Remote sensing image enhancement method and device, equipment and medium

Abstract

The invention discloses a remote sensing image enhancement method and device, equipment and a medium, and is applied to the technical field of remote sensing image processing. Obtaining prior information and spectral features of the degraded image to calculate an atmospheric light component of the remote sensing image; obtaining a gradient distribution map of the remote sensing image according to the degraded image, obtaining gradient channel characteristics of the gradient distribution map, obtaining a target gradient map of the degraded image according to the gradient channel characteristics, and obtaining an initial transmissivity map of the degraded image according to the target gradient map; and optimizing the initial transmissivity graph to obtain an optimized transmissivity graph of the degraded image, and obtaining an enhanced image of the remote sensing image according to the optimized transmissivity graph. According to the method, on the basis of joint solution of the atmospheric light component and the transmissivity, detail distortion and artifacts of the remote sensing image are avoided, and convergence, application feasibility and timeliness in the remote sensing image enhancement process are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of remote sensing image processing, and particularly relates to a remote sensing image enhancement method, device, equipment and medium. BACKGROUND

[0002] In the process of acquiring remote sensing images, the imaging light needs to pass through the atmosphere. However, in the process of passing through the atmosphere, atmospheric scattering and absorption will cause problems such as haze, contrast reduction and color distortion in the remote sensing images, especially under cloudy, hazy or low-illumination conditions. The existing remote sensing image enhancement methods are mostly based on global histogram equalization, Retinex model or dark channel prior strategy, which can improve the contrast and brightness of the remote sensing images to a certain extent, but often ignore the physical mechanism of atmospheric scattering on light transmission, making it difficult to accurately restore the true reflectance and depth information of the remote sensing images, and easily introducing over-sharpening, noise amplification or color shift in the process of dehazing or enhancing the remote sensing images, resulting in distortion of details and generation of artifacts in the remote sensing images, which is difficult to meet the stringent requirements of high-precision remote sensing applications (such as feature classification, target detection and change detection) on the quality and authenticity of remote sensing images. SUMMARY

[0003] Therefore, the present application aims to provide a remote sensing image enhancement method, device, equipment and medium, which jointly solves the atmospheric light component and transmittance map of the remote sensing image, optimizes the transmittance map, and obtains an enhanced image of the remote sensing image according to the atmospheric light component and the optimized transmittance map.

[0004] To achieve the above-mentioned purpose, the technical solution of the present application is as follows: In a first aspect, the present application provides a remote sensing image enhancement method, comprising: S1: obtaining a degraded image of a remote sensing image; S2: obtaining prior information and spectral features of the degraded image, and calculating an atmospheric light component of the remote sensing image according to the prior information and the spectral features; S3: obtaining a gradient distribution map of the remote sensing image according to the degraded image, obtaining gradient channel features of the gradient distribution map, obtaining a target gradient map of the degraded image according to the gradient channel features, and obtaining an initial transmittance map of the degraded image according to the target gradient map; S4: optimizing the initial transmittance map to obtain an optimized transmittance map of the degraded image, and obtaining an enhanced image of the remote sensing image according to the optimized transmittance map.

[0005] In some embodiments of the present application, the atmospheric light component of the remote sensing image is calculated according to the prior information and the spectral features, comprising: S21: Obtain the global mean and dark channel code value of the degraded image, and calculate the compensation parameters of the degraded image based on the global mean and dark channel code value; S22: Obtain the wavelength reference parameters, wavelength calculation parameters, and actual imaging wavelength of the degraded image based on the spectral characteristics. Calculate the atmospheric light component of the remote sensing image based on the preset Gaussian kernel, wavelength reference parameters, wavelength calculation parameters, actual imaging wavelength, atmospheric light component, and degraded image. The Gaussian kernel is used to smooth the scene details of the degraded image.

[0006] In some embodiments of the present invention, obtaining an initial transmittance map of the degraded image based on gradient channel features includes: S31: Determine the incident light wavelength of the degraded image based on the spectral characteristics; S32: Obtain the average gradient value, target gradient matrix, first weight factor, second weight factor, and number of image blocks of the degraded image; and obtain the target gradient map of the degraded image based on the degraded image, the average gradient value of the degraded image, the target gradient matrix, the first weight factor, the second weight factor, and the number of image blocks. S33: Obtain the gradient value of the target gradient map and the gradient value of the atmospheric light component; evaluate the degraded image based on the target gradient map to obtain the gradient value of the degraded image; obtain the initial transmittance map of the degraded image based on the gradient value of the degraded image, the maximum gradient value of the target gradient map, and the gradient value of the atmospheric light component. S34: Obtain the desired ideal image of the degraded image based on the degraded image, the initial transmittance map, and the atmospheric light component.

[0007] In some embodiments of the present invention, optimizing the initial transmittance map to obtain an optimized transmittance map includes: S41: Construct the mask matrix of the degraded image, and obtain the current gradient map of the degraded image based on the mask matrix and the target gradient map; S42; Obtain the regularization parameters and target transmittance map of the degraded image, and construct the first optimization model of the degraded image based on the current gradient map, regularization parameters, and target transmittance map; S43: The first optimization model optimizes the initial transmittance map of the degraded image to obtain the optimized transmittance map of the degraded image.

[0008] In some embodiments of the present invention, the first optimization model optimizes the initial transmittance map of the degraded image, as expressed by the following formula: ; Where d is the current gradient map of the degraded image, For regularization parameters, For target transmittance map, For the initial transmittance map, The minimum value of the L1 norm in the current gradient graph. The first norm regularization parameter of the current gradient graph, The second norm regularization parameter is used for the target transmittance map and the initial transmittance map.

[0009] In some embodiments of the present invention, a first optimization model optimizes the initial transmittance map of the degraded image to obtain an optimized transmittance map of the degraded image, including: S431: Obtain the preset gradient operator, and convert the first optimization model into the second optimization model based on the gradient operator, mask matrix, initial transmittance map and weighted current gradient map. The second optimization model solves the current gradient map to obtain the iterative gradient map. The gradient operator is used to quantize the rate of change of gray values ​​of the degraded image. S432: Optimize the initial transmittance map based on the iterative gradient map to obtain the optimized transmittance vector, and obtain the optimized transmittance map of the degraded image based on the optimized transmittance vector.

[0010] In some embodiments of the present invention, the initial transmittance map is optimized based on the iterative gradient map, including: S4321: Convert the gradient operator into the first sparse operator and the second sparse operator; convert the mask matrix into the mask sparse matrix; convert the vector form of the iterative gradient map into the first gradient matrix and the second gradient matrix; S4322: The initial transmittance map is optimized based on the regularization parameter, the first sparse operator, the second sparse operator, the first gradient matrix, the second gradient matrix, the mask sparse matrix, and the vector form of the initial transmittance map to obtain the optimized transmittance vector.

[0011] In a second aspect, embodiments of the present invention provide a remote sensing image enhancement apparatus, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, the instructions being executed by the at least one control processor to enable the at least one control processor to perform the remote sensing image enhancement method as described in the first aspect above.

[0012] Thirdly, embodiments of the present invention provide an electronic device including the remote sensing image enhancement device as described in the second aspect above.

[0013] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the remote sensing image enhancement method as described in the first aspect above.

[0014] The remote sensing image enhancement method according to embodiments of the present invention has at least the following beneficial effects: By introducing two key unknowns—atmospheric light component and transmittance map—the optical imaging process is decomposed into atmospheric scattering and scene reflection. Based on the gradient channel features of the ideal image, the degraded image of the remote sensing image is optimized and reconstructed, thereby obtaining a transmittance map with enhanced reconstruction details. The transmittance and the detailed features of the remote sensing image are smoothly fused across the entire image range without over-smoothing flat areas of the remote sensing image, thus accurately restoring the true reflectance and depth information of the remote sensing image.

[0015] Based on the joint solution of atmospheric light components and transmittance, the transmittance is smoothed through fast convergent optimization iteration, and the detailed features of remote sensing images are preserved, avoiding the generation of detail distortion and artifacts in remote sensing images. Through physical modeling and efficient iterative optimization, the detail performance and color restoration capability of remote sensing images are effectively improved, while also ensuring the convergence, application feasibility and timeliness of the remote sensing image enhancement process. Attached Figure Description

[0016] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of remote sensing image enhancement methods; Figure 2 This is a flowchart for calculating the atmospheric light component of a remotely sensed image; Figure 3 This is a flowchart for obtaining the initial transmittance map of a degraded image based on gradient channel features; Figure 4 This is a flowchart of optimizing the initial transmittance map to obtain the optimized transmittance map; Figure 5 This is a flowchart for obtaining an optimized transmittance map of a degraded image; Figure 6 This is a flowchart of optimizing the initial transmittance map to obtain the optimized transmittance vector. Figure 7 This is a structural diagram of a remote sensing image enhancement device provided in another embodiment of the present invention; Figure 8 The degraded image provided in the embodiments of the present invention; Figure 9 This is a schematic diagram illustrating the adjustment of the atmospheric light component in a degraded image based on spectral characteristics, provided in an embodiment of the present invention. Figure 10 The gradient distribution map of the degraded image provided in the embodiments of the present invention; Figure 11An initial transmittance map of a degraded image provided in an embodiment of the present invention; Figure 12 An optimized transmittance map of a degraded image provided in an embodiment of the present invention; Figure 13 An enhanced image of a remote sensing image provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and do not constitute a limitation thereof. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, some operations related to the invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in this invention can be combined to form various implementations. Furthermore, the order of the steps or actions in the method description can be changed or adjusted in a manner readily apparent to those skilled in the art. Therefore, the various orders in the specification and drawings are merely for the clear description of a particular embodiment and do not imply a mandatory order, unless otherwise stated that a particular order must be followed.

[0019] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] Reference Figure 1 , Figure 1 This is an overall flowchart of the remote sensing image enhancement method of the present invention, which includes the following steps: S1: Obtain a degraded image of the remote sensing image; S2: Obtain prior information and spectral features of the degraded image, and calculate the atmospheric light component of the remote sensing image based on the prior information and spectral features; S3: Obtain the gradient distribution map of the remote sensing image based on the degraded image, acquire the gradient channel features of the gradient distribution map, obtain the target gradient map of the degraded image based on the gradient channel features, and obtain the initial transmittance map of the degraded image based on the target gradient map. S4: Optimize the initial transmittance map to obtain an optimized transmittance map of the degraded image, and obtain an enhanced image of the remote sensing image based on the optimized transmittance map.

[0023] It should be noted that the atmospheric light component represents the path radiation caused by atmospheric scattering at the pixel level of the remote sensing image. This embodiment obtains the value through evaluation and calculation of a single-channel remote sensing image acquired in a single spectral band.

[0024] It should be noted that prior information represents the content information of the degraded image, or an information matrix directly or indirectly transformed from the content information of the degraded image. The degraded image represents a low-quality remote sensing image generated during the imaging, transmission, and recording process due to imperfections in the imaging system, transmission medium, and equipment. Spectral features represent the spectral band information used by the camera acquiring the remote sensing image. Atmospheric light components represent the atmospheric light intensity of the degraded image.

[0025] It should be noted that this embodiment decomposes the optical imaging process into two parts—atmospheric scattering and scene reflection—by introducing atmospheric light components and a transmittance map, thus reconstructing the enhanced degraded image from a physical perspective. Specifically, the atmospheric light components of the remote sensing image are solved based on the prior information and spectral characteristics of the degraded image. A gradient distribution map of the remote sensing image is obtained from the degraded image, and the gradient channel features of the remote sensing distribution map are acquired. A target gradient map of the degraded image is obtained based on the gradient channel features, and an initial transmittance map of the degraded image is obtained based on the target gradient map to avoid color distortion during the enhancement process of the remote sensing image. In the process of introducing the transmittance map, the gradient channel features of the ideal gradient distribution map are introduced as a constraint on the degraded image to maximize the recovery of blurred edge texture details in the degraded image. The initial transmittance map is optimized to smooth the transmittance of flat areas in the degraded image, avoiding artifacts and artifact phenomena.

[0026] Furthermore, by accurately solving for the atmospheric light component and transmittance map, and through physical modeling and efficient iterative optimization, the detail representation and color restoration capabilities of the enhanced degraded image are effectively improved, while ensuring high convergence and application feasibility during the image enhancement process. Specifically, the efficient iterative optimization method in this embodiment is the ADMM alternating direction iterative method, which calculates the atmospheric light component and transmittance map separately. Each calculation of the atmospheric light component and transmittance map constitutes one iteration. In the efficient iterative optimization process, convergence can be achieved quickly after 3-5 iterations.

[0027] It should be noted that after solving for the atmospheric light component and optimizing the transmittance map, the enhanced image of the remote sensing image is obtained based on the optimized transmittance map and the atmospheric light component, which is expressed by the following first formula: ; in, Enhanced images for remote sensing images, For degraded images, For atmospheric light components, To optimize the transmittance map.

[0028] It should be noted that, referring to Figure 8 It can be seen that the degraded remote sensing image is quite blurry at this point, and the brightness and darkness features are not very obvious. First, the atmospheric light component of the degraded image (such as...) is calculated based on spectral features and prior information. Figure 9 As shown). Simultaneously, its gradient distribution map is calculated based on the degraded image (e.g., ...). Figure 10 As shown), the target gradient map is estimated through its gradient channel characteristics, and then the initial transmittance map is solved (as shown). Figure 11 (As shown). Finally, the initial transmittance map is optimized to obtain a smoother, more detailed optimized transmittance map (as shown). Figure 12 As shown), and then based on the optimized transmittance map, the final enhanced remote sensing image is obtained (e.g., Figure 13 (As shown).

[0029] It should be noted that, Figure 11 and Figure 12 The coordinate system in all of them is the pixel coordinate system. Figure 11 and Figure 12 The X and Y axes in the graph both represent the image pixels of the degraded image. Figure 13 and Figure 8 The Z-axis in the graph represents the transmittance value of the degraded image.

[0030] It should be noted that remote sensing images are images formed by acquiring terrain features, surface objects, etc., through remote sensing technology. Remote sensing images are directly acquired as single-channel grayscale images, lacking rich color. Degraded remote sensing images refer to the image quality degradation that occurs during downloading, transmission, and other processes, leading to image degradation and a decrease in image clarity. After processing by the method of this invention, the final image obtained... Figure 13 The enhanced remote sensing image shown is compared with... Figure 13 and Figure 2 , Figure 1 The clarity and contrast are significantly improved, proving that the method of the present invention is indeed feasible.

[0031] Additionally, in one embodiment, reference is made to Figure 3 ,exist Figure 2 In step S4 of the illustrated embodiment, calculating the atmospheric light component of the remote sensing image based on prior information and spectral features further includes the following steps: S21: Obtain the global mean and dark channel code value of the degraded image, and calculate the compensation parameters of the degraded image based on the global mean and dark channel code value; S22: Obtain the wavelength reference parameters, wavelength calculation parameters, and actual imaging wavelength of the degraded image based on the spectral characteristics. Calculate the atmospheric light component of the remote sensing image based on the preset Gaussian kernel, wavelength reference parameters, wavelength calculation parameters, actual imaging wavelength, atmospheric light component, and degraded image. The Gaussian kernel is used to smooth the scene details of the degraded image.

[0032] It should be noted that the global mean and dark channel code value of the degraded image are obtained, and the compensation parameters of the degraded image are calculated based on the global mean and dark channel code value, using the following second formula: ; in, To compensate parameters, For a degraded image, mean is the global mean of the degraded image. To obtain the dark channel code value for a degraded image.

[0033] It should be noted that the atmospheric light component of the remote sensing image is calculated based on the Gaussian kernel, wavelength calculation parameters, wavelength reference parameters, actual imaging wavelength, degraded image, and compensation parameters, and is obtained through the following third formula: ; in, For atmospheric light components, For wavelength reference parameters, Parameters for wavelength calculation Where is the actual imaging wavelength, and G is the Gaussian kernel. To compensate parameters, This is a convolution operation.

[0034] It should be noted that the exponential function The value reflects that, when the wavelength reference parameter remains constant, the atmospheric scattered radiation component decreases as the wavelength calculation parameter and the actual imaging wavelength increase, which is consistent with Rayleigh scattering theory. In this embodiment, the compensation parameter... The compensation parameter is a constant. Used to compensate for color distortion in degraded images.

[0035] Additionally, in one embodiment, reference is made to Figure 4 ,exist Figure 1 In step S3 of the illustrated embodiment, obtaining the initial transmittance map of the degraded image based on the gradient channel features further includes the following steps: S31: Determine the incident light wavelength of the degraded image based on spectral characteristics; S32: Obtain the average gradient value, target gradient matrix, first weight factor, second weight factor, and number of image patches of the degraded image; and obtain the target gradient map of the degraded image based on the degraded image, the average gradient value, target gradient matrix, first weight factor, second weight factor, and number of image patches. S33: Obtain the gradient value of the target gradient map and the gradient value of the atmospheric light component. Based on the evaluation of the degraded image by the target gradient map, obtain the gradient value of the degraded image. Based on the gradient value of the degraded image, the maximum gradient value of the target gradient map, and the gradient value of the atmospheric light component, obtain the initial transmittance map of the degraded image. S34: Obtain the desired ideal image of the degraded image based on the degraded image, the initial transmittance map, and the atmospheric light component.

[0036] It should be noted that the target gradient map of the degraded image is obtained based on the degraded image, the average gradient value of the degraded image, the target gradient matrix, the first weighting factor, and the second weighting factor, and is expressed by the following fourth formula: ; In this formula, For the target gradient map, Indicates the number of pixels in an image block. For the incident light wavelength of the degraded image, For the target gradient matrix, The absolute value of the average gradient of the degraded image. As the first weighting factor, As the second weighting factor, For degraded images, For the expected gradient features of the degraded image, For the set of neighboring pixels, is a pixel that is located within the set of neighboring pixels.

[0037] It should be noted that, The degraded image is designed to ensure that the necessary edge pixels' texture degradation characteristics are reflected. A first weighting factor balances the weights of the average gradient value and the target gradient matrix, while a second weighting factor balances the weights of the degraded image itself. The incident light wavelength of the degraded image is determined by the wavelength of the spectral band used in the degraded image, typically between 0 and 1.

[0038] It should be noted that the desired ideal image, obtained from the degraded image, the initial transmittance map, and the atmospheric light component, is expressed by the following fifth formula: ; In the formula, For degraded images, To obtain the ideal image, This is the initial transmittance diagram. This refers to the atmospheric light component.

[0039] It should be noted that due to the degraded image Given a quantity, atmospheric light component As obtained through the fourth formula above, the desired ideal image is finally obtained by solving the fifth formula above. The result.

[0040] It should be noted that atmospheric transmittance For reflection component, atmospheric light component Since atmospheric transmittance and atmospheric light components are scattering components, in this embodiment, the optical imaging process of remote sensing images is decomposed into two parts: atmospheric scattering and scene reflection, so as to reconstruct the degraded image with enhanced details from a physical level.

[0041] Secondly, the gradient values ​​of the degraded image, the gradient values ​​of the atmospheric light component, and the maximum gradient value of the target gradient map are obtained. Based on the gradient values ​​of the degraded image, the maximum gradient value of the target gradient map, and the gradient values ​​of the atmospheric light component, the initial transmittance map of the degraded image is obtained, expressed by the following sixth formula: ; in, For the initial transmittance map, For degraded image gradient values, The gradient value of the atmospheric light component. This represents the maximum gradient value of the target gradient map.

[0042] Additionally, in one embodiment, reference is made to Figure 5 ,exist Figure 4 In step S3 of the illustrated embodiment, the initial transmittance map is optimized to obtain an optimized transmittance map, and the following steps are also included: S41: Construct the mask matrix of the degraded image, and obtain the current gradient map of the degraded image based on the mask matrix and the target gradient map; S42; Obtain the regularization parameters and target transmittance map of the degraded image, and construct the first optimization model of the degraded image based on the current gradient map, regularization parameters, and target transmittance map; S43: The first optimization model optimizes the initial transmittance map of the degraded image to obtain the optimized transmittance map of the degraded image.

[0043] It should be noted that the mask matrix for the degraded image is constructed based on the dark channel code values. To preserve the distinction between detailed and flat regions of the degraded image, the mask matrix is ​​defined to ensure that the first optimization model achieves high smoothing in regions with low dark channel code values ​​and preserves transmission map details in regions with high dark channel code values, further enhancing the detailed regions of the degraded image. Specifically, the mask matrix for the degraded image is constructed as shown in the following seventh formula: ; in, For the dark channel code value of remote sensing images, This is the mask matrix for the degraded image.

[0044] It should be noted that by adaptively adjusting the dark channel code value of the degraded image, the weight of each parameter in the first optimization model is different when optimizing the degraded image.

[0045] The current gradient map of the degraded image is obtained based on the mask matrix and the target gradient map, and is expressed by the following eighth formula: ; in, For the mask matrix of the degraded image, For the target gradient map of the degraded image, This is the current gradient map of the degraded image.

[0046] It should be noted that in this embodiment, the current gradient map needs to be weighted to protect the edge and texture information of the degraded image, and reduce the interference of noise and outliers on the degraded image, thereby ensuring the quality of the degraded image during the remote sensing image enhancement process.

[0047] The first optimization model for the degraded image is constructed based on the current gradient map, regularization parameters, and target transmittance map, and is expressed by the following ninth formula: ; Where d is the current gradient map after weighting, For regularization parameters, For target transmittance map, For the initial transmittance map, To find the minimum value of the L1 norm in the current gradient graph after weighting, The first norm regularization parameter of the current gradient graph, The second norm regularization parameter is used for the target transmittance map and the initial transmittance map.

[0048] It should be noted that the first norm regularization parameter is used to promote the sparsity of the current gradient map, while the second norm regularization parameter is used to smooth the target transmittance map and the initial transmittance map.

[0049] Those skilled in the art will understand that, since the first optimization model optimizes the current gradient graph and does not require the introduction of the gradient graph coordinates, in the tenth formula, the above-mentioned ninth formula... of The value is omitted. In this embodiment, the regularization parameter is set to 1.

[0050] Additionally, in one embodiment, reference is made to Figure 6 ,exist Figure 5 In step S37 of the illustrated embodiment, the first optimization model optimizes the initial transmittance map of the degraded image to obtain an optimized transmittance map of the degraded image, and further includes the following steps: S431: Obtain the preset gradient operator, and convert the first optimization model into the second optimization model based on the gradient operator, mask matrix, initial transmittance map and weighted current gradient map. The second optimization model solves the current gradient map to obtain the iterative gradient map. The gradient operator is used to quantize the rate of change of gray values ​​of the degraded image. S432: Optimize the initial transmittance map based on the iterative gradient map to obtain the optimized transmittance vector, and obtain the optimized transmittance map of the degraded image based on the optimized transmittance vector.

[0051] It should be noted that, in solving the current gradient map, i.e., solving the eighth formula mentioned above, this embodiment transforms the eighth formula. Specifically, based on the gradient operator, mask matrix, initial transmittance map, and weighted current gradient map, the first optimization model is converted into a second optimization model, as expressed by the following tenth formula: ; Where d is the current gradient map after weighting, For regularization parameters, For target transmittance map, For the initial transmittance map, For auxiliary parameters, For mask matrix, For gradient operators, For element-wise multiplication, To find the minimum value of the L1 norm in the current gradient graph after weighting, The first norm regularization parameter for the weighted current gradient graph, The second norm regularization parameter is used for the target transmittance map and the current gradient map after weighting.

[0052] To solve the ninth formula, we first solve the weighted current gradient map in the ninth formula to obtain the iterative gradient map, which is represented by the following eleventh formula: ; in, Here, d represents the iterative gradient map, and d represents the current gradient map after weighting. For the number of iterations, For mask matrix, For gradient operators, For target transmittance map, For element-wise multiplication, The first norm regularization parameter of the iterative gradient graph, represents the second norm regularization parameter for the iterative gradient map and the target transmittance map.

[0053] Solving the tenth formula above, we obtain the twelfth formula, which is expressed as follows: ; in, For mask matrix, For gradient operators, For target transmittance map, For auxiliary parameters, This represents the number of iterations.

[0054] It should be noted that in the above embodiments, by establishing a first optimization model for the initial transmittance map, the transmittance of flat areas in the degraded image is smoothed, thereby avoiding artifacts and artificial features in the enhanced remote sensing image. By preserving the initial transmittance in the detailed areas of the degraded image, color distortion in the degraded image is effectively avoided.

[0055] Additionally, in one embodiment, reference is made to Figure 7 ,exist Figure 7 In step S373 of the illustrated embodiment, the initial transmittance map is optimized based on the iterative gradient map to obtain an optimized transmittance vector, including: S4321: Convert the gradient operator into the first sparse operator and the second sparse operator; convert the mask matrix into the mask sparse matrix; convert the vector form of the iterative gradient map into the first gradient matrix and the second gradient matrix; S4322: The initial transmittance map is optimized based on the regularization parameter, the first sparse operator, the second sparse operator, the first gradient matrix, the second gradient matrix, the mask sparse matrix, and the vector form of the initial transmittance map to obtain the optimized transmittance vector.

[0056] It should be noted that after solving the equations eleven, twelve, and thirteen step by step, an iterative gradient map is obtained. This iterative gradient map is then represented as a vector to solve the target transmittance map in vector form, thus optimizing the degraded image and obtaining the optimized transmittance map. For ease of solution, this embodiment transforms matrix convolution into coefficient matrix multiplication, converting the convolution problem into a matrix calculation problem, as expressed by the following thirteenth equation: ; in, For regularization parameters, The vector form of the optimized initial transmittance map, The vector form of the target transmittance map The vector form of the initial transmittance map For auxiliary parameters, For mask matrix, The vector form of the iterative gradient map is the second norm regularization parameter.

[0057] Solving Equation 13 involves transforming the gradient operator into a first sparse operator and a second sparse operator; transforming the mask matrix into a mask sparse matrix; and transforming the vector form of the iterative gradient map into a first gradient matrix and a second gradient matrix. This is expressed by Equation 14 as follows: ; in, For regularization parameters, The vector form of the initial transmittance map For auxiliary parameters, For mask sparse matrix, The first gradient matrix in vector form of the iterative gradient map. The second gradient matrix in vector form of the iterative gradient map. For the first sparse operator, is the second sparse operator, and P is the normalization term.

[0058] like ​ As shown, ​ This is a structural diagram of a remote sensing image enhancement device provided in one embodiment of the present invention. The present invention also provides a remote sensing image enhancement device, comprising: The processor 701 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the remote sensing image enhancement method provided in the above embodiments of the present invention. The memory 702 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store the operating system and other application programs. When the remote sensing image enhancement method provided in the above embodiments of the present invention is implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701. The input / output interface 703 is used to implement information input and output; The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704); The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0059] Memory 702, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory 702 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 702 may optionally include remotely located memories 702 relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated to be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0060] This invention also provides an electronic device, including the remote sensing image enhancement device described above.

[0061] This invention also provides a storage medium, which is a computer-readable storage medium, storing a computer program that, when executed by a processor, implements the above-described remote sensing image enhancement method.

[0062] Those skilled in the art will understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0063] The systems, apparatuses, modules, or units described in one or more of the above embodiments may be implemented by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, a computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0064] It should also be noted that 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 limitation, 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.

[0065] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0066] In summary, the above description is merely a preferred embodiment of this specification and is not intended to limit the scope of protection of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A remote sensing image enhancement method, characterized in that, include: S1: Obtain a degraded image of the remote sensing image; S2: Obtain the prior information and spectral features of the degraded image, and calculate the atmospheric light component of the remote sensing image based on the prior information and the spectral features; S3: Obtain the gradient distribution map of the remote sensing image based on the degraded image, acquire the gradient channel features of the gradient distribution map, obtain the target gradient map of the degraded image based on the gradient channel features, and obtain the initial transmittance map of the degraded image based on the target gradient map. S4: Optimize the initial transmittance map to obtain an optimized transmittance map of the degraded image, and obtain an enhanced image of the remote sensing image based on the optimized transmittance map.

2. The remote sensing image enhancement method according to claim 1, characterized in that, The step of calculating the atmospheric light component of the remote sensing image based on the prior information and the spectral features includes: S21: Obtain the global mean and dark channel code value of the degraded image, and calculate the compensation parameters of the degraded image based on the global mean and the dark channel code value; S22: Obtain the wavelength reference parameters, wavelength calculation parameters, and actual imaging wavelength of the degraded image based on the spectral characteristics. Calculate the atmospheric light component of the remote sensing image based on a preset Gaussian kernel, the wavelength reference parameters, the wavelength calculation parameters, the actual imaging wavelength, and the degraded image. The Gaussian kernel is used to smooth the scene details of the degraded image.

3. The remote sensing image enhancement method according to claim 1, characterized in that, The step of obtaining the initial transmittance map of the degraded image based on the target gradient map includes: S31: Determine the incident light wavelength of the degraded image based on the spectral characteristics; S32: Obtain the average gradient value, target gradient matrix, first weight factor, second weight factor, and number of image blocks of the degraded image; and obtain the target gradient map of the degraded image based on the degraded image, the average gradient value of the degraded image, the target gradient matrix, the first weight factor, the second weight factor, and the number of image blocks. S33: Obtain the gradient value of the target gradient map and the gradient value of the atmospheric light component; evaluate the degraded image based on the target gradient map to obtain the gradient value of the degraded image; obtain the initial transmittance map of the degraded image based on the gradient value of the degraded image, the maximum gradient value of the target gradient map, and the gradient value of the atmospheric light component. S34: Obtain the desired ideal image of the degraded image based on the degraded image, the initial transmittance map, and the atmospheric light component.

4. The remote sensing image enhancement method according to claim 1, characterized in that, The optimization of the initial transmittance map to obtain the optimized transmittance map of the degraded image includes: S41: Construct a mask matrix for the degraded image, and obtain the current gradient map of the degraded image based on the mask matrix and the target gradient map; S42: Obtain the regularization parameters and target transmittance map of the degraded image, and construct a first optimization model of the degraded image based on the current gradient map, the regularization parameters, and the target transmittance map; S43: The first optimization model optimizes the initial transmittance map of the degraded image to obtain an optimized transmittance map of the degraded image.

5. The remote sensing image enhancement method according to claim 4, characterized in that, The first optimization model optimizes the initial transmittance map of the degraded image, as expressed by the following formula: ； Where d is the current gradient map of the degraded image, For regularization parameters, For target transmittance map, For the initial transmittance map, The minimum value of the L1 norm in the current gradient graph. The first norm regularization parameter of the current gradient graph, The second norm regularization parameter is used for the target transmittance map and the initial transmittance map.

6. The remote sensing image enhancement method according to claim 4, characterized in that, The first optimization model optimizes the initial transmittance map of the degraded image to obtain an optimized transmittance map of the degraded image, including: S431: Obtain a preset gradient operator, and convert the first optimization model into a second optimization model based on the gradient operator, the mask matrix, the initial transmittance map, and the weighted current gradient map. The second optimization model iterates and solves the current gradient map to obtain an iterative gradient map. The gradient operator is used to quantify the rate of change of the grayscale value of the degraded image. S432: Optimize the initial transmittance map according to the iterative gradient map to obtain an optimized transmittance vector, and obtain the optimized transmittance map of the degraded image according to the optimized transmittance vector.

7. The remote sensing image enhancement method according to claim 6, characterized in that, The step of optimizing the initial transmittance map based on the iterative gradient map to obtain an optimized transmittance vector includes: S4321: Convert the gradient operator into a first sparse operator and a second sparse operator; convert the mask matrix into a mask sparse matrix; convert the vector form of the iterative gradient map into a first gradient matrix and a second gradient matrix; S4322: The optimized transmittance vector is obtained based on the regularization parameter, the first sparse operator, the second sparse operator, the first gradient matrix, the second gradient matrix, the mask sparse matrix, and the vector form of the initial transmittance map.

8. A remote sensing image enhancement device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform the remote sensing image enhancement method as described in any one of claims 1 to 7.

9. An electronic device, characterized in that, Includes the remote sensing image enhancement device as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the remote sensing image enhancement method as described in any one of claims 1 to 7.

Images