Unmanned aerial vehicle multi-spectral image pixel level anisotropy correction method

Abstract

The application discloses a UAV multi-spectral image pixel-level anisotropy correction method, and belongs to the technical field of radiation correction of UAV multi-spectral images. The application aims at the problem that the existing anisotropy correction method has limitations in the processing of complex ground cover and high-resolution images, and causes poor image radiation precision. The method comprises the following steps: three-dimensional reconstruction is carried out based on a UAV multi-spectral sub-image, geographical position data and attitude data to obtain a spliced image, a digital elevation model and optimized camera parameters of the sub-image; pixel region coordinates in the spliced image are converted from a geocentric geodetic coordinate system through a camera coordinate system to an image coordinate system; multi-angle reflectivity is obtained by extracting a projection region of the pixel region in the sub-image; multi-angle data are obtained by calculating an observation imaging angle; a fitting result is obtained by using a general BRDF model to fit the multi-angle data; and the pixel-level anisotropy correction is realized by extracting a downward-looking reflectivity to replace original reflectivity of the spliced image. The application is used for anisotropy correction of UAV multi-spectral images.

Description

Technical Field

[0001] This invention relates to a pixel-level anisotropy correction method for UAV multispectral images, belonging to the field of radiometric correction technology for UAV multispectral images. Background Technology

[0002] Unmanned aerial vehicle (UAV) multispectral imagery acquires spatial and spectral characteristics of ground features by collecting their reflectance spectra using multispectral sensors. Unlike traditional remote sensing imagery, UAV multispectral imagery offers higher spatial resolution and flexible viewing angles, providing more detailed information about ground features. These images contain the spectral characteristics of ground features across various bands, reflecting subtle differences in spatial and spectral dimensions. Due to the anisotropic reflectance properties of ground surfaces, variations in viewing angle and lighting conditions can lead to significant radiometric errors in the images, affecting their accuracy and consistency. Therefore, anisotropic correction plays a crucial role in improving the quantitative accuracy of remote sensing imagery.

[0003] Anisotropy correction of UAV multispectral images can effectively eliminate radiometric errors caused by changes in observation angle, ensuring consistency and comparability of image data under different conditions. Researching pixel-level anisotropy correction methods has significant theoretical and practical value. Existing anisotropy correction methods mainly focus on "regional" or "coarse" correction methods, which have limitations in processing complex land cover and high-resolution images. Summary of the Invention

[0004] To address the limitations of existing anisotropic correction methods in processing complex surface cover and high-resolution images, which results in poor image radiometric accuracy, this invention provides a pixel-level anisotropic correction method for UAV multispectral images.

[0005] The present invention provides a pixel-level anisotropy correction method for multispectral images of unmanned aerial vehicles, comprising:

[0006] 3D reconstruction is performed based on UAV multispectral sub-images, geographic location data and attitude data to obtain stitched images, digital elevation models and sub-image optimized camera parameters;

[0007] For each pixel region in the stitched image, the three-dimensional coordinates of the pixel region in the geocentric coordinate system are converted into three-dimensional coordinates in the camera coordinate system based on the digital elevation model and the sub-image optimized camera parameters. Then, the three-dimensional coordinates in the camera coordinate system are converted into coordinates in the image coordinate system.

[0008] The projection region of the pixel region in the corresponding UAV multispectral sub-image is extracted based on the coordinates of the pixel region in the image coordinate system to obtain the multi-view reflectance of the pixel region; at the same time, the observation imaging angle of the pixel region is calculated based on the three-dimensional coordinates of the pixel region in the geocentric-ground-fixed coordinate system and the three-dimensional coordinates of the corresponding camera position in the geocentric-ground-fixed coordinate system; the multi-view reflectance and the observation imaging angle constitute multi-view data.

[0009] The multi-view data is fitted using a general BRDF model based on basis functions to obtain the multi-view data fitting results for each pixel region; then, the downward reflectance is extracted from the multi-view data fitting results, and the downward reflectance is used to replace the original reflectance of the stitched image to achieve pixel-level anisotropy correction of the UAV multispectral image.

[0010] The method for pixel-level anisotropy correction of UAV multispectral images according to the present invention converts the three-dimensional coordinates of the pixel region in the geocentric-fixed coordinate system to the three-dimensional coordinates in the camera coordinate system as follows:

[0011]

[0012] In the formula [x c ,y c ,z c [x] represents the three-dimensional coordinates of the center point of the pixel region in the camera coordinate system. e ,y e ,z e [ ] represents the three-dimensional coordinates of the center point of the pixel region in the geocentric coordinate system. This is the camera extrinsic parameter matrix transferred from the geocentric coordinate system to the camera coordinate system;

[0013]

[0014] In the formula Let be the rotation matrix from the geocentric coordinate system to the camera coordinate system. This is the translation vector from the geocentric coordinate system to the camera coordinate system;

[0015]

[0016] In the formula [t x ,t y ,t z [R] represents the three-dimensional displacement from the geocentric coordinate system to the camera coordinate system. x ,R y ,R z [ ] is the three-dimensional rotation matrix from the geocentric coordinate system to the camera coordinate system. The angle is the three-dimensional rotation from the geocentric coordinate system to the camera coordinate system.

[0017] According to the UAV multispectral image pixel-level anisotropy correction method of the present invention, the method for converting three-dimensional coordinates in the camera coordinate system to coordinates in the image coordinate system is as follows:

[0018]

[0019] In the formula [u i ,v i ] represents the coordinates of the pixel region in the image coordinate system, α represents the pixel resolution along the X-axis of the image coordinate system, β represents the pixel resolution along the Y-axis of the image coordinate system, f represents the camera focal length, [c x ,c y [ ] represents the coordinates of the camera center point in the image coordinate system. Camera intrinsic parameter matrix from camera coordinate system to image coordinate system:

[0020]

[0021] The pixel-level anisotropy correction method for UAV multispectral images according to the present invention obtains the transformation relationship between the geocentric-fixed coordinate system and the image coordinate system based on the transformation relationship between the geocentric-fixed coordinate system and the camera coordinate system, and the transformation relationship between the camera coordinate system and the image coordinate system:

[0022]

[0023] The method for pixel-level anisotropy correction of UAV multispectral images according to the present invention includes the following method for obtaining the multi-view reflectance of the pixel region:

[0024] A 3×3 pixel region is selected and denoted as S={P(p,q)p∈[x-1,x+1],q∈[y-1,y+1]}, where P(p,q) represents the pixel point of the pixel region in the stitched image, (p,q) represents the pixel point coordinates of the pixel region, and (x,y) represents the pixel point coordinates of the stitched image in the image coordinate system;

[0025] The projection coordinates of the boundary points of the pixel region S on the corresponding UAV multispectral sub-image are calculated based on the expression of the pixel region S, thereby extracting the projection region of the pixel region on the corresponding UAV multispectral sub-image and obtaining the multi-view reflectance of the pixel region.

[0026]

[0027] In the formula, R represents the multi-view reflectance of a pixel region in each UAV multispectral sub-image. This represents the projection region of pixel region S in the w-th UAV multispectral sub-image. Indicates the projection area The mean is the average reflectance, where mean represents the arithmetic mean.

[0028] According to the UAV multispectral image pixel-level anisotropy correction method of the present invention, the observation imaging angle of the pixel region includes the solar zenith angle θ. s Sun azimuth angle φ s Observing the zenith angle θ v and the observed azimuth angle φ v Solar zenith angle θ s And solar azimuth φ s These are the horizontal angles of the sun and the camera relative to the ground, measured clockwise from true north.

[0029] Observation of zenith angle θ v and the observed azimuth angle φ v These represent the perpendicular angles of the sun and the camera to the observation point, respectively.

[0030]

[0031] In the formula This represents the three-dimensional coordinates of the camera position in the Earth-centered, Earth-fixed coordinate system.

[0032] The pixel-level anisotropy correction method for UAV multispectral images according to the present invention uses infinite series basis functions to approximate the theoretical BRDF model and uses hemispherical harmonic basis functions to represent the BRDF model:

[0033]

[0034] In the formula Let be a hemispherical harmonic basis function of order l with components m. Let be the coefficients of a hemispherical harmonic basis function of order l and components m;

[0035]

[0036] In the formula The normalization coefficient is... cosθ v The related Legendre polynomial.

[0037] According to the UAV multispectral image pixel-level anisotropy correction method of the present invention, given the order n, the BRDF model is obtained as follows:

[0038]

[0039] set up:

[0040]

[0041] Where k∈[1,2n+1];

[0042] Let M represent the total number of multi-view data, and ignore spectral correlation. Establish a system of linear equations for each band:

[0043]

[0044] In the formula R M The multi-view reflectivity of the Mth viewpoint; and This represents the Mth value corresponding to the variable;

[0045] Solve the system of linear equations to obtain the multi-view data fitting results for each pixel region.

[0046] The method for pixel-level anisotropy correction of UAV multispectral images according to the present invention includes the following method for solving the linear equation system:

[0047] Representing the system of linear equations in matrix form:

[0048] HA = R, where:

[0049]

[0050] and

[0051] The problem of solving a system of linear equations is transformed into a convex optimization problem; the objective function is constructed as follows:

[0052]

[0053] The least squares method was used to obtain the multi-view data fitting results.

[0054]

[0055] In the formula These are the optimal coefficients.

[0056] The method for extracting downward-looking reflectance using the UAV multispectral image pixel-level anisotropy correction method according to the present invention is as follows:

[0057] Let θ v =0, φ v =0, extracting the downward reflectivity R nadir :

[0058]

[0059] The beneficial effects of this invention are as follows: The pixel-level anisotropy correction method of this invention, by extracting pixel-level multi-view reflectance data, can more accurately process the local features of ground objects, overcoming the shortcomings of traditional methods in detail capture. In the application of high-resolution remote sensing imagery, the method of this invention can effectively alleviate the radiometric errors introduced by changes in imaging perspective during the acquisition of UAV multispectral images, thereby more effectively improving the radiometric accuracy of the images. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating the pixel-level anisotropy correction method for UAV multispectral images described in this invention.

[0061] Figure 2 This is a schematic diagram of a stitched image that has not undergone anisotropic correction;

[0062] Figure 3 This is a schematic diagram of the stitched image after being corrected by the method of this invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0065] The present invention will be further described below with reference to the accompanying drawings, but this should not be construed as limiting the invention.

[0066] Specific Implementation Method 1: Combination Figure 1 As shown, this invention provides a pixel-level anisotropy correction method for multispectral images of unmanned aerial vehicles, including,

[0067] 3D reconstruction is performed based on UAV multispectral sub-images, geographic location data and attitude data to obtain stitched images, digital elevation models and sub-image optimized camera parameters;

[0068] For each pixel region in the stitched image, the three-dimensional coordinates of the pixel region in the geocentric coordinate system are converted into three-dimensional coordinates in the camera coordinate system based on the digital elevation model and the sub-image optimized camera parameters. Then, the three-dimensional coordinates in the camera coordinate system are converted into coordinates in the image coordinate system.

[0069] The projection region of the pixel region in the corresponding UAV multispectral sub-image is extracted based on the coordinates of the pixel region in the image coordinate system to obtain the multi-view reflectance of the pixel region; at the same time, the observation imaging angle of the pixel region is calculated based on the three-dimensional coordinates of the pixel region in the geocentric-ground-fixed coordinate system and the three-dimensional coordinates of the corresponding camera position in the geocentric-ground-fixed coordinate system; the multi-view reflectance and the observation imaging angle constitute multi-view data.

[0070] The multi-view data is fitted using a general BRDF model based on basis functions to obtain the multi-view data fitting results for each pixel region; then, the downward reflectance is extracted from the multi-view data fitting results, and the downward reflectance is used to replace the original reflectance of the stitched image to achieve pixel-level anisotropy correction of the UAV multispectral image.

[0071] This implementation first processes the input data, then extracts pixel-level multi-view data for each pixel region based on the stitched image, digital elevation model, and optimized camera parameters. Next, it performs BRDF model fitting and correction, using a basis function-based universal bidirectional reflectance distribution function (BRDF) model to fit the extracted multi-view data and predict downward reflectance, achieving pixel-level anisotropy correction. This implementation effectively mitigates radiometric errors introduced by changes in imaging angle during the acquisition of UAV multispectral images.

[0072] Furthermore, the method for converting the 3D coordinates of the pixel region in the Earth-Centered Earth-Fixed (ECEF) coordinate system to the 3D coordinates in the camera coordinate system is as follows:

[0073]

[0074] In the formula [x c ,y c ,z c [x] represents the three-dimensional coordinates of the center point of the pixel region in the camera coordinate system. e ,y e ,z e [ ] represents the three-dimensional coordinates of the center point of the pixel region in the geocentric coordinate system. This is the camera extrinsic parameter matrix transferred from the geocentric coordinate system to the camera coordinate system;

[0075]

[0076] In the formula Let be the rotation matrix from the geocentric coordinate system to the camera coordinate system. This is the translation vector from the geocentric coordinate system to the camera coordinate system;

[0077]

[0078] In the formula [t x ,t y ,t z [R] represents the three-dimensional displacement from the geocentric coordinate system to the camera coordinate system. x ,R y ,R z [ ] is the three-dimensional rotation matrix from the geocentric coordinate system to the camera coordinate system. is the three-dimensional rotation angle from the geocentric coordinate system to the camera coordinate system. [t] x ,t y ,t z ]and These are the optimized camera parameters.

[0079] The method for converting 3D coordinates in the camera coordinate system to coordinates in the image coordinate system is as follows:

[0080]

[0081] In the formula [u i ,v i ] represents the coordinates of the pixel region in the image coordinate system, α represents the pixel resolution along the X-axis of the image coordinate system, β represents the pixel resolution along the Y-axis of the image coordinate system, f represents the camera focal length, [c x ,c y [ ] represents the coordinates of the camera center point in the image coordinate system. Camera intrinsic parameter matrix from camera coordinate system to image coordinate system:

[0082]

[0083] In summary, based on the transformation relationship between the geocentric-fixed coordinate system and the camera coordinate system, and the transformation relationship between the camera coordinate system and the image coordinate system, the transformation relationship between the geocentric-fixed coordinate system and the image coordinate system is obtained as follows:

[0084]

[0085] Based on the above formula, the projection area of ​​a given pixel region onto the corresponding sub-image can be extracted, thus realizing the extraction of pixel-level reflectance data.

[0086] Furthermore, the method for obtaining the multi-view reflectance of the pixel region includes:

[0087] A 3×3 pixel region is selected and denoted as S={P(p,q)p∈[x-1,x+1],q∈[y-1,y+1]}, where P(p,q) represents the pixel point of the pixel region in the stitched image, (p,q) represents the pixel point coordinates of the pixel region, and (x,y) represents the pixel point coordinates of the stitched image in the image coordinate system;

[0088] The projection coordinates of the boundary points of the pixel region S on the corresponding UAV multispectral sub-image are calculated based on the expression of the pixel region S, thereby extracting the projection region of the pixel region on the corresponding UAV multispectral sub-image and obtaining the multi-view reflectance of the pixel region.

[0089]

[0090] In the formula, R represents the multi-view reflectance of a pixel region in each UAV multispectral sub-image. This represents the projection region of pixel region S in the w-th UAV multispectral sub-image. Indicates the projection area The mean is the average reflectance, where mean represents the arithmetic mean.

[0091] In this embodiment, the imaging perspective is a set of angular parameters that define the spatial relationship between the sun, the camera, and the ground target; the observation imaging perspective of the pixel region includes the solar zenith angle θ. s Sun azimuth angle φ s Observing the zenith angle θ v and the observed azimuth angle φ v Solar zenith angle θ s And solar azimuth φ s These are the horizontal angles of the sun and the camera relative to the ground, measured clockwise from true north.

[0092] Observation of zenith angle θ v and the observed azimuth angle φ v These represent the perpendicular angles of the sun and the camera to the observation point, respectively.

[0093]

[0094] In the formula This represents the three-dimensional coordinates of the camera position in the Earth-centered, Earth-fixed coordinate system.

[0095] The theoretical BRDF model is approximated using infinite series basis functions. To ensure alignment of the domain, hemispherical harmonic basis functions are used to represent the BRDF model.

[0096]

[0097] In the formula Let be a hemispherical harmonic basis function of order l with components m. Let be the coefficients of a hemispherical harmonic basis function of order l and components m;

[0098]

[0099]

[0100] In the formula The normalization coefficient is... cosθ v The related Legendre polynomial.

[0101] The parameter fitting and downward reflectance extraction process of the BRDF model is as follows:

[0102] Ideally, BRDF is represented by an infinite order, but in practice, it is usually approximated with a given order.

[0103] Given the order n, the BRDF model is as follows:

[0104]

[0105] set up:

[0106]

[0107] Where k∈[1,2n+1];

[0108] Let M represent the total number of multi-view data, and n represent the highest order of the HSH approximation. Ignoring spectral correlation, a system of linear equations is established for each band:

[0109]

[0110] In the formula R M The multi-view reflectivity of the Mth viewpoint; and This represents the Mth value corresponding to the variable;

[0111] Solve the system of linear equations to obtain the multi-view data fitting results for each pixel region.

[0112] Furthermore, the method for solving a system of linear equations is as follows:

[0113] Representing the system of linear equations in matrix form:

[0114] HA = R, where:

[0115]

[0116] and

[0117] Considering that the system of equations is overdetermined and does not have a unique solution, the problem of solving the linear equation system is transformed into a convex optimization problem; by constructing an objective function and minimizing it, the optimal solution is obtained:

[0118]

[0119] The least squares method was used to obtain the multi-view data fitting results.

[0120]

[0121] In the formula These are the optimal coefficients.

[0122] The method for extracting downward reflectance is as follows:

[0123] Let θ v =0, φ v =0, extracting the downward reflectivity R nadir :

[0124]

[0125] Anisotropy correction can be achieved by replacing the original reflectivity with the downward reflectivity.

[0126] Verification experiment:

[0127] The system acquired multispectral image data from the UAV, containing 10 bands with a wavelength range of 400-840 nm and a ground resolution of 3.5 cm. The data underwent radiometric correction preprocessing and was converted into reflectance data. Figure 2 This is a stitched image that has not undergone anisotropy correction. Figure 3 The image shown is a stitched image after correction using the method of this invention. Table 1 compares the reflectance errors of the ground samples before and after correction.

[0128] Table 1

[0129]

[0130] It can be seen that the image corrected by the method of the present invention can effectively alleviate the radiation error introduced by the change of imaging angle, which proves the effectiveness of the method of the present invention.

[0131] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A pixel-level anisotropy correction method for multispectral images from unmanned aerial vehicles (UAVs), characterized in that, include, 3D reconstruction is performed based on UAV multispectral sub-images, geographic location data and attitude data to obtain stitched images, digital elevation models and sub-image optimized camera parameters; For each pixel region in the stitched image, the three-dimensional coordinates of the pixel region in the geocentric coordinate system are converted into three-dimensional coordinates in the camera coordinate system based on the digital elevation model and the sub-image optimized camera parameters. Then, the three-dimensional coordinates in the camera coordinate system are converted into coordinates in the image coordinate system. Based on the coordinates of the pixel region in the image coordinate system, the projection region of the pixel region in the corresponding UAV multispectral sub-image is extracted to obtain the multi-view reflectance of the pixel region; Simultaneously, the observation imaging angle of the pixel region is calculated based on the three-dimensional coordinates of the pixel region in the geocentric-fixed coordinate system and the three-dimensional coordinates of the corresponding camera position in the geocentric-fixed coordinate system; the multi-view reflectivity and the observation imaging angle constitute multi-view data; The multi-view data is fitted using a general BRDF model based on basis functions to obtain the multi-view data fitting results for each pixel region; then, the downward reflectance is extracted from the multi-view data fitting results, and the downward reflectance is used to replace the original reflectance of the stitched image to achieve pixel-level anisotropy correction of the UAV multispectral image.

2. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 1, characterized in that, The method for converting the 3D coordinates of a pixel region in the geocentric coordinate system to the 3D coordinates in the camera coordinate system is as follows: In the formula [x c ,y c ,z c [x] represents the three-dimensional coordinates of the center point of the pixel region in the camera coordinate system. e ,y e ,z e [ ] represents the three-dimensional coordinates of the center point of the pixel region in the geocentric coordinate system. This is the camera extrinsic parameter matrix transferred from the geocentric coordinate system to the camera coordinate system; In the formula Let be the rotation matrix from the geocentric coordinate system to the camera coordinate system. This is the translation vector from the geocentric coordinate system to the camera coordinate system; In the formula [t x ,t y ,t z [R] represents the three-dimensional displacement from the geocentric coordinate system to the camera coordinate system. x ,R y ,R z [ ] is the three-dimensional rotation matrix from the geocentric coordinate system to the camera coordinate system. The angle is the three-dimensional rotation from the geocentric coordinate system to the camera coordinate system.

3. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 2, characterized in that, The method for converting 3D coordinates in the camera coordinate system to coordinates in the image coordinate system is as follows: In the formula [u i ,v i ] represents the coordinates of the pixel region in the image coordinate system, α represents the pixel resolution along the X-axis of the image coordinate system, β represents the pixel resolution along the Y-axis of the image coordinate system, f represents the camera focal length, [c x ,c y [ ] represents the coordinates of the camera center point in the image coordinate system. Camera intrinsic parameter matrix from camera coordinate system to image coordinate system:

4. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 3, characterized in that, Based on the transformation relationships between the geocentric-fixed coordinate system and the camera coordinate system, and between the camera coordinate system and the image coordinate system, the transformation relationship between the geocentric-fixed coordinate system and the image coordinate system is obtained:

5. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 4, characterized in that, Methods for obtaining the multi-view reflectance of the pixel region include: A 3×3 pixel region is selected and denoted as S={P(p,q)|p∈[x-1,x+1],q∈[y-1,y+1]}, where P(p,q) represents the pixel point of the pixel region in the stitched image, (p,q) represents the pixel coordinates of the pixel point in the pixel region, and (x,y) represents the pixel coordinates of the stitched image in the image coordinate system; The projection coordinates of the boundary points of the pixel region S on the corresponding UAV multispectral sub-image are calculated based on the expression of the pixel region S, thereby extracting the projection region of the pixel region on the corresponding UAV multispectral sub-image and obtaining the multi-view reflectance of the pixel region. In the formula, R represents the multi-view reflectance of a pixel region in each UAV multispectral sub-image. This represents the projection region of pixel region S in the w-th UAV multispectral sub-image. Indicates the projection area The mean is the average reflectance, where mean represents the arithmetic mean.

6. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 5, characterized in that, The observation imaging angle of the pixel region includes the solar zenith angle θ. s Sun azimuth angle φ s Observing the zenith angle θ v and the observed azimuth angle φ v Solar zenith angle θ s And solar azimuth φ s These are the horizontal angles of the sun and the camera relative to the ground, measured clockwise from true north. Observation of zenith angle θ v and the observed azimuth angle φ v These represent the perpendicular angles of the sun and the camera to the observation point, respectively. In the formula This represents the three-dimensional coordinates of the camera position in the Earth-centered, Earth-fixed coordinate system.

7. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 6, characterized in that, The theoretical BRDF model is approximated using infinite series basis functions, and the BRDF model is represented using hemispherical harmonic basis functions: In the formula Let be a hemispherical harmonic basis function of order l with components m. Let be the coefficients of a hemispherical harmonic basis function of order l and components m; In the formula The normalization coefficient is... cosθ v The related Legendre polynomial.

8. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 7, characterized in that, Given the order n, the BRDF model is as follows: set up: Where k∈[1,2n+1]; Let M represent the total number of multi-view data, and ignore spectral correlation. Establish a system of linear equations for each band: In the formula R M θ represents the multi-view reflectivity from the Mth viewpoint; v M and φ v M This represents the Mth value corresponding to the variable; Solve the system of linear equations to obtain the multi-view data fitting results for each pixel region.

9. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 8, characterized in that, The method for solving a system of linear equations is as follows: Representing the system of linear equations in matrix form: HA = R, where: and The problem of solving a system of linear equations is transformed into a convex optimization problem; the objective function is constructed as follows: The least squares method was used to obtain the multi-view data fitting results. In the formula These are the optimal coefficients.

10. The method for pixel-level anisotropy correction of UAV multispectral images according to claim 9, characterized in that, The method for extracting downward reflectance is as follows: Let θ v =0, φ v =0, extracting the downward reflectivity R nadir :

Images