Correction method and system for anisotropic reflection of GF6-WFV data

Abstract

The invention provides a correction method and system for anisotropic reflection of GF6-WFV data, and belongs to the technical field of spatial information.The correction method comprises the steps that the GF6-WFV data is subjected to block processing, data of a block center is compared with data extracted from satellite navigation file data, atmospheric correction is carried out on the block data one by one, and the correction efficiency is improved; the processing efficiency and the correction precision are remarkably improved, image data are divided into small blocks with 2000 * 2000 pixels through blocking processing, the pixel amount calculated each time is reduced, the method is more suitable for rapid processing of large-data-amount and high-resolution images, the blocking coverage range is large, local and overall consistency can be taken into account, and secondly, the processing efficiency and the correction precision are greatly improved. And the zenith angle and observation angle parameters of each block are standardized by the angle normalization model, so that the angle difference between different areas is reduced, and the accuracy of a correction result is improved. And finally, in combination with a lookup table technology, computing resource consumption can be remarkably reduced, and the method is convenient to operate in practical application of radiation correction and has the advantages of high efficiency and convenience.

Description

Technical Field

[0001] The present invention relates to the field of spatial information technology, and in particular to a correction method and system for anisotropic reflection of GF6-WFV data. Background Art

[0002] GF6 (Gaofen-6 Satellite) was officially put into use on March 21, 2019. As an important component of my country's low-orbit high-resolution optical remote sensing satellite, it has the characteristics of high resolution, wide coverage, high quality and efficient imaging. The satellite is equipped with WFV (multi-spectral medium-resolution wide-band camera) with an observation width of more than 800 kilometers, a spatial resolution of 16 meters, and covers 8 spectral bands. GF6-WFV (multi-spectral medium-resolution wide-band camera of Gaofen-6 Satellite) has advanced imaging performance, and its remote sensing data is easily affected by atmospheric factors during transmission, including the absorption and scattering of atmospheric molecules and aerosols, which can cause signal distortion of remote sensing images. Therefore, in remote sensing quantitative analysis, in order to obtain real surface information, it is necessary to perform atmospheric correction on the image.

[0003] Surface reflectivity refers to the amount of solar radiation reflected by a surface in a specific direction and band, and is a key factor in studying global environmental changes. In the natural environment, the surface's reflection of solar radiation has anisotropic characteristics, that is, the reflection is different at different angles.

[0004] Among the atmospheric correction methods, the radiation transfer model is widely used due to its high accuracy. Common radiation transfer models include the 6S radiation transfer model, the MODTRAN model, the LOWTRAN model, and the FLAASH model, which are widely used in the field of remote sensing. However, the atmospheric correction method faces shortcomings such as low computational efficiency, poor spatial adaptability, and insufficient consideration of aerosol changes. For example, directly using the radiation transfer model to perform atmospheric correction on the entire remote sensing image is computationally intensive and time-consuming, and it is difficult to meet the needs of large-scale processing of high-resolution data. In addition, some methods rely on the spectral characteristics of specific objects (such as cement pavement and vegetation areas), which is less applicable in areas with large differences in the distribution of objects and cannot accurately restore the surface reflectance. Finally, some models do not fully consider the changes in the spatial distribution of aerosols, so the correction accuracy is low in areas with complex aerosols; and most studies assume that the aerosol optical thickness is consistent under clear weather conditions, and may also ignore the errors caused by the uneven distribution of aerosols in actual images. Summary of the invention

[0005] In order to solve the problems in the above-mentioned prior art, the present invention provides a correction method and system for anisotropic reflection of GF6-WFV data. The invention processes GF6-WFV data in blocks, compares the data at the center of the block with the data extracted from the satellite navigation file data, selects a correction coefficient of the nearest point as the correction coefficient of the block, and performs atmospheric correction on the block data one by one, which significantly improves the processing efficiency and correction accuracy. The block processing divides the image data into small blocks of 2000*2000 pixels, reducing the number of pixels calculated each time, making it more suitable for fast processing of large amounts of data and high-resolution images, and the block coverage is large, which is conducive to taking into account the consistency of local and overall. Secondly, the angle normalization model standardizes the zenith angle and observation angle parameters of each block, reducing the angle difference between different regions, so that the accuracy of the correction result can be improved. Finally, combined with the lookup table technology, it can significantly reduce the consumption of computing resources, is suitable for batch processing of large-scale remote sensing images, is easy to operate in the actual application of radiation correction, and has the advantages of high efficiency and convenience. To achieve the above purpose, the technical solution is as follows:

[0006] In one aspect, the present invention provides a method for correcting anisotropic reflection of GF6-WFV data, the method comprising:

[0007] S1. Extract multi-angle remote sensing data based on historical data of MODIS BRDF (MODIS Bidirectional Reflectance Distribution Function) to obtain a multi-angle remote sensing data set;

[0008] S2, obtaining a lookup table according to the multi-angle remote sensing data set through a kernel driving model and a normalization model;

[0009] S3, collecting data through the GF6-WFV to obtain original image data;

[0010] S4, processing the original image data into blocks to obtain a plurality of block data;

[0011] S5. Extract data according to the block data to obtain the block 6S radiation transfer model input file;

[0012] S6, according to the 6S radiation transfer model input file of the block and the block data, obtain the correction coefficient of the block through the 6S radiation transfer model;

[0013] S7, performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data;

[0014] S8, querying the lookup table according to the plurality of corrected block data to obtain a plurality of normalized block data;

[0015] S9, splicing the plurality of normalized block data to obtain a corrected reflectivity image.

[0016] Optionally, in S2, according to the multi-angle remote sensing data set, a lookup table is obtained through a kernel-driven model and a normalization model, including:

[0017] S21. According to the multi-angle remote sensing data set, a kernel-driven model is used to extract the surface reflectance to obtain the surface reflectance in the data set. The kernel-driven model includes linear weighting of an isotropic scattering component, a volume scattering component, and a geometric optical scattering component, as shown in formula (1);

[0018] R(θ s ,θ v ,δ,Λ)=f iso (Λ)+f vol (Λ)K vol (θ s ,θ v ,δ)+f geo (Λ)K geo (θ s ,θ v ,δ) (1)

[0019] In the formula, R(θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, K vol (θ s ,θ v ,δ) is the volume scattering component kernel, K geo (θ s ,θ v ,δ) is the geometric optics scattering component kernel, f iso (Λ) is the weight of the isotropic scattering component in band Λ, f vol (Λ) is the weight of the volume scattering component in band Λ, f geo (Λ) is the weight of the geometrical optics scattering component in band Λ;

[0020] S22, extracting effective information according to the surface reflectivity in the data set, and obtaining a BRDF (Bidirectional Reflectance Distribution Function, a mathematical model that describes the light reflected from the surface of an object) prototype library;

[0021] S23, according to the BRDF prototype library, a lookup table is obtained by normalizing the model, and the method of the normalized model is shown in formula (2);

[0022]

[0023] In the formula, R n (θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function at the relative azimuth angle δ in the band Λ, R ′ (θ s ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, R ′ n (θ s ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function in band Λ at relative azimuth δ.

[0024] Optionally, the solar zenith angle θ corresponding to the lookup table s The range is 0°~75°, with a resolution of 1°; the observation zenith angle θ v The range is 0°~75°, with a resolution of 2°; the relative azimuth angle δ ranges from 0°~360°.

[0025] Optionally, the lookup table is a two-dimensional matrix symmetrical about the main plane, the matrix has 91 rows and 91 columns, and the numbers of the two-dimensional matrix represent the index of the optimal prototype as formula (3);

[0026]

[0027] Wherein, L is the row of the two-dimensional matrix, and S is the column of the two-dimensional matrix.

[0028] Optionally, in S4, the original image data is processed into blocks to obtain a plurality of block data, including: the original image data is processed into 2000*2000 blocks to obtain 2000*2000 block data.

[0029] Optionally, in S5, data extraction is performed according to the block data to obtain a block 6S radiation transfer model input file, including:

[0030] S51, reading the corresponding satellite navigation file data and extracting the file related to the sun angle according to the block data;

[0031] S52, analyzing and obtaining a solar zenith angle array and a solar azimuth angle array according to the file related to the solar angle;

[0032] S53, according to the solar zenith angle array and the solar azimuth angle array, selecting the minimum, middle and maximum three solar zenith angles in the solar zenith angle array and the corresponding solar azimuth angles in the solar azimuth angle array to obtain an input file of satellite navigation file data;

[0033] S54. According to the input file of the satellite navigation file data, the file data corresponding to the input file of the satellite navigation file data is input to the satellite navigation file data, and the file data is extracted to obtain the block-based 6S radiation transmission model input file.

[0034] Optionally, the block's 6S radiation transfer model input file includes: solar zenith angle, solar azimuth angle, satellite zenith angle, satellite azimuth angle, month and date of observation, aerosol optical thickness, and aerosol scattering and absorption characteristics.

[0035] Optionally, in S6, according to the 6S radiation transfer model input file of the block and the block data, a correction coefficient of the block is obtained through the 6S radiation transfer model, including:

[0036] S61, selecting a center point according to the block data to obtain data of the block center;

[0037] S62, comparing the data of the block center with the 6S radiation transfer model input file of the block, selecting the point in the 6S radiation transfer model input file that is closest to the block center, and obtaining the 6S radiation transfer model input file of the center point;

[0038] S63. According to the 6S radiation transfer model input file of the center point, the correction coefficients of the blocks are obtained through the 6S radiation transfer model.

[0039] Optionally, the calculation method of the 6S radiation transfer model is formula (4);

[0040]

[0041] In the formula, ρ(θ s ,θ v ,φ s -φ v ) is the reflectivity of the top of the atmosphere received by the sensor, ρ a is the atmospheric path radiation reflectivity composed of molecular scattering and aerosol scattering, ρ t is the surface diffuse reflectance, S is the large balloon surface reflectance, φs is the solar azimuth, φ v is the observation azimuth, T g (θ s ,θ v ) is the transmittance caused by atmospheric absorption, T(θ s ) is the linear transmittance of the sun-target path, T(θ v ) is the linear transmittance of the target-sensor path.

[0042] On the other hand, the present invention provides a correction system for anisotropic reflection of GF6-WFV data, the system is applied to a correction method for anisotropic reflection of GF6-WFV data, the system comprising:

[0043] The first extraction module is used to extract multi-angle remote sensing data according to the historical data of MODIS BRDF to obtain a multi-angle remote sensing data set;

[0044] A lookup table generation module is used to obtain a lookup table according to the multi-angle remote sensing data set through a kernel driving model and a normalization model;

[0045] The data acquisition module is used to acquire data through GF6-WFV to obtain original image data;

[0046] A block processing module is used to process the original image data in blocks to obtain a plurality of block data;

[0047] The second extraction module is used to extract data according to the block data to obtain a block 6S radiation transfer model input file;

[0048] A correction coefficient acquisition module is used to obtain the correction coefficient of the block through the 6S radiation transfer model according to the 6S radiation transfer model input file of the block and the block data;

[0049] A first data processing module, used for performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data;

[0050] A second data processing module is used to query the lookup table according to the plurality of corrected block data to obtain a plurality of normalized block data;

[0051] The third data processing module is used to splice the multiple normalized block data to obtain a corrected reflectivity image.

[0052] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0053] On the one hand, the above scheme processes the GF6-WFV data in blocks, compares the data at the center of the block with the data extracted from the satellite navigation file data, selects a correction coefficient of the nearest point as the correction coefficient of the block, and performs atmospheric correction on the block data one by one, which significantly improves the processing efficiency and correction accuracy. The block processing divides the image data into small blocks of 2000*2000 pixels, reducing the number of pixels calculated each time, making it more suitable for the rapid processing of large data volumes and high-resolution images, and the block coverage is large, which is conducive to taking into account the consistency of local and overall. On the second hand, the angle normalization model standardizes the zenith angle and observation angle parameters of each block, reducing the angle difference between different regions, so that the accuracy of the correction results can be improved. On the third hand, combined with the lookup table technology, it can significantly reduce the consumption of computing resources, which is suitable for batch processing of large-scale remote sensing images. It is easy to operate in the actual application of radiation correction and has the advantages of high efficiency and convenience. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 is a flow chart of an embodiment of a correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0056] Figure 2 It is a flow chart of obtaining a lookup table in an embodiment of a correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0057] Figure 3 It is a flow chart of obtaining a block 6S radiation transfer model input file in an embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0058] Figure 4 It is a flow chart of obtaining correction coefficients for blocks in an embodiment of a correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0059] Figure 5 It is a zenith angle distribution diagram observed by a lookup table in an embodiment of a correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0060] Figure 6 It is a relative azimuth distribution diagram of the lookup table in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0061] Figure 7 It is a lookup table for the near infrared band and the solar zenith angle of 45° in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention;

[0062] Figure 8 It is a system block diagram of an embodiment of the correction system for anisotropic reflection of GF6-WFV data of the present invention. DETAILED DESCRIPTION

[0063] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0064] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0065] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0066] like Figure 1 The flowchart of the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown in FIG. The present invention provides a correction method for anisotropic reflection of GF6-WFV data, which is implemented by a correction system for anisotropic reflection of GF6-WFV data, and the method comprises:

[0067] S1. Extract multi-angle remote sensing data based on historical data of MODIS BRDF to obtain a multi-angle remote sensing data set;

[0068] S2, obtaining a lookup table according to the multi-angle remote sensing data set through a kernel driving model and a normalization model;

[0069] Specifically, Figure 2 The flowchart of obtaining the lookup table in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown, and in S2, the lookup table is obtained according to the multi-angle remote sensing data set through the kernel driving model and the normalization model, including:

[0070] S21. According to the multi-angle remote sensing data set, a kernel-driven model is used to extract the surface reflectance to obtain the surface reflectance in the data set. The kernel-driven model includes linear weighting of an isotropic scattering component, a volume scattering component, and a geometric optical scattering component, as shown in formula (1);

[0071] R(θs ,θ v ,δ,Λ)=f iso (Λ)+f vol (Λ)K vol (θ s ,θ v ,δ)+f geo (Λ)K geo (θ s ,θ v ,δ) (1)

[0072] In the formula, R(θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, K vol (θ s ,θ v ,δ) is the volume scattering component kernel, K geo (θ s ,θ v ,δ) is the geometric optics scattering component kernel, f iso (Λ) is the weight of the isotropic scattering component in band Λ, f vol (Λ) is the weight of the volume scattering component in band Λ, f geo (Λ) is the weight of the geometrical optics scattering component in band Λ;

[0073] S22, extracting effective information according to the surface reflectance in the data set to obtain a BRDF prototype library;

[0074] S23, according to the BRDF prototype library, a lookup table is obtained by normalizing the model, and the method of the normalized model is shown in formula (2);

[0075]

[0076] In the formula, R n (θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function at the relative azimuth δ in band Λ, R ′ (θ s ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, R ′ n (θs ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function in band Λ at relative azimuth δ.

[0077] Specifically, Figure 5 The zenith angle distribution diagram of the observation table in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown as follows Figure 6 The relative azimuth distribution diagram of the lookup table in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown, and the solar zenith angle θ corresponding to the lookup table s The range is 0°~75°, with a resolution of 1°; the observation zenith angle θ v The range is 0°~75°, with a resolution of 2°; the relative azimuth angle δ ranges from 0°~360°.

[0078] Specifically, Figure 7 The lookup table in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention shown in the figure is a two-dimensional matrix symmetrical about the main plane and has 91 rows and 91 columns. The numbers of the two-dimensional matrix represent the index of the optimal prototype as formula (3);

[0079]

[0080] Wherein, L is the row of the two-dimensional matrix, and S is the column of the two-dimensional matrix.

[0081] Furthermore, the lookup table is stored in both raster and text formats for ease of use. According to the provided band and solar zenith angle, the target solar angle to be normalized is selected to obtain the corresponding lookup table. According to the lookup table, the corresponding BRDF prototype is found, and the corresponding parameters are obtained to calculate the reflectivity, and the reflectivity angle is normalized to eliminate the influence of surface anisotropy.

[0082] S3, collecting data through the GF6-WFV to obtain original image data;

[0083] S4, processing the original image data into blocks to obtain a plurality of block data;

[0084] Specifically, in S4, the original image data is processed into blocks to obtain a plurality of block data, including: the original image data is processed into 2000*2000 blocks to obtain 2000*2000 block data.

[0085] S5. Extract data according to the block data to obtain the block 6S radiation transfer model input file;

[0086] Specifically, Figure 3 The flowchart of obtaining the block 6S radiation transfer model input file in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown. In S5, data extraction is performed according to the block data to obtain the block 6S radiation transfer model input file, including:

[0087] S51, reading the corresponding satellite navigation file data and extracting the file related to the sun angle according to the block data;

[0088] Furthermore, by writing IDL (Interactive Data Language, a programming language) code, reading satellite navigation file data, extracting XML (Extensible Markup Language, Extensible Markup Language) files related to the sun angle, parsing the sun zenith angle file and the sun azimuth angle file.

[0089] S52, analyzing and obtaining a solar zenith angle array and a solar azimuth angle array according to the file related to the solar angle;

[0090] S53, according to the solar zenith angle array and the solar azimuth angle array, selecting the minimum, middle and maximum three solar zenith angles in the solar zenith angle array and the corresponding solar azimuth angles in the solar azimuth angle array to obtain an input file of satellite navigation file data;

[0091] S54. According to the input file of the satellite navigation file data, the file data corresponding to the input file of the satellite navigation file data is input to the satellite navigation file data, and the file data is extracted to obtain the block-based 6S radiation transmission model input file.

[0092] Specifically, the 6S radiation transfer model input file of this block includes: solar zenith angle, solar azimuth angle, satellite zenith angle, satellite azimuth angle, observation month and date, aerosol optical thickness, and aerosol scattering and absorption characteristics.

[0093] S6, according to the 6S radiation transfer model input file of the block and the block data, obtain the correction coefficient of the block through the 6S radiation transfer model;

[0094] Specifically, Figure 4 The flowchart of obtaining the correction coefficient of the block in the embodiment of the correction method for anisotropic reflection of GF6-WFV data of the present invention is shown, and S6 obtains the correction coefficient of the block through the 6S radiation transfer model according to the 6S radiation transfer model input file of the block and the block data, including:

[0095] S61, selecting a center point according to the block data to obtain data of the block center;

[0096] S62, comparing the data of the block center with the 6S radiation transfer model input file of the block, selecting the point in the 6S radiation transfer model input file that is closest to the block center, and obtaining the 6S radiation transfer model input file of the center point;

[0097] S63. According to the 6S radiation transfer model input file of the center point, the correction coefficients of the blocks are obtained through the 6S radiation transfer model.

[0098] Specifically, the calculation method of the 6S radiation transfer model is formula (4);

[0099]

[0100] In the formula, ρ(θ s ,θ v ,θ s -φ v ) is the reflectivity of the top of the atmosphere received by the sensor, ρ a is the atmospheric path radiation reflectivity composed of molecular scattering and aerosol scattering, ρ t is the surface diffuse reflectance, S is the large balloon surface reflectance, φ s is the solar azimuth, φ v is the observation azimuth, T g (θ s ,θ v ) is the transmittance caused by atmospheric absorption, T(θ s ) is the linear transmittance of the sun-target path, T(θ v ) is the linear transmittance of the target-sensor path.

[0101] S7, performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data;

[0102] S8, querying the lookup table according to the plurality of corrected block data to obtain a plurality of normalized block data;

[0103] S9, splicing the plurality of normalized block data to obtain a corrected reflectivity image.

[0104] Specifically, the mosaic_files function is used to complete the mosaic operation. The georef_mosaic_setup function is responsible for setting the geographic reference information of the mosaicked image to ensure that the mosaicked image is correctly aligned on the map. The mosaic_doit function will stitch all the image blocks into a large image based on the given geographic reference and pixel size. Each image block is atmospherically corrected and calibrated before being stitched together to finally obtain a calibrated reflectance image.

[0105] like Figure 8 The system block diagram of the embodiment of the correction system for anisotropic reflection of GF6-WFV data of the present invention is shown. The present invention provides a correction system for anisotropic reflection of GF6-WFV data. The system is applied to a correction method for anisotropic reflection of GF6-WFV data. The system includes a first extraction module, a lookup table generation module, a data acquisition module, a block processing module, a second extraction module, a correction coefficient acquisition module, a first data processing module, a second data processing module, and a third data processing module. Specifically,

[0106] The first extraction module is used to extract multi-angle remote sensing data according to the historical data of MODIS BRDF to obtain a multi-angle remote sensing data set;

[0107] A lookup table generation module is used to obtain a lookup table according to the multi-angle remote sensing data set through a kernel driving model and a normalization model;

[0108] The data acquisition module is used to acquire data through GF6-WFV to obtain original image data;

[0109] A block processing module is used to process the original image data in blocks to obtain a plurality of block data;

[0110] The second extraction module is used to extract data according to the block data to obtain a block 6S radiation transfer model input file;

[0111] A correction coefficient acquisition module is used to obtain the correction coefficient of the block through the 6S radiation transfer model according to the 6S radiation transfer model input file of the block and the block data;

[0112] A first data processing module, used for performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data;

[0113] A second data processing module is used to query the lookup table according to the plurality of corrected block data to obtain a plurality of normalized block data;

[0114] The third data processing module is used to splice the multiple normalized block data to obtain a corrected reflectivity image.

[0115] The present invention provides a correction method and system for anisotropic reflection of GF6-WFV data. The invention processes GF6-WFV data in blocks, compares the data at the center of the blocks with the data extracted from the satellite navigation file data, selects a correction coefficient of the nearest point as the correction coefficient of the block, and performs atmospheric correction on the block data one by one, which significantly improves the processing efficiency and correction accuracy. The block processing divides the image data into small blocks of 2000*2000 pixels, reducing the number of pixels calculated each time, making it more suitable for fast processing of large data volumes and high-resolution images, and the block coverage is large, which is conducive to taking into account the consistency of local and overall. Secondly, the angle normalization model standardizes the zenith angle and observation angle parameters of each block, reducing the angle difference between different regions, so that the accuracy of the correction result can be improved. Finally, combined with the lookup table technology, the consumption of computing resources can be significantly reduced, which is suitable for batch processing of large-scale remote sensing images. It is easy to operate in the actual application of radiation correction and has the advantages of high efficiency and convenience.

[0116] It is to be understood that the present invention is described by the above embodiments and should not be construed as limiting the embodiments of the present invention and the scope of the present invention. It is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A correction method for anisotropic reflection of GF6-WFV data, characterized in that: The method comprises: S1. Extract multi-angle remote sensing data based on historical data of MODIS BRDF to obtain a multi-angle remote sensing data set; S2. Obtaining a lookup table according to the multi-angle remote sensing data set through a kernel-driven model and a normalized model; S3, collecting data through the GF6-WFV to obtain original image data; S4, processing the original image data into blocks to obtain a plurality of block data; S5. Extract data according to the block data to obtain a block 6S radiation transfer model input file; S6, according to the 6S radiation transfer model input file of the block and the block data, obtain the correction coefficient of the block through the 6S radiation transfer model; S7, performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data; S8. According to the plurality of corrected block data, query the lookup table to obtain a plurality of normalized block data; S9, stitching the plurality of normalized block data to obtain a corrected reflectivity image.

2. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: In S2, a lookup table is obtained according to the multi-angle remote sensing data set through a kernel-driven model and a normalized model, including: S21. According to the multi-angle remote sensing data set, a kernel-driven model is used to extract the surface reflectance to obtain the surface reflectance in the data set. The kernel-driven model includes linear weighting of an isotropic scattering component, a volume scattering component, and a geometric optical scattering component, as shown in formula (1); R(θ s ,i v ,δ,Λ)=f iso (L)+f vol (L)K vol (i s ,i v ,d)+f geo (L)K geo (i s ,i v ,d) (1) In the formula, R(θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, K vol (θ s ,θ v ,δ) is the volume scattering component kernel, K geo (θ s ,θ v ,δ) is the geometric optics scattering component kernel, f iso (Λ) is the weight of the isotropic scattering component in band Λ, f vol (Λ) is the weight of the volume scattering component in band Λ, f geo (Λ) is the weight of the geometrical optics scattering component in band Λ; S22, extracting effective information according to the surface reflectance in the data set to obtain a BRDF prototype library; S23, according to the BRDF prototype library, a lookup table is obtained by normalizing the model, and the method of the normalized model is shown in formula (2); In the formula, R n (θ s ,θ v ,δ,Λ) is the solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function at the relative azimuth δ in band Λ, R ′ (θ s ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the bidirectional reflectance distribution function in band Λ at the relative azimuth δ, R ′ n (θ s ,θ v ,δ,Λ) is the simulated solar zenith angle θ s , observation zenith angle θ v , the normalized zenith reflectivity function in band Λ at relative azimuth δ.

3. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: The solar zenith angle θ corresponding to the lookup table s The range is 0°~75°, with a resolution of 1°; the observation zenith angle θ v The range is 0°~75°, with a resolution of 2°; the relative azimuth angle δ ranges from 0°~360°.

4. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: The lookup table is a two-dimensional matrix symmetrical about the main plane, the matrix has 91 rows and 91 columns, and the numbers of the two-dimensional matrix represent the index of the optimal prototype as formula (3); Wherein, L is the row of the two-dimensional matrix, and S is the column of the two-dimensional matrix.

5. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: The step S4 divides the original image data into blocks to obtain a plurality of block data, including: dividing the original image data into 2000*2000 blocks to obtain 2000*2000 block data.

6. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: In S5, data extraction is performed according to the block data to obtain a block 6S radiation transfer model input file, including: S51, reading corresponding satellite navigation file data and extracting files related to the sun angle according to the block data; S52, parsing and obtaining a solar zenith angle array and a solar azimuth angle array according to the file related to the solar angle; S53, according to the solar zenith angle array and the solar azimuth angle array, selecting the minimum, middle and maximum three solar zenith angles in the solar zenith angle array and the corresponding solar azimuth angles in the solar azimuth angle array to obtain an input file of satellite navigation file data; S54. According to the input file of the satellite navigation file data, the file data corresponding to the input file of the satellite navigation file data is input into the satellite navigation file data, and the file data is extracted to obtain the block-based 6S radiation transmission model input file.

7. The correction method for anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: The block-wise 6S radiation transfer model input file includes: solar zenith angle, solar azimuth angle, satellite zenith angle, satellite azimuth angle, month and date of observation, aerosol optical thickness, and scattering and absorption characteristics of aerosol.

8. The method for correcting anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: In S6, according to the 6S radiation transfer model input file of the block and the block data, the correction coefficient of the block is obtained through the 6S radiation transfer model, including: S61, selecting a center point according to the block data to obtain data of the block center; S62, comparing the data of the block center with the 6S radiation transfer model input file of the block, selecting the point in the 6S radiation transfer model input file closest to the block center, and obtaining the 6S radiation transfer model input file of the center point; S63. According to the 6S radiation transfer model input file of the center point, the correction coefficients of the blocks are obtained through the 6S radiation transfer model.

9. The method for correcting anisotropic reflection of GF6-WFV data according to claim 1, characterized in that: The calculation method of the 6S radiation transfer model is formula (4); In the formula, ρ(θ s ,θ v ,φ s -φ v ) is the reflectivity of the top of the atmosphere received by the sensor, ρ a is the atmospheric path radiation reflectivity composed of molecular scattering and aerosol scattering, ρ t is the surface diffuse reflectance, S is the large balloon surface reflectance, φ s is the solar azimuth, φ v is the observation azimuth, T g (θ s ,θ v ) is the transmittance caused by atmospheric absorption, T(θ s ) is the linear transmittance of the sun-target path, T(θ v ) is the linear transmittance of the target-sensor path.

10. A correction system for anisotropic reflection of GF6-WFV data, characterized in that: The system comprises: The first extraction module is used to extract multi-angle remote sensing data according to the historical data of MODIS BRDF to obtain a multi-angle remote sensing data set; A lookup table generation module, used to obtain a lookup table according to the multi-angle remote sensing data set through a kernel driving model and a normalization model; The data acquisition module is used to acquire data through GF6-WFV to obtain original image data; A block processing module, used for performing block processing on the original image data to obtain a plurality of block data; A second extraction module is used to extract data according to the block data to obtain a block 6S radiation transfer model input file; A correction coefficient acquisition module, used to obtain the correction coefficient of the block through the 6S radiation transfer model according to the 6S radiation transfer model input file of the block and the block data; A first data processing module, used for performing corrections according to the correction coefficients of the blocks and the block data, respectively, to obtain a plurality of corrected block data; A second data processing module, configured to query the lookup table according to the plurality of corrected block data to obtain a plurality of normalized block data; The third data processing module is used to splice the plurality of normalized block data to obtain a corrected reflectivity image.

Images