A method and apparatus for estimating surface reflectance in the visible red channel

Abstract

The application provides a visible red channel ground reflectivity estimation method and device, the method comprising: obtaining a remote sensing image to be processed; determining bright pixels and dark pixels in the remote sensing image to be processed based on a vegetation index; for the bright pixels, constructing a quadratic function model of the red light band ground reflectivity and the 1.64 mu m band apparent reflectivity, and calculating the red light band ground reflectivity according to the quadratic function model; for the dark pixels, constructing a linear function model of the red light band ground reflectivity and the 1.64 mu m band apparent reflectivity, and calculating the red light band ground reflectivity according to the linear function model. The visible red channel ground reflectivity estimation method provided by the scheme simultaneously realizes accurate estimation of the ground reflectivity of dark ground and bright ground under the visible red channel.

Description

Technical Field

[0001] This invention relates to the field of remote sensing technology, and in particular to a method and apparatus for estimating the surface reflectance of visible red light channels. Background Technology

[0002] In numerous remote sensing observation studies, such as atmospheric parameter inversion and ground feature identification, surface reflectance is a fundamental parameter, and its estimation accuracy directly affects the measurement accuracy of other remote sensing parameters. The visible red channel is easily affected by the atmosphere and aerosols, making it difficult to estimate the space-based downward-looking surface reflectance of this channel.

[0003] Existing remote sensing methods for estimating land surface reflectance mainly include two classic algorithms: the Deep Blue algorithm and the dark target method. The Deep Blue algorithm assumes that the land surface is stable over a certain period, thus it can directly access statistically based land surface reflectance databases. However, introducing external land surface reflectance databases inevitably introduces geometric matching errors. The classic dark target method has the advantages of simple model, universality, strong timeliness, and high accuracy, but it is not suitable for bright surfaces such as arid and semi-arid land. Therefore, existing algorithms cannot meet the application requirements, hindering the application of remote sensing data. Therefore, it is necessary to propose an accurate estimation model that can simultaneously achieve both dark and bright land surface reflectance under visible and red light channels. Summary of the Invention

[0004] This invention provides a method and apparatus for estimating surface reflectance in the visible red light channel. The method can simultaneously and accurately estimate the surface reflectance of both dark and bright surfaces in the visible red light channel.

[0005] In a first aspect, embodiments of the present invention provide a method for estimating the surface reflectance of the visible red light channel, comprising:

[0006] Acquire the remote sensing image to be processed;

[0007] Bright and dark pixels in the remote sensing image to be processed are determined based on the vegetation index.

[0008] For the bright pixel, a quadratic function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band is constructed, and the surface reflectance in the red band is calculated based on the quadratic function model.

[0009] For the dark pixels, a linear function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band is constructed, and the surface reflectance in the red band is calculated based on the linear function model.

[0010] Optionally, determining the bright and dark pixels in the remote sensing image to be processed based on vegetation indices includes:

[0011] Acquire historical remote sensing images, including apparent reflectance of each channel;

[0012] The historical remote sensing images are calculated and analyzed to obtain the normalized vegetation index for each pixel.

[0013] Based on the normalized vegetation index, the deaerosol-polluted vegetation index at the 1.64 μm band was calculated.

[0014] For each cell, perform the following:

[0015] Determine whether the deaerosol-polluted vegetation index of the pixel is within the preset deaerosol-polluted vegetation index range.

[0016] If so, then the pixel is determined to be a bright pixel;

[0017] If not, then the pixel is determined to be a dark pixel.

[0018] Optionally, the preset range of the deaerosol-polluted vegetation index is (0, 0.25).

[0019] Optionally, for the bright pixel, constructing a quadratic function model of the red band surface reflectance and the 1.64μm band apparent reflectance, and calculating the red band surface reflectance based on the quadratic function model, includes:

[0020] The quadratic function model is as follows:

[0021]

[0022] Among them, R Red R is used to characterize the surface reflectance in the red light band. 1.64μm The coefficients C, D, and E are used to characterize the apparent reflectance of the 1.64 μm band. C is used to characterize the quadratic fitting coefficient, D is used to characterize the first-order fitting coefficient, and E is used to characterize the fitting coefficient of the constant term. Among them, C, D, and E are all related to the deaerosol-polluted vegetation index at the 1.64 μm band.

[0023] Optionally, for the dark pixel, constructing a linear function model of the red band surface reflectance and the 1.64μm band apparent reflectance, and calculating the red band surface reflectance based on the linear function model, includes:

[0024] The linear function model is as follows:

[0025] R Red =SlopeR 1.64μm +Intercept

[0026] Slope = aNDVI + b

[0027]

[0028] NDVI = (R 0.87μm -R Red ) / (R 0.87μm +R Red )

[0029] wherein R Red is used to represent the surface reflectance in the red light band, R 1.64μm is used to represent the apparent reflectance in the 1.64 μm band, Slope is used to represent the slope, Intercept is used to represent the intercept, NDVI is used to represent the normalized difference vegetation index of the dark pixel, and R 0.87μm is used to represent the apparent reflectance in the 0.87 μm band; wherein both Slope and Intercept are related to the normalized difference vegetation index, a, b, d, e and f are all used to represent fitting coefficients, and c is used to represent the intercept value when 0.325 < NDVI ≤ 0.65.

[0030] In a second aspect, an embodiment of the present invention further provides an apparatus for estimating surface reflectance of a visible red light channel, comprising:

[0031] an acquisition module, configured to acquire a to-be-processed remote sensing image;

[0032] a bright and dark surface determination module, configured to determine bright pixels and dark pixels in the to-be-processed remote sensing image based on a vegetation index;

[0033] a bright surface estimation module, configured to construct, for the bright pixels, a quadratic function model of red light band surface reflectance and 1.64 μm band apparent reflectance, and calculate the red light band surface reflectance according to the quadratic function model;

[0034] a dark surface estimation module, configured to construct, for the dark pixels, a linear function model of red light band surface reflectance and 1.64 μm band apparent reflectance, and calculate the red light band surface reflectance according to the quadratic function model.

[0035] Optionally, the quadratic function model adopted by the bright surface estimation module is:

[0036]

[0037] wherein R Red is used to represent the surface reflectance in the red light band, R 1.64μm is used to represent the apparent reflectance in the 1.64 μm band, C is used to represent a quadratic fitting coefficient, D is used to represent a linear fitting coefficient, and E is used to represent a constant term fitting coefficient; wherein C, D and E are all related to the de-aerosol pollution vegetation index in the 1.64 μm band.

[0038] Optionally, the linear function model adopted by the dark surface estimation module is:

[0039] R Red =SlopeR 1.64μm +Intercept

[0040] Slope=aNDVI+b

[0041]

[0042] NDVI=(R 0.87μm -R Red ) / (R 0.87μm +R Red )

[0043] wherein, R Red is used to represent the surface reflectance in the red light band, R 1.64μm is used to represent the apparent reflectance in the 1.64μm band, Slope is used to represent the slope, Intercept is used to represent the intercept, NDVI is used to represent the normalized difference vegetation index of the dark pixel, R 0.87μm is used to represent the apparent reflectance in the 0.87μm band; wherein, both Slope and Intercept are related to the normalized difference vegetation index, a, b, d, e and f are all used to represent fitting coefficients, and c is used to represent the intercept value when 0.325<NDVI≤0.65.

[0044] In a third aspect, an embodiment of the present invention further provides a computing device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the visible red light channel surface reflectance estimation method described in any one of the above items is implemented.

[0045] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the visible red light channel surface reflectance estimation method described in any one of the above items.

[0046] This invention provides a method and apparatus for estimating surface reflectance in the visible-red light channel. Based on vegetation indices, the method divides pixels in the remote sensing image to be processed into bright pixels corresponding to bright surfaces and dark pixels corresponding to dark surfaces. Then, it constructs quadratic function models for the corresponding bright pixels and linear function models for the corresponding dark pixels to estimate the surface reflectance in the red light band. Thus, by using these two function models, accurate estimation of surface reflectance for both dark and bright surfaces in the visible-red light channel can be achieved simultaneously. This solves the problem that introducing geometric matching errors in bright surface reflectance when using external databases leads to low accuracy in estimating surface reflectance in the red light band, thereby improving the accuracy of subsequent remote sensing parameter inversion using remote sensing data. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a method for estimating surface reflectance in the visible red light channel according to an embodiment of the present invention;

[0049] Figures 2a to 2d This is a scatter plot of the relationship between the 0.67μm band surface reflectance and the 1.64μm apparent reflectance provided in an embodiment of the present invention;

[0050] Figure 3 The absolute coefficients for the estimation model under bright ground surfaces provided in one embodiment of the present invention are the first-order and second-order fittings, respectively.

[0051] Figure 4 The quadratic fitting coefficients and AFRI in the quadratic function model provided in one embodiment of the present invention are... 1.64 The functional relationship;

[0052] Figure 5 The first-order fitting coefficients and AFRI in the quadratic function model provided in one embodiment of the present invention are 1.64 The functional relationship;

[0053] Figure 6 The constant term fitting coefficient and AFRI in the quadratic function model provided in one embodiment of the present invention are... 1.64 The functional relationship;

[0054] Figure 7 This is a functional relationship between the slope and NDVI in a linear function model provided by an embodiment of the present invention;

[0055] Figure 8 This is a functional relationship between the intercept and NDVI in a linear function model provided by an embodiment of the present invention;

[0056] Figure 9 This is a hardware architecture diagram of a computing device provided in an embodiment of the present invention;

[0057] Figure 10 This is a structural diagram of a visible red light channel surface reflectance estimation device provided in an embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, this embodiment of the invention provides a method for estimating the surface reflectance of the visible red light channel, the method comprising:

[0060] Step 100: Obtain the remote sensing image to be processed;

[0061] Step 102: Determine the bright and dark pixels in the remote sensing image to be processed based on the vegetation index;

[0062] Step 104: For bright pixels, construct a quadratic function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band, and calculate the surface reflectance in the red band based on the quadratic function model;

[0063] Step 106: For dark pixels, construct a linear function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band, and calculate the surface reflectance in the red band based on the linear function model.

[0064] In this embodiment of the invention, based on the vegetation index, pixels in the remote sensing image to be processed are divided into bright pixels corresponding to bright surfaces and dark pixels corresponding to dark surfaces. Then, a quadratic function model for the corresponding bright pixels and a linear function model for the corresponding dark pixels are constructed to estimate the surface reflectance in the red band. Thus, using these two function models, accurate estimation of surface reflectance for both dark and bright surfaces in the visible red light channel can be achieved simultaneously. This solves the problem that introducing geometric matching errors in bright surface reflectance when using external databases leads to low accuracy in estimating surface reflectance in the red band, thereby improving the accuracy of subsequent remote sensing parameter inversion using remote sensing data.

[0065] The following description Figure 1 The execution method of each step is shown.

[0066] First, for step 100, the remote sensing image to be processed is acquired. This image includes apparent reflectance values ​​at the 0.87 μm band and apparent reflectance values ​​at the 1.64 μm band.

[0067] For step 102, determining the bright and dark pixels in the remote sensing image to be processed based on vegetation indices includes:

[0068] Acquire historical remote sensing images, including apparent reflectance of each channel;

[0069] The normalized vegetation index for each pixel is obtained by calculating and analyzing historical remote sensing images.

[0070] Based on the normalized vegetation index, the deaerosol-polluted vegetation index at the 1.64 μm band was calculated.

[0071] For each cell, perform the following:

[0072] Determine whether the deaerosol-polluted vegetation index of the pixel is within the preset deaerosol-polluted vegetation index range.

[0073] If so, then the pixel is determined to be a bright pixel;

[0074] If not, then the pixel is determined to be a dark pixel.

[0075] In a preferred embodiment, the preset range of the deaerosol-polluted vegetation index is (0, 0.25).

[0076] It should be noted that the preset aerosol-pollution-free vegetation index range corresponds to the preset normalized vegetation index used to distinguish between bright and dark pixels; the value range of this preset normalized vegetation index range is (0, 0.325). Bright surfaces include sparsely vegetated arid and semi-arid surfaces; dark surfaces include densely vegetated surfaces.

[0077] Specifically, in this invention, the land surface type corresponding to the remote sensing data of each pixel is first determined. The Normalized Difference Vegetation Index (NDVI) is used to describe the cover density of land vegetation. The higher the vegetation density, the higher the NDVI, which is obtained through the following formula:

[0078] NDVI = (R 0.87μm -R Red ) / (R 0.87μm +R Red (1)

[0079] For the condition 0 < NDVI ≤ 0.325, the surface reflectance in the 0.67 μm band and the apparent reflectance in the 1.64 μm band have the following relationship:

[0080]

[0081] Based on the relationship between surface reflectance in the visible red light channel and apparent reflectance in the shortwave infrared channel of a bright surface according to formula (2), the aerosol-free vegetation index (AFRI) at the 1.64 μm band can be further constructed, and its expression is as follows:

[0082]

[0083] In formulas (1) to (3), NDVI is used to characterize the normalized vegetation index; R 0.87μm Used to characterize the apparent reflectance in the 0.87μm band; R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band; R Red Used to characterize the surface reflectance in the red band to be estimated; AFRI 1.64 Used to characterize the deaerosol-polluted vegetation index at the 1.64 μm band;

[0084] For example, by analyzing historical remote sensing image set I, the preset range of de-aerosol-polluted vegetation index can be determined, as shown in Figure 2 (including...). Figure 2a , Figure 2b , Figure 2c and Figure 2d As shown in the figure, when 0 < NDVI ≤ 0.325, the corresponding AFRI 1.64 The value range is (0, 0.25], therefore AFRI is adopted. 1.64 Surface delineation: 0 < AFRI 1.64 When ≤0.25, the ground surface is a bright surface, and the corresponding pixel is a bright pixel; AFRI 1.64 When the value is greater than 0.25, the surface is dark, and the corresponding pixel is a dark pixel.

[0085] In step 104, the quadratic function model is:

[0086]

[0087] Among them, R Red R is used to characterize the surface reflectance in the red light band. 1.64μm C is used to characterize the apparent reflectance in the 1.64 μm band, D is used to characterize the quadratic fitting coefficient, E is used to characterize the linear fitting coefficient, and E is used to characterize the constant term fitting coefficient; among them, C, D and E are all related to the deaerosol-polluted vegetation index in the 1.64 μm band.

[0088] It should be noted that the specific values ​​of C, D, and E were obtained by fitting historical remote sensing data from the same remote sensing device.

[0089] In step 106, the linear function model is:

[0090] R Red =SlopeR 1.64μm +Intercept (5)

[0091] Slope=aNDVI+b (6)

[0092]

[0093] NDVI=(R 0.87μm -R Red ) / (R 0.87μm +R Red ) (1)

[0094] wherein, R Red is used to represent the surface reflectance in the red light band, R 1.64μm is used to represent the apparent reflectance in the 1.64 μm band, Slope is used to represent the slope, Intercept is used to represent the intercept, NDVI is used to represent the normalized difference vegetation index of the dark pixel, R 0.87μm is used to represent the apparent reflectance in the 0.87 μm band; wherein, both Slope and Intercept are related to the normalized difference vegetation index, a, b, d, e and f are all used to represent fitting coefficients, and c is used to represent the intercept value when 0.325<NDVI≤0.65.

[0095] In the present invention, for bright surfaces and dark surfaces, by analyzing the influence of the normalized difference vegetation index and the aerosol-free vegetation index in historical remote sensing images on the surface reflectance of the visible red channel and the surface reflectance of the shortwave infrared channel, it is found that when 0<AFRI 1.64 ≤0.25, 0<NDVI≤0.325, the relationship between the surface reflectance of the visible red channel and the apparent reflectance at 1.64 μm is a quadratic function related to AFRI 1.64 ; when AFRI 1.64 >0.25, NDVI>0.325, the relationship between the surface reflectance of the visible red channel and the apparent reflectance at 1.64 μm is a linear function related to NDVI. Therefore, a quadratic function model and a linear function model are established respectively to estimate the surface reflectance of the red light band (i.e., the surface reflectance of the visible red channel to be estimated). The function model thus obtained is simple, has strong universality and higher accuracy. It should be noted that Figure 3 shows the coefficient of determination R for linear fitting and quadratic fitting of the relationship between the surface reflectance of the visible red channel and the apparent reflectance at 1.64 μm when 0<AFRI 1.64 ≤0.25 in the historical remote sensing image set I 2The change in Ri of the quadratic fit is obvious. 2 The higher value further confirms that the quadratic fit under the bright surface is more consistent with the distribution pattern between the two.

[0096] Specifically, 0 < AFRI 1.64 When ≤0.25, a quadratic function model is used. Figures 4 to 6 The fitting coefficients C, D, and E obtained based on historical remote sensing image set I are shown in relation to AFRI. 1.64 The relationship.

[0097] Specifically, AFRI 1.64 When the value is greater than 0.25, a linear function model is used. Figure 7 This shows the functional relationship between the slope and NDVI obtained from historical remote sensing image set I. Figure 8 This shows the functional relationship between the intercept and NDVI obtained from historical remote sensing image set I. The linear function model in this case is:

[0098] R Red =SlopeR 1.64μm +Intercept

[0099] Slope = -0.51422NDVI + 0.56505

[0100]

[0101] NDVI = (R 0.87μm -R Red ) / (R 0.87μm +R Red );

[0102] Then, based on the four formulas in the linear function model, the corresponding surface reflectance in the red band is calculated.

[0103] At the same time, by Figure 7 and Figure 8 It can also be seen that the coefficients of determination |R| or R0 for each functional relationship 2 All values ​​are greater than 0.9, further confirming that the next fit of the dark surface is more consistent with the data distribution pattern between the surface reflectance in the red band and the apparent reflectance in the 1.64μm band.

[0104] like Figure 9 , Figure 10 As shown, this embodiment of the invention provides a device for estimating the surface reflectance in the visible red light channel. The device embodiment can be implemented through software, hardware, or a combination of both. From a hardware perspective, such as... Figure 9 The diagram shown is a hardware architecture diagram of a computing device housing a visible red light channel surface reflectance estimation device provided in an embodiment of the present invention, except for... Figure 9 In addition to the processor, memory, network interface, and non-volatile memory shown, the computing device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 10 As shown, as a logical device, it is formed by the CPU of its computing device reading the corresponding computer program from the non-volatile memory into the memory for execution. This embodiment provides a visible red light channel surface reflectance estimation device, including: an acquisition module 1000, a bright / dark surface determination module 1002, a bright surface estimation module 1004, and a dark surface estimation module 1006;

[0105] Acquisition module 1000 is used to acquire remote sensing images to be processed;

[0106] Bright and dark surface determination module 1002 is used to determine bright pixels and dark pixels in the remote sensing image to be processed based on vegetation index;

[0107] The bright surface estimation module 1004 is used to construct a quadratic function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band for bright pixels, and calculate the surface reflectance in the red band based on the quadratic function model.

[0108] The dark surface estimation module 1006 is used to construct a linear function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band for dark pixels, and calculate the surface reflectance in the red band based on the linear function model.

[0109] In some specific implementations, the acquisition module 1000 can be used to perform the above step 100, the bright and dark surface determination module 1002 can be used to perform the above step 102, the bright surface estimation module 1004 can be used to perform the above step 104, and the dark surface estimation module 1006 can be used to perform the above step 106.

[0110] In some specific implementations, the light / dark surface determination module 1002 is also used to perform the following operations:

[0111] Acquire historical remote sensing images, including apparent reflectance of each channel;

[0112] The normalized vegetation index for each pixel is obtained by calculating and analyzing historical remote sensing images.

[0113] Based on the normalized vegetation index, the deaerosol-polluted vegetation index at the 1.64 μm band was calculated.

[0114] For each cell, perform the following:

[0115] Determine whether the deaerosol-polluted vegetation index of this pixel is within the range of (0, 0.25);

[0116] If yes, determining the pixel is a bright pixel;

[0117] If no, determining the pixel is a dark pixel.

[0118] In some specific implementation manners, in the bright surface estimation module 1004, the quadratic function model is:

[0119]

[0120] wherein R Red is used to represent the red band surface reflectance, R 1.64μm is used to represent the 1.64μm band apparent reflectance, C is used to represent a quadratic fitting coefficient, D is used to represent a linear fitting coefficient, and E is used to represent a constant term fitting coefficient; wherein C, D and E are all correlated with the de-aerosol pollution vegetation index at the 1.64μm band.

[0121] In some specific implementation manners, in the dark surface estimation module 1006, the linear function model is:

[0122] R Red = Slope R 1.64μm + Intercept

[0123] Slope = aNDVI + b

[0124]

[0125] NDVI = (R 0.87μm - R Red ) / (R 0.87μm + R Red )

[0126] wherein R Red is used to represent the red band surface reflectance, R 1.64μm is used to represent the 1.64μm band apparent reflectance, Slope is used to represent a slope, Intercept is used to represent an intercept, NDVI is used to represent the normalized difference vegetation index of the dark pixel, R 0.87μm is used to represent the 0.87μm band apparent reflectance; wherein Slope and Intercept are both correlated with the normalized difference vegetation index, a, b, d, e and f are all used to represent fitting coefficients, and c is used to represent an intercept value when 0.325 < NDVI ≤ 0.65.

[0127] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on a visible-red light channel surface reflectance estimation device. In other embodiments of the present invention, a visible-red light channel surface reflectance estimation device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0128] The information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific details can be found in the description of the method embodiment of the present invention, and will not be repeated here.

[0129] This invention also provides a computing device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a visible red light channel surface reflectance estimation method according to any embodiment of this invention.

[0130] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform a visible red light channel surface reflectance estimation method according to any embodiment of this invention.

[0131] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0132] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0133] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0134] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0135] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.

[0136] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0137] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating surface reflectance in the visible red light channel, characterized in that, include: Acquire the remote sensing image to be processed; Acquire historical remote sensing images, including reflectance of each channel; The historical remote sensing images are calculated and analyzed to obtain the normalized vegetation index for each pixel. The normalized vegetation index is obtained using the following formula: Based on the normalized vegetation index, the de-aerosol-polluted vegetation index at the 1.64 μm band was calculated, and the expression is as follows: NDVI Used to characterize the normalized vegetation index; R 0.87μm Used to characterize the apparent reflectance in the 0.87μm band; R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band; R Red Used to characterize the surface reflectance in the red band to be estimated; AFRI 1.64 Used to characterize aerosol-de-polluted vegetation index based on the 1.64 μm band; For each cell, perform the following: Determine whether the deaerosol-polluted vegetation index of this pixel is within (0, 0.25]. If so, then the pixel is determined to be a bright pixel; the bright surface corresponding to a bright pixel includes sparsely vegetated arid and semi-arid surfaces. If not, then the pixel is determined to be a dark pixel; the dark surface corresponding to a dark pixel includes a surface with dense vegetation. For the bright pixels, a quadratic function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band is constructed, and the surface reflectance in the red band is calculated based on the quadratic function model; the quadratic function model is as follows: in, R Red Used to characterize the surface reflectivity in the red light band. R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band, C Used to characterize the quadratic fitting coefficients D Used to characterize the first-order fitting coefficient. E Used to characterize the fitting coefficients of the constant term; where... C , D and E All were correlated with the deaerosol-polluted vegetation index at the 1.64 μm band; For the dark pixels, a linear function model of the surface reflectance in the red band and the apparent reflectance in the 1.64μm band is constructed, and the surface reflectance in the red band is calculated based on the linear function model; the linear function model is as follows: in, R Red Used to characterize the surface reflectivity in the red light band. R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band, Slope Used to characterize slope, Intercept Used to characterize the intercept. NDVI The normalized vegetation index is used to characterize the dark pixel; among which, Slope and Intercept All are correlated with the normalized difference in vegetation index (NDVI). a, b, d, e, and f are used to characterize the fitting coefficients, and c is used to characterize 0.325 < 0. NDVI The intercept value when ≤0.

65.

2. A visible-red light channel surface reflectance estimation device, characterized in that, To implement the method as described in claim 1, comprising: The acquisition module is used to acquire the remote sensing image to be processed; The bright and dark surface determination module is used to determine the bright and dark pixels in the remote sensing image to be processed based on the vegetation index. The bright surface estimation module is used to construct a quadratic function model of the red band surface reflectance and the 1.64μm band apparent reflectance for the bright pixels, and calculate the red band surface reflectance based on the quadratic function model; the quadratic function model used by the bright surface estimation module is: in, R Red Used to characterize the surface reflectivity in the red light band. R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band, C Used to characterize the quadratic fitting coefficients D Used to characterize the first-order fitting coefficient. E Used to characterize the fitting coefficients of the constant term; where... C , D and E All were correlated with the deaerosol-polluted vegetation index at the 1.64 μm band; The dark surface estimation module is used to construct a linear function model of the red band surface reflectance and the 1.64μm band apparent reflectance for the dark pixels, and calculate the red band surface reflectance based on the linear function model; the linear function model used by the dark surface estimation module is: in, R Red Used to characterize the surface reflectivity in the red light band. R 1.64μm Used to characterize the apparent reflectivity of the 1.64μm band, Slope Used to characterize slope, Intercept Used to characterize the intercept. NDVI The normalized vegetation index used to characterize this dark pixel. R 0.87μm Used to characterize the apparent reflectance in the 0.87μm band; where, Slope and Intercept All are correlated with the normalized difference in vegetation index (NDVI). a, b, d, e, and f are used to characterize the fitting coefficients, and c is used to characterize 0.325 < 0. NDVI The intercept value when ≤0.

65.

3. A computing device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor, when executing the computer program, implements the method as claimed in claim 1.

4. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method of claim 1.

Images