Quantitative retrieval method of aerosol remote sensing based on nonlinear land surface reflectance model

Abstract

The present application relates to the technical field of Internet of Things, and more particularly to an aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectivity model, wherein samples with AOD values less than 0.2 are screened out from auxiliary satellite-load AOD data sets as background values for atmospheric correction of target loads, surface reflectivity in different channels is obtained by means of an atmospheric radiation transmission model in cycles of solar, satellite observation angle and azimuth variation, a mathematical model suitable for each set of surface reflectivity relationship is established by analyzing the variation law and significance of surface reflectivity between channels with respect to key parameters such as observation geometry, surface type and surface vegetation parameters in different seasons, the nonlinear variation law of surface reflectivity with respect to observation geometry and surface parameters is considered, and the AOD inversion precision is improved. Meanwhile, the variation law of surface reflectivity with respect to surface parameters is analyzed, and the variation law of surface reflectivity is better described, which can be used for aerosol remote sensing quantitative inversion of various satellite remote sensing data.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and more specifically, to a quantitative inversion method for aerosol remote sensing based on a nonlinear surface reflectance model. Background Technology

[0002] In today's era of Earth observation, the entire process, from the theory of radiative transfer of atmospheric particles to satellite detection technology, aerosol inversion principles, and the commercialization of products, is relatively mature. A wealth of globally covered aerosol optical depth (AOD) products developed using satellite remote sensing are readily available, providing data support for studying air quality at any scale. Aerosol inversion algorithms primarily utilize the classic Dark Target (DT) algorithm and the Deep Blue (DB) algorithm (Hsu et al., 2013; Levy et al., 2013). The former is used for dark pixels, and the latter for bright surface pixels.

[0003] The DT algorithm assumes that aerosol optical thickness generally decreases with increasing wavelength, and that the shortwave near-infrared channel is almost unaffected by atmospheric signals. It also assumes a linear statistical relationship between the surface reflectance (0.47 and 0.65) in the visible light channel and the apparent reflectance or surface reflectance in the shortwave near-infrared channel. Given the surface reflectance, aerosol inversion can then be achieved. This idea has been applied by the MODIS team to the operational aerosol algorithms for Terra / MODIS and Aqua / MODIS. The DT algorithm has also been ported to many sensors with similar observation capabilities, such as Suomi-NPP / VIIRS and FY3D / MERSI-II. Some researchers have also applied the DT algorithm to the H8 geostationary satellite.

[0004] The Dark Blue Algorithm (DB) is primarily designed for bright surfaces in deserts, arid regions, and semi-arid areas. It leverages the low surface reflectance in the deep blue (or near-ultraviolet) bands, where aerosol signals contribute significantly, to select the minimum value from long-term apparent reflectance data. After atmospheric correction, a surface reflectance database is obtained. Over the past few decades, numerous studies have been conducted based on the DT algorithm, with establishing surface reflectance models tailored to actual load conditions being a significant improvement. Overall, the newly improved surface reflectance model still exhibits a linear relationship with changes in observation geometry and vegetation indices. When there are significant changes in the surface type of neighboring pixels or the observation angle, abrupt changes in surface reflectance occur, leading to anomalies in the quantitatively retrieved AOD. Summary of the Invention

[0005] To address the aforementioned issues, this invention aims to provide a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model. This method selects samples with a reflectance less than 0.2 from the auxiliary AOD dataset as background values ​​for atmospheric correction of the target payload, and iterates through changes in solar, satellite observation angles, and azimuth angles. Using an atmospheric radiative transfer model, it obtains the surface reflectance of different channels. By analyzing the variation patterns and significance of surface reflectance between channels under different seasons with key parameters such as observation geometry, surface type, and surface vegetation parameters, a mathematical model applicable to the surface reflectance relationship of each group is established. This model accurately characterizes the variation patterns of surface reflectance with seasons and angles, thereby achieving high-precision AOD inversion.

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

[0007] According to a first aspect of the present invention, a quantitative inversion method for aerosol remote sensing based on a nonlinear surface reflectance model is provided, the method comprising the following steps:

[0008] 1. Read the observation time of the target satellite-payload orbit-by-orbit first-level data and the observation time of the auxiliary satellite-payload AOD product. If the time difference between the two is less than 5 minutes, proceed with the following process; otherwise, terminate.

[0009] 2. Select samples with AOD values ​​less than 0.2 from the auxiliary satellite-payload AOD dataset as background values ​​for atmospheric correction of the target payload;

[0010] 3. Read the count value of each pixel in the target satellite-payload observation orbital level one data and convert it into apparent reflectance of the top of the atmosphere. After correction, obtain the apparent reflectance after gas absorption correction.

[0011] 4. Read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is not marked as "ice and snow", then read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is marked as "definitely clear sky", proceed with the following process; otherwise, terminate.

[0012] 5. Read the reflectivity of multiple channels of the target satellite-payload. If it does not meet the identification criteria for inland water bodies, proceed with the following process; otherwise, terminate.

[0013] 6. Read the four datasets of longitude, latitude, land and sea identification and AOD in the auxiliary satellite-payload AOD product, filter out the effective pixels of the target satellite-payload orbital level data within the preset range, and average the effective pixels of the selected target satellite-payload for each pixel of the auxiliary satellite-payload to obtain the spatiotemporally matched observation records of the target satellite-payload and the auxiliary satellite-payload.

[0014] 7. Based on the observation records, the surface reflectance of the target satellite-payload is obtained by cyclically changing the solar and satellite observation angles and azimuth angles. The breakpoints for the observation angle classification settings and the breakpoints for the surface parameter classification settings are determined through analysis, and a surface reflectance model is established.

[0015] As a further aspect of the present invention, samples with AOD values ​​less than 0.2 are selected from the auxiliary satellite-payload AOD dataset, including:

[0016] Read the AOD per orbit product of the auxiliary satellite-payload. For each pixel, perform the following judgment condition. If the condition is met, continue the following process; otherwise, terminate.

[0017] AOD auxiliary <0.2

[0018] Among them, AOD auxiliary To supplement satellite-payload AOD data, 0.2 is the threshold for selecting samples.

[0019] As a further aspect of the present invention, reading the count value of each pixel in the target satellite-payload observation orbit-by-orbit first-level data and converting it into the apparent reflectance of the top of the atmosphere includes:

[0020] Read the count values ​​of each pixel in multiple visible light channels, near-infrared channels, and short-wave near-infrared channels from the target satellite-payload observation orbital level data, and convert the count values ​​into the apparent reflectance of the top of the atmosphere through calibration coefficients and physical quantity conversion formulas;

[0021] The visible light channel includes blue light, green light and red light bands, and the near-infrared channel includes two near-infrared bands.

[0022] As a further aspect of the present invention, when obtaining the apparent reflectance after gas absorption correction, gas absorption correction is performed on the apparent reflectance of the target satellite-payload related bands using reanalysis data containing water vapor, ozone, and carbon dioxide to obtain the apparent reflectance after gas absorption correction. The bands of the apparent reflectance include the visible light channel, the near-infrared channel, and the short-wave near-infrared channel.

[0023] As a further aspect of the present invention, the target satellite-payload observation level 2 cloud detection product is read. For each pixel, if the cloud detection product is marked as "ice and snow", the following process is not performed, and the process is terminated.

[0024] As a further aspect of the present invention, the target satellite-payload observation level 2 cloud detection product is read. If the cloud detection product is marked as "definitely clear sky", for each pixel, the following inland water body judgment conditions are executed. If the conditions are not met, the following process is continued; otherwise, the process is terminated.

[0025] Determine whether it is an inland water body. If the following conditions are met, the process terminates; otherwise, proceed with the following process.

[0026]

[0027] Where, ρ nir1 and ρ red These represent the near-infrared channel 1 and the apparent reflectance of red light, respectively; t1 is the threshold for identifying inland water bodies; ρ swir t1 represents the apparent reflectance of the short-wave near-infrared channel, and t2 represents the threshold value for judging the reflectance of the short-wave near-infrared channel.

[0028] As a further aspect of the present invention, the effective pixels of the selected target satellite-payload are averaged to obtain the spatiotemporal matching {time, longitude, latitude, θ}. s θ v , Θ, Land and Sea Signage, AOD, ρ blue ρ green ρ red ρ nir1 ρ nir2 ρ swir NDVI, NDVI swir AFRI observation records, where θ s For the solar zenith angle, θ v For satellite zenith angle, Θ is the relative azimuth angle, ρ is the scattering angle, and ρ is the relative azimuth angle. blue ρ green ρ red These represent the apparent reflectance of the target satellite-payload channels in the blue, green, and red light channels, respectively, ρ. nir1 and ρ nir2 The apparent reflectance of the target satellite-payload near-infrared channel 1 and near-infrared channel 2 are ρ, respectively. swir The apparent reflectance of the shortwave near-infrared channel of the target satellite-payload is given by NDVI, which is calculated from the apparent reflectance of near-infrared channel 1 and the apparent reflectance of the red channel. swir The apparent reflectance of the near-infrared channel 1 and the short-wave near-infrared channel are calculated, while the apparent reflectance of the blue light channel and the short-wave near-infrared channel are calculated.

[0029] As a further aspect of the present invention, before establishing the surface reflectance model, based on the observation angles where the relationship between the target satellite and payload surface reflectance changes significantly, combined with... and and Two sets of relationship analysis were used to determine the inflection points of the fitting parameters and correlation changes, and to identify the breakpoints for the observation angle classification setting and the surface parameter classification setting. Specifically, the parameters were divided into primary and secondary variables based on the strength of the chosen correlation. Within each category of the primary variables, the data for each category were statistically analyzed according to the classification criteria of the secondary variables. and and The most strongly correlated fitting relationship is used to determine the surface reflectance model; To calculate the surface reflectance of the blue light channel, To calculate the surface reflectance of the red light channel, To calculate the surface reflectance of the shortwave near-infrared channel.

[0030] As a further aspect of the present invention, the surface reflectance model is used to substitute into the aerosol inversion algorithm to perform aerosol optical thickness inversion, wherein the aerosol inversion algorithm is an aerosol inversion algorithm based on multi-angle reflectance analysis.

[0031] As a further aspect of the present invention, the aerosol inversion algorithm is an aerosol inversion algorithm based on visible light and near-infrared reflectance.

[0032] According to a second aspect of the invention, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, which, when executed by the processor, performs any of the above-described aerosol remote sensing quantitative inversion methods based on a nonlinear surface reflectance model according to the invention.

[0033] According to a third aspect of the invention, a computer-readable storage medium is also provided, storing computer program instructions that, when executed, implement any of the above-described aerosol remote sensing quantitative inversion methods based on a nonlinear surface reflectance model according to the invention.

[0034] Compared with existing technologies, the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model provided by this invention has the following advantages:

[0035] This invention establishes a nonlinear surface reflectance model by selecting samples with a reflectance less than 0.2 as background values. This model considers the nonlinear variation of surface reflectance with observation geometry and surface parameters, thus improving the accuracy of AOD inversion. Simultaneously, this invention utilizes NDVI and NDVI... swir The method, along with surface parameters such as AFRI, was analyzed to better characterize the variation of surface reflectance. This invention is simple, easy to implement, and widely applicable, and can be used for quantitative inversion of aerosol remote sensing data from various satellite remote sensing sources.

[0036] These or other aspects of this application will become more apparent from the following description of embodiments. It should be understood that the foregoing general description and the following detailed description are merely illustrative and explanatory, and are not intended to limit the scope of this application. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. In the drawings:

[0038] Figure 1 This is a flowchart illustrating a quantitative inversion method for aerosol remote sensing based on a nonlinear surface reflectance model, as described in an embodiment of the present invention.

[0039] Figure 2 This invention relates to an embodiment of a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model, using FY3D / MERSI-II. and With θ s A schematic diagram of the fitting relationship between Θ and Θ;

[0040] Figure 3 This invention relates to an embodiment of a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model, using FY3D / MERSI-II. and With θ s A schematic diagram of the fitting relationship between Θ and Θ;

[0041] Figure 4 This invention relates to an embodiment of a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model, using FY3D / MERSI-II. and With NDVI, NDVI swir A schematic diagram of the AFRI fitting relationship;

[0042] Figure 5 This invention relates to an embodiment of a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model, using FY3D / MERSI-II. and With NDVI, NDVI swir A schematic diagram of the AFRI fitting relationship;

[0043] Figure 6 In an embodiment of the present invention, an aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model is used to retrieve FY3D / MERSI-II at different θ values. s Down and With NDVIswir A schematic diagram of the fitting relationship;

[0044] Figure 7 In an embodiment of the present invention, an aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model is used to retrieve FY3D / MERSI-II at different θ values. s Down and With NDVI swir A schematic diagram of the fitting relationship.

[0045] Figure 8 This is a schematic diagram of the hardware structure of an embodiment of the computer device for executing the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model in this invention.

[0046] Figure 9 A schematic diagram of a computer-readable storage medium for a quantitative inversion method of aerosol remote sensing based on a nonlinear surface reflectance model provided in an embodiment of the present invention.

[0047] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0048] The present application will now be further described in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0049] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0052] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0053] Because the newly improved surface reflectance model still exhibits a linear relationship with changes in observation geometry and vegetation index, abrupt changes in surface reflectance occur when there are significant variations in the surface type of neighboring pixels or the observation angle, leading to anomalies in the quantitatively retrieved AOD. To address this issue, this invention provides a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model.

[0054] In some implementations, the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model can be applied to computer equipment, which can be a PC, a laptop, a mobile terminal, or other device with display and processing capabilities, but is not limited to these.

[0055] This application provides a method for quantitative inversion of aerosol remote sensing based on a nonlinear surface reflectance model. This method includes the following steps:

[0056] Step S10: Read the observation time of the target satellite-payload orbital level one data and the observation time of the auxiliary satellite-payload AOD product. If the time difference between the two is less than 5 minutes, proceed with the following process; otherwise, terminate.

[0057] Step S20: Select samples with AOD values ​​less than 0.2 from the auxiliary satellite-payload AOD dataset as background values ​​for atmospheric correction of the target payload;

[0058] Step S30: Read the count value of each pixel of the target satellite-payload observation orbital level one data and convert it into apparent reflectance of the top of the atmosphere. After correction, the apparent reflectance after gas absorption correction is obtained.

[0059] Step S40: Read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is not marked as "ice and snow", then read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is marked as "definitely clear sky", proceed with the following process; otherwise, terminate.

[0060] Step S50: Read the reflectivity of multiple channels of the target satellite-payload. If it does not meet the identification conditions for inland water bodies, proceed with the following process; otherwise, terminate.

[0061] Step S60: Read the four datasets of longitude, latitude, land and sea identification and AOD in the auxiliary satellite-payload AOD product, filter out the effective pixels of the target satellite-payload orbital level data within the preset range, and average the effective pixels of the selected target satellite-payload for each pixel of the auxiliary satellite-payload to obtain the spatiotemporally matched observation records of the target satellite-payload and the auxiliary satellite-payload.

[0062] Step S70: Based on the observation records, cycle through the changes in solar and satellite observation angles and azimuth angles to obtain the surface reflectance of the target satellite-payload. By analyzing and determining the breakpoints for the observation angle classification settings and the breakpoints for the surface parameter classification settings, a surface reflectance model is established.

[0063] This invention provides a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model. By selecting samples with a value less than 0.2 in the auxiliary AOD dataset as background values ​​for atmospheric correction of the target payload, and cyclically varying the solar, satellite observation angle, and azimuth angle, the surface reflectance of different channels is obtained using an atmospheric radiative transfer model. By analyzing the variation patterns and significance of surface reflectance between channels under different seasons with key parameters such as observation geometry, surface type, and surface vegetation parameters, a mathematical model applicable to the surface reflectance relationship of each group is established to characterize the variation patterns of surface reflectance with season and angle as accurately as possible, thereby achieving high-precision AOD inversion.

[0064] The following steps are included in the quantitative inversion operation of aerosol remote sensing:

[0065] 1) Read the observation time of the target satellite-payload orbital-level data and the observation time of the auxiliary satellite-payload AOD product. If the difference between the former and the latter observation time is less than 5 minutes, proceed with the following process; otherwise, terminate.

[0066] 2) Read the Auxiliary Satellite-Payload AOD per orbit product. For each pixel, perform the following judgment condition. If the condition is met, continue the following process; otherwise, terminate.

[0067] AOD auxiliary <0.2

[0068] Among them, AOD auxiliary To supplement satellite-payload AOD data, 0.2 is the threshold for selecting samples.

[0069] 3) Read the target satellite-payload observation orbit-by-orbit first-level data. For each pixel, collect counts in multiple visible, near-infrared, and short-wave near-infrared bands. Using calibration coefficients and physical quantity conversion formulas, convert the counts into apparent reflectance at the top of the atmosphere. The visible light channel includes blue, green, and red bands, abbreviated as blue, green, and red, respectively. The near-infrared channel includes two bands, abbreviated as nir1 and nir2, respectively. The short-wave near-infrared channel is abbreviated as swir.

[0070] 4) Using auxiliary data such as water vapor, ozone, and carbon dioxide, the apparent reflectance of the target satellite-payload correlation bands is corrected for gas absorption to obtain the gas absorption-corrected apparent reflectance. The correlation bands include blue, green, red, nir1, nir2, and swir.

[0071] 5) Read the target satellite-payload observation level 2 cloud detection product. For each pixel, if the cloud detection product is marked as "ice and snow", do not perform the following process; otherwise, continue to perform the following process.

[0072] 6) Read the target satellite-payload observation level 2 cloud detection product. For each pixel, if the cloud detection product is marked as "definitely clear sky", proceed with the following process; otherwise, terminate.

[0073] 7) For each pixel, execute the following inland water body judgment conditions. If not met, continue the following process; otherwise, terminate. Determine whether it is an inland water body. If the following conditions are met, the process terminates; otherwise, proceed with the following process.

[0074]

[0075] Where, ρ nir1 and ρ red These represent the near-infrared channel 1 and the apparent reflectance of red light, respectively; t1 is the threshold for identifying inland water bodies; ρ swir t1 represents the apparent reflectance of the short-wave near-infrared channel, and t2 represents the threshold value for judging the reflectance of the short-wave near-infrared channel.

[0076] 8) Read the four datasets of longitude, latitude, land and sea identification and AOD in the auxiliary satellite-payload AOD product. For each pixel, filter out the valid pixels of the target satellite-payload orbital level data within a radius of 5km. If the number of pixels is greater than 2, execute the following judgment condition; otherwise, terminate.

[0077] 9) For each pixel of the auxiliary satellite-payload pair, average the effective pixels of the selected target satellite-payload pair to obtain the spatiotemporal matching parameters {time, longitude, latitude, θ}. s θ v , Θ, Land and Sea Signage, AOD, ρ blue ρ green ρ red ρ nir1 ρ nir2 ρ swir NDVI, NDVI swir AFRI observation records; where θ s For the solar zenith angle, θ v For satellite zenith angle, Θ is the relative azimuth angle, ρ is the scattering angle, and ρ is the relative azimuth angle. blue ρ green ρ red These represent the apparent reflectance of the target satellite-payload channels in the blue, green, and red light channels, respectively, ρ. nir1 and ρ nir2The apparent reflectance of the target satellite-payload near-infrared channel 1 and near-infrared channel 2 are ρ, respectively. swir The apparent reflectance of the shortwave near-infrared channel of the target satellite-payload is given by NDVI, which is calculated from the apparent reflectance of near-infrared channel 1 and the apparent reflectance of the red channel. swir The apparent reflectance of the near-infrared channel 1 and the apparent reflectance of the short-wave near-infrared channel are calculated, while the apparent reflectance of the blue light channel and the apparent reflectance of the short-wave near-infrared channel are calculated.

[0078] 10) Statistically calculate θ respectively s θ v Θ, NDVI, NDVI swir The distribution characteristics of the extreme values ​​and average values ​​of the six AFRI parameters were analyzed. One parameter was selected as the independent variable, and 19 breakpoints were set for each parameter, dividing the data into 20 classes. The distribution of each parameter in each class was statistically analyzed. and and The linear fit relationship (slope, intercept) and correlation are determined; the independent variables are changed sequentially until all 6 parameters have been traversed.

[0079] 11) For the observation geometry θ s θ v Θ, comparison and and The fitting relationship and correlation of the target satellite-payload surface reflectance relationship change under these three variables. Based on the correlation level, determine which observation angle the relationship of target satellite-payload surface reflectance changes significantly with.

[0080] 12) For NDVI, NDVI swir AFRI, comparison and and The fitting relationship and correlation of the target satellite-payload surface reflectance relationship are analyzed under these three variables. Based on the correlation level, it is determined which surface parameter significantly changes the relationship between the target satellite and payload surface reflectance.

[0081] 13) For observation angles where the relationship between the target satellite and payload surface reflectivity changes significantly, the combined... and and The inflection points of the changes in fitting parameters and correlations in the two sets of relationship analysis were used to determine the breakpoints in the observation angle classification settings;

[0082] 14) For surface parameters where the relationship between the target satellite and payload surface reflectance changes significantly, the combined... and and The inflection points of the changes in fitting parameters and correlations in the two sets of relationship analysis were used to determine the breakpoints in the classification settings of surface parameters.

[0083] 15) Compare the correlations in the two sets of relationships in 13) and 14), select the parameter with the strongest correlation as the main variable, and the parameters with weaker correlations as secondary variables; under each category of the main variable, according to the classification criteria of the secondary variables, statistically analyze the samples in each category. and and The most strongly correlated fitting relationship (which can be one of the power, logarithmic, polynomial, linear, etc.) is used to determine the entire surface reflectance model.

[0084] 16) Based on the above surface reflectance model, it is substituted into the aerosol inversion algorithm to realize the inversion of aerosol optical thickness.

[0085] In this embodiment, the surface reflectance model is used to perform aerosol optical thickness inversion in the aerosol inversion algorithm. The aerosol inversion algorithm is an aerosol inversion algorithm based on multi-angle reflectance analysis, and the aerosol inversion algorithm is an aerosol inversion algorithm based on visible light and near-infrared reflectance.

[0086] To facilitate understanding and implementation of the present invention by those skilled in the art, the following describes the embodiments of the present invention in detail by applying the method of the present invention to the MERSI-II sensor (Medium Resolution Spectral Imager) of the Fengyun-3 satellite and selecting the Aqua / MODIS sensor as an auxiliary data source.

[0087] 1) Read the observation time from the MERSI track-by-track primary data file name, and read the observation time of the Aqua / MODIS L2 AOD product track-by-track data. If the difference between the former and the latter observation time is less than 5 minutes, execute the following procedure; otherwise, terminate.

[0088] 2) Read the Aqua / MODIS L2 AOD track-by-track product. For each pixel, perform the following judgment condition. If the condition is met, continue the following process; otherwise, terminate.

[0089] AOD MODIS <0.2

[0090] Among them, AOD MODIS To supplement satellite-payload AOD data, 0.2 is the threshold for selecting samples.

[0091] 3) Read the count values ​​of each pixel in FY-3D / MERSI-II L1 data in three visible light channels (0.47μm, 0.55μm and 0.65μm), two near-infrared channels (0.865μm and 1.03μm), and one short-wave near-infrared channel (2.13μm). Convert the above channel values ​​into apparent reflectance of the top of the atmosphere using calibration coefficients and physical quantity conversion formulas.

[0092] 4) Using auxiliary data such as water vapor, ozone, and carbon dioxide, corrections were made for water vapor absorption, ozone absorption, and carbon dioxide absorption at 0.47μm, 0.55μm and 0.65μm, 0.865μm, 1.03μm and 2.13μm, respectively, to obtain the apparent reflectance after gas absorption correction.

[0093] 5) Read the MERSI-II Level 2 cloud inspection product. For each pixel, if the cloud inspection product is marked as "ice and snow", do not perform the following process; otherwise, continue to perform the following process.

[0094] 6) Read the MERSI-II Level 2 cloud detection product. For each pixel, if the cloud detection product is marked as "definitely clear sky", proceed with the following process; otherwise, terminate.

[0095] 7) For each pixel, execute the following inland water body judgment conditions. If not met, continue the following process; otherwise, terminate. Determine whether it is an inland water body. If the following conditions are met, the process terminates; otherwise, proceed with the following process.

[0096]

[0097] Where, ρ nir1 and ρ red The apparent reflectances are 0.865 μm and 0.65 μm, respectively, with t1 being 0.1 and t2 being 0.08.

[0098] 8) Read the four datasets of longitude, latitude, land and sea identification and AOD from the Aqua / AOD product. For each pixel, filter out the valid MERSI-II L1 pixels within a radius of 5km. If the number of pixels is greater than 2, execute the following judgment condition; otherwise, terminate.

[0099] 9) For each Aqua / AOD cell, average the selected MERSI-II valid cells to obtain the spatiotemporal matching of the two: {time, longitude, latitude, θ}. s θ v , Θ, Land and Sea Signage, AOD, ρ blue ρ green ρ red ρ nir1 ρ nir2ρ swir NDVI, NDVI swir AFRI observation records; where θ s For the solar zenith angle, θ v For satellite zenith angle, Θ is the relative azimuth angle, ρ is the scattering angle, and ρ is the relative azimuth angle. blue ρ green ρ red These represent the apparent reflectance of the MERSI-II blue, green, and red light channels, respectively, ρ. nir1 and ρ nir2 The apparent reflectance of MERSI-II near-infrared channel 1 and near-infrared channel 2 are ρ, respectively. swir NDVI is the apparent reflectance of the MERSI-II short-wave near-infrared channel, calculated from the apparent reflectance of near-infrared channel 1 and the apparent reflectance of the red channel. swir The apparent reflectance of the near-infrared channel 1 and the apparent reflectance of the short-wave near-infrared channel are calculated, while the apparent reflectance of the blue light channel and the apparent reflectance of the short-wave near-infrared channel are calculated.

[0100] 10) Statistically calculate θ respectively s θ v Θ, NDVI, NDVI swir The distribution characteristics of the extreme values ​​and average values ​​of the six AFRI parameters were analyzed. One parameter was selected as the independent variable, and 19 breakpoints were set for each parameter, dividing the data into 20 classes. The distribution of each parameter in each class was statistically analyzed. and and The linear fit relationship (slope, intercept) and correlation are determined; the independent variables are changed sequentially until all 6 parameters have been traversed.

[0101] 11) For the observation geometry θ s θ v Θ, comparison and and The fitting relationship and correlation of the MERSI-II surface reflectance relationship under these three variables were analyzed. Based on the correlation strength, it was determined which observation angle significantly affected the MERSI-II surface reflectance relationship. (The last part, θ, appears to be a separate, unrelated sentence fragment.) s For example, Θ Figure 2 and Figure 3 They were shown respectively and and The changing trend of the two sets of fitted relationships, from the correlation R 2 Let's look at the two sets of relationships as θ s Significant changes.

[0102] 12) For NDVI, NDVI swir AFRI, comparison and and The fitting relationship and correlation of the MERSI-II surface reflectance relationship under these three variables were analyzed. Based on the correlation level, it was determined which surface parameter significantly changed the MERSI-II surface reflectance relationship. Figure 4 They were shown respectively and With NDVI, NDVI swir The changing trend of AFRI fitting parameters Figure 5 They were shown respectively and With NDVI, NDVI swir The changing trend of AFRI fitting parameters; combined with Figure 4 and Figure 5 The performance of NDVI and NDVI swir The distribution of AFRI values ​​in each sample class, and the relationship between MERSI-II surface reflectance and NDVI. swir Parameter changes.

[0103] 13) For observation angles where the MERSI-II surface reflectance relationship changes significantly, combined with and and The inflection points of the changes in fitting parameters and correlations in the two sets of relationship analysis were used to determine the breakpoints in the observation angle classification settings; combined with Figure 2 and Figure 3 θ can be determined s Breakpoints were set at 30° and 55°. Figure 2 For July 2022 FY3D / MERSI-II and With θ s A schematic diagram of the fitting relationship between (top) and Θ (bottom). Figure 3 For July 2022 FY3D / MERSI-II and With θ s A schematic diagram of the fitting relationship between (top) and Θ (bottom);

[0104] 14) For surface parameters where the MERSI-II surface reflectance relationship changes significantly, the combined... and and The inflection points of the changes in fitting parameters and correlations in the two sets of relationship analysis were used to determine the breakpoints in the land surface parameter classification settings; combined with... Figure 4 NDVI can be determined swir The breakpoints are -0.05, 0.25, and 0.75, respectively. Figure 4 For July 2022 FY3D / MERSI-II and With NDVI (pictured above), NDVI swir (Middle image) and AFRI (bottom image) fitting relationship schematic diagram; combined with Figure 5 NDVI can be determined swir The breakpoints are -0.05 and 0.75 respectively. Figure 5 For July 2022 FY3D / MERSI-II and With NDVI (pictured above), NDVI swir (Middle figure) and AFRI (bottom figure) fitting relationship diagram;

[0105] 15) Compare the correlations in the two sets of relationships in 13) and 14), select the parameter with the strongest correlation as the main variable, and the parameters with weaker correlations as secondary variables; under each category of the main variable, according to the classification criteria of the secondary variables, statistically analyze the samples in each category. and and The most strongly correlated fitting relationship (which can be one of the power, logarithmic, polynomial, linear, etc.) is used to determine the entire surface reflectance model. Figure 6 θ were shown respectively s Three groups of samples: less than 30°, 30–55°, and greater than 55°. and Fitting relationship (first column: slope, second column: intercept, third column: correlation) with NDVI swir Changes Figure 7 and Figure 6 Similar, but targeting and Figure 6 for Figure 6 FY3D / MERSI-II at different θ s Down and With NDVI swir A schematic diagram of the fitting relationship. Figure 6 The first line is θ s Less than 30°, second row θ s The angle is 30–55°, and the third row is θ. s Greater than 55°. Figure 7 For FY3D / MERSI-II at different θ s Down and With NDVI swir A schematic diagram of the fitting relationship. Figure 7 The first line is θ s Less than 30°, second row θs The angle is 30–55°, and the third row is θ. s Greater than 55°.

[0106] Based on the above surface reflectance model, it is substituted into the aerosol inversion algorithm to realize the inversion of aerosol optical thickness.

[0107] This invention establishes a nonlinear surface reflectance model by selecting samples with a reflectance less than 0.2 as background values. This model considers the nonlinear variation of surface reflectance with observation geometry and surface parameters, thus improving the accuracy of AOD inversion. Simultaneously, this invention utilizes NDVI and NDVI... swir The method, along with surface parameters such as AFRI, was analyzed to better characterize the variation of surface reflectance. This invention is simple, easy to implement, and widely applicable, and can be used for quantitative inversion of aerosol remote sensing data from various satellite remote sensing sources.

[0108] A third aspect of the present invention also provides a computer device 1000, including a memory 1001 and a processor 1002, wherein the memory stores a computer program, which, when executed by the processor, implements the method of any of the above embodiments.

[0109] like Figure 8 The diagram shown is a hardware structure schematic of an embodiment of a computer device provided by the present invention for executing a quantitative inversion method for aerosol remote sensing based on a nonlinear surface reflectance model. Figure 8 Taking the computer device 1000 shown as an example, this computer device includes a processor 1002 and a memory 1001, and may also include an input device and an output device. The processor 1002, memory 1001, input device, and output device can be connected via a bus or other means. Figure 8 Taking a bus connection as an example, the input device can receive input digital or character information, as well as generate signal inputs related to the quantitative inversion of aerosol remote sensing based on a nonlinear surface reflectance model. The output device may include display devices such as a screen.

[0110] Memory 1001, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model in this embodiment. Memory 1001 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created by using the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model, etc. In addition, memory 1001 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 1001 may optionally include memory remotely located relative to processor 1002, and these remote memories can be connected to the local module via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0111] In some embodiments, processor 1002 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. Processor 1002 is typically used to control the overall operation of computer device 1000. In this embodiment, processor 1002 is used to run program code stored in memory 1001 or process data. In this embodiment, the processors 1002 of multiple computer devices 1000 execute various server functions and data processing by running non-volatile software programs, instructions, and modules stored in memory 1001, thereby implementing the steps of the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model described in the above-described method embodiment.

[0112] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software and a general-purpose hardware platform, or of course, using hardware. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware.

[0113] A fourth aspect of the present invention also provides a computer-readable storage medium. Figure 9 A schematic diagram of a computer-readable storage medium for a quantitative aerosol remote sensing inversion method based on a nonlinear surface reflectance model, provided by an embodiment of the present invention, is shown. Figure 9As shown, the computer-readable storage medium 2000 stores computer program instructions 2001, which can be executed by a processor. When the computer program instructions 2001 are executed, they implement the method of any of the above embodiments, that is, implement the steps of the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model of the above method embodiments.

[0114] It should be understood that, where there is no conflict, all the embodiments, features and advantages described above for the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to the present invention are equally applicable to the aerosol remote sensing quantitative inversion system and storage medium based on a nonlinear surface reflectance model according to the present invention.

[0115] Those skilled in the art will also understand that the various logic blocks, modules, circuits, and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the functionality of various illustrative components, blocks, modules, circuits, and steps has been generally described. Whether this functionality is implemented as software or as hardware depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the embodiments disclosed herein.

[0116] Finally, it should be noted that the computer-readable storage medium (e.g., memory) described herein can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. By way of example, and not limitation, non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which can act as external cache memory. By way of example, and not limitation, RAM can be obtained in various forms, such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The storage devices disclosed herein are intended to include, but are not limited to, these and other suitable types of memory.

[0117] The various logic blocks, modules, and circuits described herein can be implemented or executed using the following components designed to perform the functions herein: general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination of these components. A general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP, and / or any other such configuration.

[0118] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A quantitative inversion method for aerosol remote sensing based on a nonlinear surface reflectance model, characterized in that, The quantitative inversion method for aerosol remote sensing includes the following steps: Read the observation time of the target satellite-payload orbit-by-orbit first-level data and the observation time of the auxiliary satellite-payload AOD product. If the time difference between the two is less than 5 minutes, proceed with the following process; otherwise, terminate. Samples with AOD values ​​less than 0.2 were selected from the auxiliary satellite-payload AOD dataset and used as background values ​​for atmospheric correction of the target payload. The count value of each pixel in the target satellite-payload observation orbital level data is read and converted into the apparent reflectance of the top of the atmosphere. After correction, the apparent reflectance after gas absorption correction is obtained. Read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is not marked as "ice and snow", then read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is marked as "definitely clear sky", proceed with the following process; otherwise, terminate. Read the reflectivity of multiple channels of the target satellite-payload. If it does not meet the identification criteria for inland water bodies, proceed with the following process; otherwise, terminate the process. Read the four datasets of longitude, latitude, land and sea identification and AOD in the auxiliary satellite-payload AOD product, filter out the effective pixels of the target satellite-payload orbit-level data within the preset range, and average the effective pixels of the selected target satellite-payload for each pixel of the auxiliary satellite-payload to obtain the spatiotemporally matched observation records of the target satellite-payload and the auxiliary satellite-payload. Based on observation records, the surface reflectance of the target satellite-payload is obtained by cyclically varying the solar and satellite observation angles and azimuth angles. By analyzing these changes, the breakpoints for the observation angle classification settings and the breakpoints for the surface parameter classification settings are determined, and a surface reflectance model is established.

2. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 1, characterized in that, Samples with an AOD value less than 0.2 were selected from the auxiliary satellite-payload AOD dataset, including: Read the AOD per orbit product of the auxiliary satellite-payload. For each pixel, perform the following judgment condition. If the condition is met, continue; otherwise, terminate. AOD auxiliary <0.2 Among them, AOD auxiliary To supplement satellite-payload AOD data, 0.2 is the threshold for selecting samples.

3. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 2, characterized in that, Read the count value of each pixel from the target satellite-payload observation orbit-by-orbit first-level data and convert it into apparent reflectance at the top of the atmosphere, including: Read the count values ​​of each pixel in multiple visible light channels, near-infrared channels, and short-wave near-infrared channels from the target satellite-payload observation orbital level data, and convert the count values ​​into the apparent reflectance of the top of the atmosphere through calibration coefficients and physical quantity conversion formulas; The visible light channel includes blue light, green light and red light bands, and the near-infrared channel includes two near-infrared bands.

4. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 3, characterized in that, When obtaining the apparent reflectance after gas absorption correction, the apparent reflectance of the target satellite-payload related bands is corrected for gas absorption using reanalysis data containing water vapor, ozone, and carbon dioxide to obtain the apparent reflectance after gas absorption correction. The bands of the apparent reflectance include the visible light channel, the near-infrared channel, and the short-wave near-infrared channel.

5. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 1, characterized in that, Read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is marked as "ice and snow" for each pixel, the process terminates.

6. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 5, characterized in that, Read the target satellite-payload observation level 2 cloud detection product. If the cloud detection product is marked as "definitely clear sky", for each pixel, execute the following inland water body judgment conditions. If not met, continue to execute the following process; otherwise, terminate. Determine whether it is an inland water body. If the following conditions are met, the process terminates; otherwise, proceed with the following process: Where, ρ nir1 and ρ red These represent the near-infrared channel 1 and the apparent reflectance of red light, respectively; t1 is the threshold for identifying inland water bodies; ρ swir t1 represents the apparent reflectance of the short-wave near-infrared channel, and t2 represents the threshold value for judging the reflectance of the short-wave near-infrared channel.

7. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 6, characterized in that, The effective pixels of the selected target satellite-payload are averaged to obtain the spatiotemporal matching {time, longitude, latitude, θ}. s θ v , Θ, Land and Sea Signage, AOD, ρ blue ρ green ρ red ρ nir1 ρ nir2 ρ swir NDVI, NDVI swir AFRI observation records; Where, θ s For the solar zenith angle, θ v For satellite zenith angle, Θ is the relative azimuth angle, ρ is the scattering angle, and ρ is the relative azimuth angle blue ρ green ρ red These represent the apparent reflectance of the target satellite-payload channels in the blue, green, and red light channels, respectively, ρ. nir1 and ρ nir2 The apparent reflectance of the target satellite-payload near-infrared channel 1 and near-infrared channel 2 are ρ, respectively. swir The apparent reflectance of the shortwave near-infrared channel of the target satellite-payload is given by NDVI, which is calculated from the apparent reflectance of near-infrared channel 1 and the apparent reflectance of the red channel. swir The apparent reflectance of the near-infrared channel 1 and the short-wave near-infrared channel are calculated, while the apparent reflectance of the blue light channel and the short-wave near-infrared channel are calculated.

8. The aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model according to claim 7, characterized in that, Also includes: Statistical analysis of θ s θ v Θ, NDVI, NDVI swir The distribution characteristics of the extreme values ​​and average values ​​of the AFRI parameters were analyzed. One parameter was selected as the independent variable, and 19 breakpoints were set for each parameter, dividing the data into 20 classes. The distribution of each parameter in each class was statistically analyzed. and and The linear fitting relationship and correlation are obtained by changing the independent variables in turn until all 6 parameters are traversed; the linear fitting relationship includes the slope and intercept. To calculate the surface reflectance of the blue light channel, To calculate the surface reflectance of the red light channel, To calculate the surface reflectance of the shortwave near-infrared channel.

9. A computer device, characterized in that, The computer device includes multiple computer devices, each computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processors of the multiple computer devices execute the computer program, they jointly implement the steps of the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program stored in the storage medium is executed by a processor, it implements the steps of the aerosol remote sensing quantitative inversion method based on a nonlinear surface reflectance model as described in any one of claims 1 to 8.

Images