and robustness of data processing algorithms such as atmospheric correction and surface parameter inversion.

[0005] There are generally three methods for simulating satellite atmospheric top radiation: The first is based on radiative transfer models to accurately simulate atmospheric top radiance. This method, which simulates the impact of various physical processes in the atmosphere on radiation by numerically solving the radiative transfer equations, is the most fundamental and accurate approach. The second is based on lookup tables. This method pre-runs the physical model, calculates the corresponding atmospheric component content, and stores the results in a multidimensional dataset. Its computational speed is much faster than that of physical models and is currently the most mainstream method in engineering applications. The third is neural network methods. In recent years, with increased computing power and data accumulation, deep learning or neural networks have been used. These networks use the input parameters of the physical model as the input layer and the atmospheric top radiance as the output layer, training with a large number of samples to learn complex nonlinear mapping relationships.

[0006] However, simulations based on radiative transfer models require enormous resources and time when performing simulations on global-scale, high-resolution, hyperspectral data. Lookup table-based methods, in practical applications, convert a continuous function into a series of discrete grid points, thus requiring interpolation. This interpolation approximation of the true irrational function can mask certain phenomena. Neural network-based methods are a "black box" process, lacking clear physical interpretability and traceability; furthermore, generating a universal, high-precision simulation dataset is extremely time-consuming. Summary of the Invention

[0007] To address the technical problems of low reliability and time-consuming simulation of atmospheric top radiation by existing satellite remote sensing, this invention proposes a general method for simulating atmospheric top radiation by satellite remote sensing, which can solve the above problems.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0009] A general method for simulating atmospheric top radiation using satellite remote sensing includes:

[0010] (1) Calculate the transit area of ​​the satellite within a set time period based on the satellite orbit information, and calculate the latitude, longitude and observation geometry of the transit area;

[0011] (2) Obtain multi-source satellite seasonal aerosol optical thickness, multi-source satellite seasonal marine biological optical characteristics data, climatological atmospheric parameter data, global land cover type products and typical land cover reflectance data from the ENVI spectral library to construct global environmental parameter background field data;

[0012] (3) Based on the global environmental parameter background field data, radiation simulations are performed on the atmospheric and water parameters of the marine area and the atmospheric and surface parameters of the land area within the transit area. The marine area radiation simulation steps include obtaining the simulated water radiance. Simulated solar flare radiance Simulated Rayleigh scattering radiance and simulated aerosol scattering radiance The steps for terrestrial radiation simulation include obtaining atmospheric radiance and surface reflectance, respectively.

[0013] The simulated aerosol scattering radiance The methods for obtaining it include:

[0014] (311) Obtaining the optical thickness of aerosols in the 555nm band Select the aerosol type;

[0015] (312) will The aerosol type and observation geometry information are input into the aerosol deep neural network model, which outputs simulated aerosol scattering radiance. ;

[0016] (4) Calculate the atmospheric top radiance of the ocean region based on the simulated Rayleigh scattering radiance, simulated aerosol scattering radiance, simulated solar flare radiance and simulated water-free radiance, and calculate the atmospheric top radiance of the land region based on the radiance on the atmospheric transport path and the surface reflection radiance;

[0017] (5) The atmospheric top radiance data of the ocean area and the atmospheric top radiance data of the land area are stitched together and displayed on the same map to obtain the atmospheric top radiance data of the whole scene image.

[0018] Compared with existing technologies, the advantages and positive effects of this invention are as follows: The general satellite remote sensing atmospheric top radiation data simulation method of this invention adopts a "hybrid modeling" approach. Specifically, it uses traditional radiative transfer physical models for simulating data such as water radiance, solar flare radiance, and Rayleigh scattering radiance. For aerosol parameters, which exhibit the greatest variability and uncertainty among all atmospheric parameters, this invention introduces a modular neural network (NN) for high-precision processing. This NN module focuses on solving the modeling problem of complex optical characteristics of aerosol data. This hybrid modeling strategy significantly improves the accuracy and robustness of radiation simulation in complex scenarios. This invention cleverly combines the reliability of physical models with the modeling capabilities of neural networks, solving the problem of low reliability in existing satellite remote sensing atmospheric top radiation simulations while consuming less time compared to full-data neural network simulations.

[0019] Other features and advantages of the present invention will become clearer after reading the detailed description of the embodiments of the present invention in conjunction with the accompanying drawings. Attached Figure Description

[0020] Figure 1 This is a block diagram illustrating the principle of one embodiment of the general satellite remote sensing atmospheric top radiation data simulation method proposed in this invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be noted that in the description of this invention, terms such as "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," indicating directional or positional relationships, are based on the directional or positional relationships shown in the accompanying drawings. These are merely for ease of description and do not indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation; therefore, they should not be construed as limitations on this invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0024] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0025] Example 1, see Figure 1 As shown, the satellite remote sensing atmospheric top radiation simulation method of this embodiment includes:

[0026] (1) Calculate the satellite's transit area within a set time period based on the satellite orbit information, and calculate the latitude, longitude, and observation geometry information of the transit area. For example, the latitude, longitude, and observation geometry information of a specific area can be calculated using the set satellite parameters (such as swath width, resolution, scanning angle, etc.).

[0027] (2) Obtain multi-source satellite seasonal aerosol optical thickness, multi-source satellite seasonal marine biological optical properties data, climatological atmospheric parameter data, global land cover type products, and typical land cover reflectance data from the ENVI spectral library to construct global environmental parameter background field data. Global environmental parameter background field data may include aerosol optical thickness, Rayleigh optical thickness, wind speed, air pressure, water vapor, land surface reflectance, and ocean parameter data. All of the above data are currently commonly used known data products in atmospheric and oceanic research.

[0028] (3) Based on the background field data of global environmental parameters, radiation simulations are performed on the atmospheric and water parameters of the marine area and the atmospheric and surface parameters of the land area within the transit area. The radiation simulation steps for the marine area include obtaining the simulated water radiance. Simulated solar flare radiance Simulated Rayleigh scattering radiance and simulated aerosol scattering radiance The steps for terrestrial radiation simulation include obtaining the radiance along the atmospheric transport path and the surface reflected radiance, respectively.

[0029] (4) Calculate the top atmospheric radiance of the ocean region based on the simulated Rayleigh scattering radiance, simulated aerosol scattering radiance, simulated solar flare radiance and simulated water-free radiance, and calculate the top atmospheric radiance of the land region based on the radiance on the atmospheric transport path and the surface reflection radiance.

[0030] (5) The atmospheric top radiance data of the ocean area and the atmospheric top radiance data of the land area are stitched together and displayed on the same map to obtain the atmospheric top radiance data of the whole scene image.

[0031] This embodiment presents a general satellite remote sensing atmospheric top radiation data simulation method using a "hybrid modeling" approach. This involves employing traditional radiative transfer physical models for simulating data such as water radiance, solar flare radiance, and Rayleigh scattering radiance. For aerosol parameters, which exhibit the highest variability and uncertainty among all atmospheric parameters, this invention introduces a modular neural network (NN) for high-precision processing. This NN module focuses on modeling the complex optical characteristics of aerosol data. This hybrid modeling strategy significantly improves the accuracy and robustness of radiation simulations in complex scenarios. This invention cleverly combines the reliability of physical models with the modeling capabilities of neural networks, addressing the low reliability of existing satellite remote sensing atmospheric top radiation simulations while being less time-consuming compared to full-data neural network simulations.

[0032] In some embodiments, aerosol scattering radiance is simulated. The methods for obtaining it include:

[0033] (311) Obtaining the optical thickness of aerosols in the 555nm band Select the aerosol type.

[0034] (312) will The aerosol type and observation geometry information are input into the aerosol deep neural network model, and the aerosol deep neural network model outputs the simulated aerosol scattering radiance. .

[0035] The aerosol types used were 80 typical aerosol types (including 10 types: Arctic, Desert, Antarctic, Continental Mean, Continental Clean, Continental Polluted, Ocean Clean, Ocean Polluted, Marine Tropical, and Urban; each type was combined with 8 relative humidity levels: 0, 50, 70, 80, 90, 95, 98, and 99%) provided by the OPAC (Optical Properties of Aerosols and Clouds) database, covering various atmospheric environments such as marine and terrestrial types. Subsequently, each aerosol type was input into the 6SV (Second Simulation of a Satellite Signal in the Solar Spectrum) radiative transfer model to calculate multiple sets of simulated spectral reflectance data under different observation angles and aerosol optical thickness conditions.

[0036] The input to the aerosol deep neural network model consists of three parts: 1. 555nm aerosol optical thickness. 2. Aerosol type; 3. Observation geometry. The model output is the atmospheric aerosol radiance corresponding to each target band, used for subsequent calculation of the atmospheric top radiance information. The network uses an MLP (Multilayer Feedforward Neural Network) model. This neural network model employs standardization, cross-validation, and loss function constraints during training to avoid overfitting, enabling it to stably reconstruct multi-band aerosol contributions in actual ocean remote sensing inversion, thereby improving the accuracy and robustness of the simulation algorithm. It contains three hidden layers, each with 64-256 neurons, using ReLU as the activation function. K-fold cross-validation is also used to evaluate the model's generalization ability.

[0037] In some embodiments, the training method for the aerosol deep neural network model includes:

[0038] The training sample set is constructed by inputting various aerosol types into the 6SV radiative transfer model, calculating the radiance under different observation geometry and aerosol optical thickness conditions, and constructing the training sample set.

[0039] A deep neural network model for aerosols was constructed, with observational geometric information, aerosol optical thickness, and corresponding aerosol types as inputs and radiance as output, and the aerosol deep neural network model was trained.

[0040] In some embodiments, the satellite remote sensing atmospheric top radiation simulation method also includes establishing background field data for 10 types of aerosols globally. In step (311), the aerosol type of a specific region is selected by combining the atmospheric data and relative humidity of the simulation area.

[0041] In step (1), the provided satellite two-line report data needs to be input, and the satellite's position and velocity in the geocentric inertial coordinate system (ECI) are calculated using the SGP4 / SDP4 model. Further calculation of the nadir trajectory is performed to convert the satellite position from the ECI coordinate system to the geocentric Earth-fixed coordinate system (ECEF). This step considers factors such as Earth's rotation and polar motion. The satellite position vector in the ECEF coordinate system is then extended in the reverse direction and intersected with the Earth's reference ellipsoid (WGS84).

[0042] Step (2) involves collecting and organizing background field data on global environmental parameters. This includes acquiring seasonal aerosol optical thickness from multiple satellite sources and seasonal marine biological optical property data from multiple satellite sources, such as the phytoplankton absorption coefficient at 443 nm. Absorption coefficient of yellow substance and debris at 443 nm and backscattering coefficient at 555 nm Climate-state atmospheric parameter data, such as wind speed, air pressure, and columnar water vapor data; global land cover type products in 2023; reflectance of typical land features in the ENVI spectral library.

[0043] In some embodiments, the satellite remote sensing atmospheric top radiation simulation method also includes establishing a hyperspectral Rayleigh scattering lookup table using the MODTRAN radiative transfer model, and simulating Rayleigh scattering radiance in step (3). The methods for obtaining the data include: acquiring Rayleigh optical thickness, wind speed, and air pressure in the ocean region from global environmental parameter background field data; combining this with observational geometric information; and searching for Rayleigh scattering radiance in a hyperspectral Rayleigh scattering lookup table to simulate Rayleigh scattering radiance. .

[0044] When the simulated Rayleigh scattering radiance cannot be directly found in the hyperspectral Rayleigh scattering lookup table If the condition is met, then find the value that is closest to the condition, and then obtain the desired value through interpolation.

[0045] The simulation methods for water radiation and land radiation differ in step (3), and are therefore performed separately in this scheme. The radiance values ​​calculated during the simulation are the radiance values ​​of the corresponding pixels.

[0046] In some embodiments, a bio-optical model is constructed to simulate the radiance of water separation. ,include:

[0047] .

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] .

[0054] Where λ is the wavelength. The water absorption coefficient, The backscattering coefficient of water body Solar irradiance at the top of the atmosphere. The zenith angle of the sun. Water reflectance For subsurface remote sensing reflectance, The absorption coefficient of pure water. The absorption coefficient of phytoplankton pigments. The detritus absorption coefficient is... and Let η be the empirical coefficient. spectral slope, The backscattering coefficient of water body , , Y is obtained from global environmental parameter background field data, representing the spectral distribution of the particle backscattering coefficient, and is derived from empirical relationships.

[0055] .

[0056] In some embodiments, the Cox & Munk model is used to simulate the radiance of solar flares. ,include:

[0057] Calculate the normalized solar flare radiance :

[0058] .

[0059] Let be the Fresnel reflection coefficient, β be the zenith angle of the inclined surface, and p be the probability distribution function of the sea surface slope. This is the solar zenith angle.

[0060] Calculate the radiance of a solar flare :

[0061] .

[0062] Where T(λ) is the solar direct transmittance.

[0063] In some embodiments, the atmospheric top radiance of the ocean region is calculated. :

[0064] .

[0065] Where λ is the wavelength. Indicates the permeability of the absorbed gas. This represents the diffuse transmittance along the path from the sun to the ocean. T(λ) represents the diffuse transmittance along the path from the ocean to the sensor, and T(λ) represents the direct transmittance of sunlight.

[0066] It can be accurately calculated based on Rayleigh optics thickness, wind speed, air pressure, observation geometry, etc. To simulate aerosol scattering radiance, including the interaction between atmospheric molecules and aerosols, the aerosol type can be determined using the Angstrom index of the background field, and then calculated using an aerosol neural network model. The aerosol optical thickness at a known reference wavelength of 555 nm is also considered. The observation geometry and aerosol type can be used to calculate the aerosol optical thickness for each band.

[0067] Calculate the radiance of the top of the atmosphere over land areas , which is the sum of the radiance PATH(λ) along the atmospheric transport path and the surface reflected radiance GRFL(λ).

[0068] .

[0069] Calculate the radiance of the top of the atmosphere over land areas When the Earth's surface is assumed to be a homogeneous Lambertian surface, the atmospheric radiative transfer equation can be written as:

[0070] .

[0071] in, This represents the transmittance from the top of the atmosphere to the bottom. It represents the transmittance from the top of the atmosphere to a certain reflection point in the atmosphere. The equivalent reflectance is expressed as the background. This is expressed as atmospheric transmittance from the background to the sensor. Represented as the reflectance of the target ground object, This is expressed as the atmospheric transmittance from the target object to the sensor. It represents the equivalent reflectance, which is the background scattering to the atmosphere and the atmospheric secondary scattering. The equivalent reflectivity of the sensor is the atmospheric reflection. This represents the solar radiation flux at the top of the atmosphere.

[0072] In some embodiments, the radiance PATH(λ) along the atmospheric transport path is calculated as follows:

[0073] ;

[0074] The method for calculating the surface reflectance radiance GRFL(λ) is as follows:

[0075] ;

[0076] ;

[0077] in, Let S be the radiance of the top of the atmosphere, S be the reflectance of the balloon surface, and ρ(λ) be the reflectance of the Earth's surface. and These are the gain coefficients, This represents the solar radiation flux at the top of the atmosphere. The zenith angle of the sun. The equivalent reflectivity transmitted to the sensor by atmospheric reflection.

[0078] The calculation methods for the latitude and longitude of the transit area in step (1) include:

[0079] Calculate the longitude of the nadir point trajectory and latitude :

[0080] .

[0081] .

[0082] .

[0083] .

[0084] .

[0085] .

[0086] .

[0087] Where t represents time. , , These represent the spatial position components of the satellite in the Earth's inertial coordinate system along the inertial reference axes X, Y, and Z, respectively. This represents the position vector of the satellite relative to the Earth's center of mass at time t. This represents the position vector of the satellite relative to the Earth's center of mass at time t. , , These represent the spatial position components of the satellite in the ECEF coordinate system. The Earth's rotation angle is calculated using the Julian day JD(t) at time t. , , Let a and b represent the unit direction vectors pointing from the Earth's center to the satellite, respectively. Let a and b represent the major and minor semi-axes of the WGS-84 ellipsoid, respectively. Let a = 6378137.0 m and b = a(1 - 1 / 298.257223563). , , Let represent the spatial coordinates of the point below the star, and let e be defined as the first eccentricity.

[0088] In some embodiments, the observed geometric information in the transit area includes the longitude of each pixel. and latitude The calculation method is as follows:

[0089] .

[0090] .

[0091] .

[0092] .

[0093] .

[0094] .

[0095] .

[0096] .

[0097] Where i and j represent the i-th row and j-th column of the satellite image, respectively. Let it be the vector of the sub-satellite point at time t. Represented as a time interval, Represented as a unit vector along the orbital direction. It is represented as a unit vector in the width direction, which is perpendicular to the track direction. and These represent the ground resolution in the vertical orbital direction and the orbital direction, respectively. The number of pixels in the orbital direction is calculated using the swath width and detection time. and the number of pixels perpendicular to the track , It is represented as a vector of pixels in row i and column j in the geocentric coordinate system. , , These represent the unit direction vectors from the Earth's center to the satellite for that pixel. , , This represents the spatial coordinates of the pixel in the i-th row and j-th column.

[0098] Compared with the prior art, this embodiment has the following significant advantages:

[0099] Significantly improves simulation efficiency: Through the "pre-computed LUT + intelligent driving" model, complex radiative transfer calculations are transformed into rapid lookup and interpolation operations, realizing a leap from "days" to "minutes" in global-scale simulation, perfectly meeting the needs of rapid iteration in satellite mission planning.

[0100] Achieving a fully automated process: Users only need to provide the track file, and the system can automatically complete the preparation of all environmental parameters, the determination of land and sea types, and the driving of LUTs, completely eliminating the tedious manual parameter preparation process and making large-scale, business-oriented simulation applications possible.

[0101] Ensuring simulation realism: The environmental parameters driving the simulation are derived from real global data, enabling the output results to accurately reflect the geographical differences and spatiotemporal changes in the real world, providing high-quality and highly reliable simulation data for remote sensing algorithm verification and data assimilation research.

[0102] With a minimalist interface and high versatility, this method uses standard satellite orbit files as the sole input and outputs standard spectral data products. It can be seamlessly integrated with various satellite mission design toolchains and data processing platforms, and has strong versatility and engineering application value.

[0103] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A general method for simulating atmospheric top radiation using satellite remote sensing, characterized in that, include: (1) Calculate the transit area of ​​the satellite within a set time period based on the satellite orbit information, and calculate the latitude and longitude and observation geometry information of the transit area; (2) Obtain multi-source satellite seasonal aerosol optical thickness, multi-source satellite seasonal marine biological optical characteristics data, climatological atmospheric parameter data, global land cover type products and typical land cover reflectance data from the ENVI spectral library to construct global environmental parameter background field data; (3) Based on the global environmental parameter background field data, radiation simulations are performed on the atmospheric and water parameters of the marine area and the atmospheric and surface parameters of the land area within the transit area. The marine area radiation simulation steps include obtaining the simulated water radiance. Simulated solar flare radiance Simulated Rayleigh scattering radiance and simulated aerosol scattering radiance The steps for terrestrial radiation simulation include obtaining atmospheric radiance and surface reflectance, respectively. The simulated aerosol scattering radiance The methods for obtaining it include: (311) Obtaining the optical thickness of aerosols in the 555nm band Select the aerosol type; (312) will The aerosol type and observation geometry information are input into the aerosol deep neural network model, which outputs simulated aerosol scattering radiance. ; (4) Calculate the atmospheric top radiance of the ocean region based on the simulated Rayleigh scattering radiance, simulated aerosol scattering radiance, simulated solar flare radiance and simulated water-free radiance, and calculate the atmospheric top radiance of the land region based on the radiance on the atmospheric transport path and the surface reflection radiance; (5) The atmospheric top radiance data of the ocean area and the atmospheric top radiance data of the land area are stitched together and displayed on the same map to obtain the atmospheric top radiance data of the whole scene image.

2. The satellite remote sensing atmospheric top radiation simulation method according to claim 1, characterized in that, The training method for the aerosol deep neural network model includes: Constructing a training sample set involves inputting various aerosol types into the 6SV radiative transfer model, calculating the radiance under different observation geometry and aerosol optical thickness conditions, and constructing the training sample set. A deep neural network model for aerosols is constructed, taking the observed geometric information, aerosol optical thickness, and corresponding aerosol type as inputs and radiance as output, and then training the aerosol deep neural network model.

3. The satellite remote sensing atmospheric top radiation simulation method according to claim 2, characterized in that, The satellite remote sensing atmospheric top radiation simulation method also includes establishing background field data for 10 types of aerosols globally. In step (311), the aerosol type of a specific region is selected by combining the atmospheric data and relative humidity of the simulation area.

4. The satellite remote sensing atmospheric top radiation simulation method according to claim 1, characterized in that, The satellite remote sensing atmospheric top radiation simulation method also includes establishing a hyperspectral Rayleigh scattering lookup table using the MODTRAN radiative transfer model, and simulating Rayleigh scattering radiance in step (3). The acquisition method includes: obtaining the Rayleigh optical thickness, wind speed, and air pressure of the ocean region from global environmental parameter background field data; combining the observation geometric information; and looking up the Rayleigh scattering radiance from the hyperspectral Rayleigh scattering lookup table to simulate the Rayleigh scattering radiance. .

5. The satellite remote sensing atmospheric top radiation simulation method according to claim 1, characterized in that, Simulating water radiance by constructing a bio-optical model ,include: ; ； ； ; ; ; ; Where λ is the wavelength. The water absorption coefficient, The backscattering coefficient of water body Solar irradiance at the top of the atmosphere. The zenith angle of the sun. Water reflectance For subsurface remote sensing reflectance, The absorption coefficient of pure water. The absorption coefficient of phytoplankton pigments. The detritus absorption coefficient is... and Let η be the empirical coefficient. spectral slope, , , Y was obtained from global environmental parameter background field data, and it represents the spectral distribution of the particle backscattering coefficient.

6. The satellite remote sensing atmospheric top radiation simulation method according to claim 5, characterized in that, Calculation of solar flare radiance using the Cox & Munk model ,include: Calculate the normalized solar flare radiance : ; in, Let be the Fresnel reflection coefficient, β be the zenith angle of the inclined surface, and p be the probability distribution function of the sea surface slope. The solar zenith angle; Simulated solar flare radiance : 。 7. The satellite remote sensing atmospheric top radiation simulation method according to claim 1, characterized in that, Calculate the radiance of the top of the atmosphere in the ocean region : ; Where λ is the wavelength. Indicates the permeability of the absorbed gas. This represents the diffuse transmittance along the path from the sun to the ocean. T(λ) represents the diffuse transmittance along the path from the ocean to the sensor, and T(λ) represents the direct transmittance of sunlight. Calculate the radiance of the top of the atmosphere over land areas , which is the sum of the radiance PATH(λ) along the atmospheric transport path and the surface reflected radiance GRFL(λ).

8. The satellite remote sensing atmospheric top radiation simulation method according to claim 7, characterized in that, The method for calculating the radiance PATH(λ) along the atmospheric transport path is as follows: ； The method for calculating the surface reflectance radiance GRFL(λ) is as follows: ； ； in, Let S be the radiance of the top of the atmosphere, S be the reflectance of the balloon surface, and ρ(λ) be the reflectance of the Earth's surface. and These are the gain coefficients, This represents the solar radiation flux at the top of the atmosphere. The zenith angle of the sun. The equivalent reflectivity transmitted to the sensor by atmospheric reflection.

9. The satellite remote sensing atmospheric top radiation simulation method according to claim 1, characterized in that, The calculation methods for the latitude and longitude of the transit area in step (1) include: Calculate the longitude of the nadir point trajectory and latitude : ； ； ； ； ； ； ； Where t represents time. , , These represent the spatial position components of the satellite in the Earth's inertial coordinate system along the inertial reference coordinate axes X, Y, and Z, respectively. This represents the position vector of the satellite relative to the Earth's center of mass at time t. , , These represent the spatial position components of the satellite in the ECEF coordinate system. The Earth's rotation angle is calculated using the Julian day JD(t) at time t. , , Let a and b represent the unit direction vectors pointing from the Earth's center to the satellite, respectively. Let a and b represent the major and minor semi-axes of the WGS-84 ellipsoid, respectively. Let a = 6378137.0 m and b = a(1 - 1 / 298.257223563). , , Let represent the spatial coordinates of the point below the star, and let e be defined as the first eccentricity.

10. The satellite remote sensing atmospheric top radiation simulation method according to claim 9, characterized in that, Geometric information observed in the transit area includes pixel longitude. and latitude The calculation method is as follows: ； ； ； ； ； ； ； ； Where i and j represent the i-th row and j-th column of the satellite image, respectively. Let it be the vector of the sub-satellite point at time t. Represented as a time interval, Represented as a unit vector along the orbital direction. It is represented as a unit vector in the width direction, which is perpendicular to the track direction. and These represent the ground resolution in the vertical orbital direction and the orbital direction, respectively. The number of pixels in the orbital direction is calculated using the swath width and detection time. and the number of pixels perpendicular to the track , It is represented as a vector of pixels in row i and column j in geocentric coordinates. , , These represent the unit direction vectors from the Earth's center to the satellite for that pixel. , , This represents the spatial coordinates of the pixel in the i-th row and j-th column.

Images