Panchromatic sharpening method and device of multi-spectral image, electronic equipment and storage medium

Abstract

Embodiments of the present application disclose a kind of multispectral image panchromatic sharpening method, device, electronic equipment and storage medium, method includes: obtaining the low frequency component and high frequency component of panchromatic image and low resolution multispectral image;Residual detail image of the panchromatic image relative to multispectral image is obtained, and the low frequency component of residual detail image is obtained;New multispectral image is obtained;New residual detail image is obtained;The standard deviation and correlation coefficient of new multispectral image and new residual detail image are calculated, and injection coefficient is calculated according to the standard deviation of panchromatic image, injection coefficient is optimized, to obtain optimal fused high resolution multispectral image according to the injection coefficient after optimization. The effect of image fusion can be effectively improved in the embodiments of the present application.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a panchromatic sharpening method and device for multispectral images, an electronic device and a storage medium. BACKGROUND

[0002] Multispectral images have low spatial resolution and moderate spectral resolution, while panchromatic images have high spatial resolution but poor spectral resolution. Panchromatic sharpening refers to fusing a multispectral image and a panchromatic image obtained from the same scene to obtain a multispectral image with high spatial resolution. Common panchromatic sharpening methods can be divided into four categories: (1) component replacement-based methods; (2) multi-resolution analysis-based methods; (3) variational optimization-based methods; and (4) deep learning-based methods.

[0003] Among them, the component replacement-based method replaces the components representing spatial information in the multispectral image with the spatial information of the panchromatic image, which loses spectral information and causes distortion. The multi-resolution analysis-based method loses spatial information, resulting in blurring and artifacts on the fused image. The variational optimization-based method has high time complexity. The deep learning-based panchromatic sharpening method lacks training data, has high training difficulty, and lacks consideration of the essential characteristics of remote sensing images during the process. SUMMARY

[0004] Based on the problems in the prior art, the embodiments of the present application provide a panchromatic sharpening method and device for multispectral images, an electronic device and a storage medium.

[0005] In a first aspect, the embodiments of the present application provide a panchromatic sharpening method for multispectral images, comprising:

[0006] obtaining a low-frequency component and a high-frequency component of a panchromatic image and a low-resolution multispectral image of the same scene;

[0007] based on the high-frequency components of the panchromatic image and the multispectral image, obtaining a residual detail image of the panchromatic image relative to the multispectral image, and obtaining a low-frequency component of the residual detail image;

[0008] based on the high-frequency component of the multispectral image and a predetermined weight, obtaining a new multispectral image;

[0009] based on the low-frequency component of the multispectral image and a predetermined weight, obtaining a new residual detail image;

[0010] The standard deviation and correlation coefficient of the new multi-spectral image and the new residual detail image are calculated, an injection coefficient is calculated according to the standard deviation of the panchromatic image, and the injection coefficient is optimized to obtain an optimal fused high-resolution multi-spectral image according to the optimized injection coefficient.

[0011] In some examples, before the low-frequency component and the high-frequency component of the panchromatic image and the low-resolution multi-spectral image of the same scene are acquired, the method further comprises:

[0012] The low-resolution multi-spectral image is up-sampled based on the resolution of the panchromatic image.

[0013] In some examples, the calculation of the standard deviation and correlation coefficient of the new multi-spectral image and the new residual detail image, the calculation of the injection coefficient according to the standard deviation of the panchromatic image, and the optimization of the injection coefficient to obtain the optimal fused high-resolution multi-spectral image according to the optimized injection coefficient comprise:

[0014] The standard deviation and correlation coefficient of the new multi-spectral image and the new residual detail image are calculated, and an initial injection coefficient is calculated according to the standard deviation of the panchromatic image;

[0015] The fused high-resolution multi-spectral image is calculated according to the initial injection coefficient, and the injection coefficient is optimized according to the fused high-resolution multi-spectral image through iterative recursion;

[0016] The optimal fused high-resolution multi-spectral image is obtained according to the optimized injection coefficient.

[0017] In some examples, an adaptive Gram-Schmidt and an adaptive generalized method are used to extract the high-frequency component and the low-frequency component in the fusion of the high-resolution multi-spectral image.

[0018] In a second aspect, embodiments of the present application provide a panchromatic sharpening device for multi-spectral images, comprising:

[0019] An acquisition module is configured to acquire a low-frequency component and a high-frequency component of a panchromatic image and a low-resolution multi-spectral image of the same scene;

[0020] A residual detail image calculation module is configured to obtain a residual detail image of the panchromatic image relative to the multi-spectral image based on the high-frequency components of the panchromatic image and the multi-spectral image, and to obtain a low-frequency component of the residual detail image;

[0021] A multi-spectral image recalculation module is configured to obtain a new multi-spectral image based on the multi-spectral image and a high-frequency component of the multi-spectral image under a predetermined weight;

[0022] a residual detail image re-computing module configured to obtain a new residual detail image based on the residual detail image and a low-frequency component of the multispectral image under a predetermined weight;

[0023] a fusion module configured to calculate a standard deviation and a correlation coefficient of the new multispectral image and the new residual detail image, and obtain an injection coefficient according to a standard deviation of the panchromatic image, and optimize the injection coefficient to obtain an optimal fused high-resolution multispectral image according to the optimized injection coefficient.

[0024] In some examples, the obtaining module is further configured to, before obtaining the panchromatic image and the low-frequency component and the high-frequency component of the low-resolution multispectral image of the same scene, up-sample the low-resolution multispectral image based on a resolution of the panchromatic image.

[0025] In some examples, the fusion module is configured to:

[0026] calculate a standard deviation and a correlation coefficient of the new multispectral image and the new residual detail image, and obtain an initial injection coefficient according to a standard deviation of the panchromatic image;

[0027] obtain a fused high-resolution multispectral image according to the initial injection coefficient, and optimize the injection coefficient by iterative recursion according to the fused high-resolution multispectral image;

[0028] obtain an optimal fused high-resolution multispectral image according to the optimized injection coefficient.

[0029] In some examples, an adaptive Gram-Schmidt and an adaptive generalized method are used to extract the high-frequency component and the low-frequency component when fusing the high-resolution multispectral image.

[0030] In a third aspect, an embodiment of the present application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the panchromatic sharpening method of the multispectral image according to the first aspect when executing the computer program.

[0031] In a fourth aspect, an embodiment of the present application further provides a non-transitory computer readable storage medium having a computer program stored thereon, and the computer program is executable on a processor to implement the panchromatic sharpening method of the multispectral image according to the first aspect.

[0032] From the above technical solutions, the multispectral image panchromatic sharpening method, device, electronic equipment and storage medium provided by the embodiment of the present application utilize the high-frequency components of the multispectral image and the high-frequency components of the residual detail image, control the amount of introduced high-frequency information by calculating the correlation coefficient, and improve the result of image fusion. In addition, iteration recursion is used to improve the injection coefficient, the termination condition is calculated according to the characteristics of the image itself, the optimal injection coefficient is achieved, and the final effect of image fusion is improved. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0034] Figure 1 is a flow chart of the multispectral image panchromatic sharpening method provided by an embodiment of the present application;

[0035] Figure 2 is a structural block diagram of the multispectral image panchromatic sharpening device provided by an embodiment of the present application;

[0036] Figure 3 is a structural schematic diagram of the electronic equipment provided by an embodiment of the present application. DETAILED DESCRIPTION

[0037] The specific embodiments of the present application will be further described below in combination with the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and cannot be used to limit the protection scope of the present application.

[0038] The multispectral image panchromatic sharpening method, device, electronic equipment and storage medium according to the embodiment of the present application are described below in combination with the drawings.

[0039] Figure 1 shows a flow chart of the multispectral image panchromatic sharpening method provided by an embodiment of the present application. As shown in Figure 1 The multispectral image panchromatic sharpening method provided by an embodiment of the present application includes the following contents:

[0040] S101: Obtain the low-frequency components and high-frequency components of the panchromatic image and the low-resolution multispectral image of the same scene.

[0041] Specifically, before obtaining the low-frequency component and the high-frequency component of the low-resolution multispectral image of the same scene, it further comprises: up-sampling the low-resolution multispectral image based on the resolution of the panchromatic image. That is: obtaining the panchromatic image (PAN) and the low-resolution multispectral image (LRMS) of the same scene, and up-sampling the LRMS image so that the PAN image and the LRMS image have the same resolution.

[0042] Obtaining the low-frequency component and the high-frequency component of the panchromatic image and the low-resolution multispectral image of the same scene, that is, calculating the low-frequency component and the high-frequency component of the PAN image and the LRMS image.

[0043] S102: obtaining the residual detail image of the panchromatic image relative to the multispectral image based on the high-frequency component of the panchromatic image and the multispectral image, and obtaining the low-frequency component of the residual detail image. That is, by calculating the high-frequency component of the PAN image and the LRMS, the residual detail image of the PAN image relative to the LRMS image is obtained, and then the low-frequency component of the residual detail image is calculated.

[0044] S103: obtaining a new multispectral image based on the multispectral image and the high-frequency component of the multispectral image under a predetermined weight. That is: by calculating the LRMS image and the high-frequency component of the LRMS image under a certain weight, a new multispectral image is obtained.

[0045] S104: obtaining a new residual detail image based on the residual detail image and the low-frequency component of the multispectral image under a predetermined weight.

[0046] S105: calculating the standard deviation and the correlation coefficient of the new multispectral image and the new residual detail image, and calculating the injection coefficient according to the standard deviation of the panchromatic image, and optimizing the injection coefficient to obtain the optimal fused high-resolution multispectral image according to the optimized injection coefficient.

[0047] Specifically, calculating the standard deviation and the correlation coefficient of the new multispectral image and the new residual detail image, and calculating the injection coefficient according to the standard deviation of the panchromatic image, and optimizing the injection coefficient to obtain the optimal fused high-resolution multispectral image according to the optimized injection coefficient, comprising: calculating the standard deviation and the correlation coefficient of the new multispectral image and the new residual detail image, and calculating the initial injection coefficient according to the standard deviation of the panchromatic image; calculating the fused high-resolution multispectral image according to the initial injection coefficient, and iterating and recursing according to the fused high-resolution multispectral image to optimize the injection coefficient; obtaining the optimal fused high-resolution multispectral image according to the optimized injection coefficient.

[0048] In this example, the adaptive Gram-Schmidt and the adaptive generalized method are used to extract high-frequency components and low-frequency components when fusing high-resolution multispectral images.

[0049] That is, the standard deviation and the correlation coefficient of the new multispectral image and the new residual detail image are calculated, and then the initial injection coefficient is calculated by using the standard deviation of the PAN image; according to the initial injection coefficient, the fused high-resolution multispectral image (HRMS) is calculated, and then the injection coefficient is optimized by iterative recursion according to the HRMS image; and according to the final injection coefficient, the optimal fused high-resolution multispectral image is obtained.

[0050] The present application is based on the linear correlation of LRMS, PAN and HRMS in the local scale.

[0051] F b =M b +G b ·D b (1)

[0052] The present application improves formula (1) by using the linear correlation of LRMS, PAN and HRMS in the local space, extracting the high-frequency component of the LRMS image and the high-frequency component information of the residual image, and then adding them to the original LRMS image to achieve the effect of image fusion.

[0053] F b =M b +μ(M b -f b (M b ))+g b (D b -μf b (D b ))+b b (2)

[0054] Where F b refers to the fused high-resolution multispectral image, M b represents the up-sampled LRMS image, D b represents the calculated residual detail image. (M b -f b (M b )) represents the high-frequency component of the LRMS image, f b (D b ) represents the low-frequency component of the residual detail image. Equation (2) can be simplified as equation (3):

[0055]

[0056] wherein and respectively represent M b +μ(M b -f b (M b )) and (D b -μf b (D b )). represents a new multispectral image enhanced by the high frequency component of the LRMS, represents a new residual detail image, which is mainly the high frequency component of the original residual detail image.

[0057] By calculating the injection coefficient g b and b b , the fused high-resolution multispectral image can be obtained. In the present application, an iterative recursive method is used to improve the injection coefficient, thereby optimizing the result of image fusion.

[0058] In specific applications, the following applies:

[0059] (1) Obtain a LRMS image and a PAN image of the same scene. First, geometrically precisely register the two images, and then use the nearest neighbor method, bilinear interpolation or cubic spline interpolation method to up-sample the LRMS image to obtain the same resolution as the PAN image.

[0060] (2) The adaptive Gram–Schmidt (GSA) and adaptive generalized IHS (GIHSA) methods are used to extract the high frequency component and the low frequency component of the image in the image fusion process.

[0061] (3) The calculation formula of the synthetic intensity component of the multispectral image is equation (4). Wherein ω b and ω0 represent the weight number of the bth band of the image and the offset coefficient of the image, respectively.

[0062]

[0063] Then the residual detail image of the PAN image relative to the LRMS image can be obtained by equation (5):

[0064] D=P-I (5)

[0065] (4) By calculating the correlation coefficient between the low frequency component of the LRMS image and the low frequency component of the PAN image, the variable μ in equation (2) is obtained:

[0066]

[0067] Then, the parameter μ is calculated by the calculated parameter μ, and the new multispectral image is calculated by equation (3) That is, the new multispectral image enhanced by the high frequency component of the LRMS.

[0068] (5) The parameter μ is calculated by the parameter μ obtained in step (4), and the new multispectral image is calculated by equation (4) That is, the new residual detail image.

[0069] (6) The present application uses the minimum-variance unbiased estimator (MVUE) to estimate the unknown variable g b The variance of the fused image obtained in equation (3) is:

[0070]

[0071] Where σ and ρ b represent the standard deviation and correlation coefficient of and respectively. The solution of equation (7) is:

[0072]

[0073] In the first step of initialization, because there is no fused image, we use the variance of the PAN image to replace the variance of the fused image in the above equation, and combine the standard deviation and correlation coefficient of the new multispectral image and the new residual image to obtain the initial injection coefficient.

[0074]

[0075]

[0076] (7) The fused high-resolution multispectral image (HRMS) is calculated according to the initial injection coefficient obtained in step 6 and equation (3). Equation (8) is used as the termination criterion for iterative recursion in the present application:

[0077]

[0078] The value of is calculated in each iteration, and when this value is smaller than the previous value, the recursive model is terminated.

[0079] (8) After the kth iteration recursion, the injection coefficient is optimized as:

[0080]

[0081]

[0082] Finally, the optimal image fusion result is calculated according to the optimized injection coefficient.

[0083] According to the multispectral image panchromatic sharpening method provided by the embodiment of the present application, the high frequency components of the multispectral image and the high frequency components of the residual detail image are used, the number of introduced high frequency information is controlled by calculating the correlation coefficient, and the image fusion result is improved. In addition, the injection coefficient is improved by using iterative recursion, the termination condition is calculated according to the characteristics of the image itself, the optimal injection coefficient is obtained, and the final image fusion effect is improved.

[0084] Figure 2 The structure schematic diagram of the multispectral image panchromatic sharpening device provided by the embodiment of the present application is shown as Figure 2 The multispectral image panchromatic sharpening device provided by the embodiment of the present application comprises an acquisition module 210, a residual detail image calculation module 220, a multispectral image recalculation module 230, a residual detail image recalculation module 240 and a fusion module 250, wherein:

[0085] The acquisition module 210 is used for acquiring the low frequency components and the high frequency components of the panchromatic image and the low resolution multispectral image of the same scene.

[0086] The residual detail image calculation module 220 is used for obtaining the residual detail image of the panchromatic image relative to the multispectral image based on the high frequency components of the panchromatic image and the multispectral image, and obtaining the low frequency components of the residual detail image.

[0087] The multispectral image recalculation module 230 is used for obtaining a new multispectral image based on the high frequency components of the multispectral image under a predetermined weight.

[0088] The residual detail image recalculation module 240 is used for obtaining a new residual detail image based on the low frequency components of the multispectral image under a predetermined weight.

[0089] The fusion module 250 is used for calculating the standard deviation and the correlation coefficient of the new multispectral image and the new residual detail image, calculating the injection coefficient according to the standard deviation of the panchromatic image, optimizing the injection coefficient, and obtaining the optimal fused high resolution multispectral image according to the optimized injection coefficient.

[0090] In an embodiment of the present application, the acquisition module is further used for performing up-sampling on the low resolution multispectral image based on the resolution of the panchromatic image before acquiring the low frequency components and the high frequency components of the panchromatic image and the low resolution multispectral image of the same scene.

[0091] In one embodiment of the present application, the fusion module is used for:

[0092] calculating the standard deviation and correlation coefficient of the new multi-spectral image and the new residual detail image, and calculating an initial injection coefficient according to the standard deviation of the panchromatic image;

[0093] calculating a fused high-resolution multi-spectral image according to the initial injection coefficient, and performing iterative recursion according to the fused high-resolution multi-spectral image to optimize the injection coefficient;

[0094] obtaining an optimal fused high-resolution multi-spectral image according to the optimized injection coefficient.

[0095] In one embodiment of the present application, the adaptive Gram-Schmidt and adaptive generalized methods are used to extract high-frequency components and low-frequency components when fusing high-resolution multi-spectral images.

[0096] The panchromatic sharpening device for multi-spectral images according to the embodiments of the present application uses the high-frequency components of the multi-spectral images and the high-frequency components of the residual detail images, controls the amount of introduced high-frequency information by calculating the correlation coefficient, and improves the results of image fusion. In addition, iterative recursion is used to improve the injection coefficient, the termination condition is calculated according to the characteristics of the image itself, the optimal injection coefficient is obtained, and the final effect of image fusion is improved.

[0097] It should be noted that the specific implementation of the panchromatic sharpening device for multi-spectral images according to the embodiments of the present application is similar to the specific implementation of the panchromatic sharpening method for multi-spectral images according to the embodiments of the present application, and specific details are described in the method section. In order to reduce redundancy, this part will not be described here.

[0098] Based on the same inventive concept, another embodiment of the present application provides an electronic device, as shown in Figure 3 , which specifically includes the following contents: a processor 401, a memory 402, a communication interface 403, and a communication bus 404.

[0099] The processor 401, the memory 402, and the communication interface 403 communicate with each other through the communication bus 404; the communication interface 403 is used to realize information transmission between devices.

[0100] The processor 401 is configured to invoke a computer program in the memory 402, and the processor implements all steps of the panchromatic sharpening method of multi-spectral images when executing the computer program. For example, the processor implements the following steps when executing the computer program: obtaining a low-frequency component and a high-frequency component of a panchromatic image and a low-resolution multi-spectral image of the same scene; obtaining a residual detail image of the panchromatic image relative to the multi-spectral image based on the high-frequency components of the panchromatic image and the multi-spectral image, and obtaining a low-frequency component of the residual detail image; obtaining a new multi-spectral image based on the high-frequency component of the multi-spectral image and a predetermined weight of the multi-spectral image; obtaining a new residual detail image based on the low-frequency component of the multi-spectral image and a predetermined weight of the multi-spectral image; calculating a standard deviation and a correlation coefficient of the new multi-spectral image and the new residual detail image, and obtaining an injection coefficient according to a standard deviation of the panchromatic image, and optimizing the injection coefficient to obtain an optimal fused high-resolution multi-spectral image according to the optimized injection coefficient.

[0101] Based on the same inventive concept, another embodiment of the present application provides a non-transitory computer readable storage medium having a computer program stored thereon, and the computer program implements all steps of the panchromatic sharpening method of multi-spectral images when executed by a processor. For example, the processor implements the following steps when executing the computer program: obtaining a low-frequency component and a high-frequency component of a panchromatic image and a low-resolution multi-spectral image of the same scene; obtaining a residual detail image of the panchromatic image relative to the multi-spectral image based on the high-frequency components of the panchromatic image and the multi-spectral image, and obtaining a low-frequency component of the residual detail image; obtaining a new multi-spectral image based on the high-frequency component of the multi-spectral image and a predetermined weight of the multi-spectral image; obtaining a new residual detail image based on the low-frequency component of the multi-spectral image and a predetermined weight of the multi-spectral image; calculating a standard deviation and a correlation coefficient of the new multi-spectral image and the new residual detail image, and obtaining an injection coefficient according to a standard deviation of the panchromatic image, and optimizing the injection coefficient to obtain an optimal fused high-resolution multi-spectral image according to the optimized injection coefficient.

[0102] In addition, the logic instructions in the memory described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0103] The device embodiments described above are only illustrative, wherein the units illustrated as separate components can or can not be physically separated, and the components illustrated as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment of the present application according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0104] From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions essentially or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the index monitoring method described in each embodiment or some parts of the embodiment.

[0105] In addition, in the present application, such as "first", "second" is only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise specifically limited.

[0106] Moreover, in the subject specification, the term "engagement" simply denotes the relationship between or among multiple entities or operations, and does not necessarily require any such actual relationship or order between or among such entities or operations. Also, the terminology "includes," "has," "holds," "contains" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to those elements, but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0107] Furthermore, in the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the particular feature, structure, material or characteristic being described in connection with the embodiment or example contains in at least one embodiment or example of the present application. The illustrative representation of the above terms in the specification does not necessarily refer to the same embodiment or example. Moreover, the particular feature, structure, material, or characteristic being described can be combined in any suitable manner in one or more embodiments or examples. Furthermore, different embodiments or examples described in the specification can be combined and combined with features of other embodiments or examples, if such combination does not result in a contradiction.

[0108] Finally, it should be noted that the above embodiments are merely used to illustrate the technical solutions of the present application, rather than limiting them; even though the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and such modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A panchromatic sharpening method for multispectral images, characterized in that, include: Acquire the low-frequency and high-frequency components of a panchromatic image and a low-resolution multispectral image of the same scene; Based on the high-frequency components of the panchromatic image and the multispectral image, a residual detail image of the panchromatic image relative to the multispectral image is obtained, and the low-frequency components of the residual detail image are obtained. A new multispectral image is obtained based on the multispectral image and the high-frequency components of the multispectral image under a predetermined weight. A new residual detail image is obtained based on the residual detail image and the low-frequency components of the multispectral image under a predetermined weight. The standard deviation and correlation coefficient of the new multispectral image and the new residual detail image are calculated, and the injection coefficient is calculated based on the standard deviation of the panchromatic image. The injection coefficient is optimized to obtain the optimal fused high-resolution multispectral image based on the optimized injection coefficient.

2. The panchromatic sharpening method for multispectral images according to claim 1, characterized in that, Before acquiring the low-frequency and high-frequency components of the panchromatic image and the low-resolution multispectral image of the same scene, the method further includes: Based on the resolution of the panchromatic image, the low-resolution multispectral image is upsampled.

3. The panchromatic sharpening method for multispectral images according to claim 1, characterized in that, The process of calculating the standard deviation and correlation coefficient of the new multispectral image and the new residual detail image, calculating the injection coefficient based on the standard deviation of the panchromatic image, and optimizing the injection coefficient to obtain the optimal fused high-resolution multispectral image based on the optimized injection coefficient includes: Calculate the standard deviation and correlation coefficient of the new multispectral image and the new residual detail image, and calculate the initial injection coefficient based on the standard deviation of the panchromatic image; Based on the initial injection coefficients, a fused high-resolution multispectral image is calculated, and the injection coefficients are optimized iteratively based on the fused high-resolution multispectral image. Based on the optimized injection coefficients, the optimal fused high-resolution multispectral image is obtained.

4. The panchromatic sharpening method for multispectral images according to any one of claims 1-3, characterized in that, An adaptive Gram-Schmidt and adaptive generalized approach is used to extract high-frequency and low-frequency components when fusing high-resolution multispectral images.

5. A panchromatic sharpening device for multispectral images, characterized in that, include: The acquisition module is used to acquire the low-frequency and high-frequency components of panchromatic images and low-resolution multispectral images of the same scene; The residual detail image calculation module is used to obtain the residual detail image of the panchromatic image relative to the multispectral image based on the high-frequency components of the panchromatic image and the multispectral image, and to obtain the low-frequency components of the residual detail image. A multispectral image recalculation module is used to obtain a new multispectral image based on the multispectral image and the high-frequency components of the multispectral image under a predetermined weight; The residual detail image recalculation module is used to obtain a new residual detail image based on the residual detail image and the low-frequency component of the multispectral image under a predetermined weight. The fusion module is used to calculate the standard deviation and correlation coefficient of the new multispectral image and the new residual detail image, and to calculate the injection coefficient based on the standard deviation of the panchromatic image. The injection coefficient is then optimized to obtain the optimal fused high-resolution multispectral image based on the optimized injection coefficient.

6. The panchromatic sharpening device for multispectral images according to claim 5, characterized in that, The acquisition module is further configured to upsample the low-resolution multispectral image based on the resolution of the panchromatic image before acquiring the low-frequency and high-frequency components of the panchromatic image and the low-resolution multispectral image of the same scene.

7. The panchromatic sharpening device for multispectral images according to claim 5, characterized in that, The fusion module is used for: Calculate the standard deviation and correlation coefficient of the new multispectral image and the new residual detail image, and calculate the initial injection coefficient based on the standard deviation of the panchromatic image; Based on the initial injection coefficients, a fused high-resolution multispectral image is calculated, and the injection coefficients are optimized iteratively based on the fused high-resolution multispectral image. Based on the optimized injection coefficients, the optimal fused high-resolution multispectral image is obtained.

8. The panchromatic sharpening apparatus for multispectral images according to any one of claims 5-7, characterized in that, An adaptive Gram-Schmidt and adaptive generalized approach is used to extract high-frequency and low-frequency components when fusing high-resolution multispectral images.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the panchromatic sharpening method for multispectral images according to any one of claims 1 to 4.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the panchromatic sharpening method for multispectral images according to any one of claims 1 to 4.

Images