Image parallel color correction method, device and storage medium

Abstract

The present application relates to image processing technology, and provides an image parallel color correction method, device and storage medium. The method extracts matching colors of matching points or matching areas between multiple images to be corrected, and presets an auxiliary parameter to express the same target color corresponding to the matching colors of different images after correction, so that the colors of different images can be consistent after correction; the auxiliary parameter is paired with the corresponding matching color to obtain a color correspondence corresponding only to each image to be corrected, providing a parallel basis for parallel correction of multiple images; based on the color correspondence of multiple images, the multiple images jointly construct a parallel energy equation, and the energy equation is solved in parallel to obtain color mapping function parameters; the color mapping function parameters are substituted into the color mapping function to obtain multiple images with consistent colors, thereby improving the color correction efficiency of multiple images.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an image parallel color correction method, device and computer-readable storage medium. Background Art

[0002] When the image acquisition system obtains digital images, there are color differences between the multiple images collected due to factors such as lighting and equipment. In order to ensure the geometric accuracy and visual effects of common visual tasks (such as image stitching and 3D reconstruction), it is generally necessary to perform color consistency correction on the multiple images collected to obtain multiple images with consistent colors.

[0003] Existing methods for color consistency correction of multiple images are mainly divided into two categories: 1) Path propagation method: According to the adjacency relationship, the color correction between two adjacent images is continuously performed, and the color of the reference image is gradually transferred to other images. This type of method has cumulative errors and may fall into local optimality, and the calculation time increases with the increase of the number of images. 2) Global optimization method: Multiple images construct a total energy equation, minimize the energy, and obtain the color mapping function parameters corresponding to each image. This type of method can obtain the global optimal solution, but the calculation time and memory consumption increase with the increase of the number of images. However, both of the above two methods are difficult to parallelize, and the problems of calculation speed and memory consumption are inevitable, and the processing capacity of large-scale images is limited. At present, color consistency correction for large-scale images is still an unresolved problem. It is urgent to propose a parallel global color consistency correction method that takes into account global optimality, calculation speed and memory consumption, and improves the processing capacity of large-scale images. Summary of the invention

[0004] The main purpose of the present invention is to provide an image parallel color correction method, device and storage medium, aiming to solve the technical problem that the existing global color consistency correction method is difficult to be parallel.

[0005] To achieve the above-mentioned purpose, the present invention provides an image parallel color correction method, which includes: obtaining at least two images to be corrected, presetting a color mapping function for each image, and extracting matching colors corresponding to matching points or matching areas between the images to be corrected; presetting auxiliary parameters as the same target color corresponding to at least two of the matching colors after correction, and pairing the auxiliary parameters with the corresponding matching colors one by one to obtain a color correspondence relationship; according to the images to be corrected, the color correspondence relationship, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a parallel energy equation; alternately and parallelly optimize the parallel energy equation to obtain color mapping function parameters and the auxiliary parameters; correct the colors of the images to be corrected according to the color mapping function to obtain multiple images with consistent colors.

[0006] In addition, to achieve the above-mentioned purpose, the present invention also provides an image parallel color correction device, which includes a processor, a memory, and an image parallel color correction program stored in the memory and executable by the processor, wherein when the image parallel color correction program is executed by the processor, the steps of the image parallel color correction method as described above are implemented.

[0007] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, on which an image parallel color correction program is stored, wherein when the image parallel color correction program is executed by a processor, the steps of the image parallel color correction method as described above are implemented.

[0008] The present invention provides an image parallel color correction method, which obtains at least two images to be corrected, presets a color mapping function for each image, and extracts matching colors corresponding to matching points or matching areas between the images to be corrected; presets auxiliary parameters as the same target color corresponding to at least two matching colors after correction, and pairs the auxiliary parameters with the corresponding matching colors one by one to obtain a color correspondence relationship; according to the images to be corrected, the color correspondence relationship, a corresponding color consistency criterion and an image quality lossless criterion, a plurality of images jointly construct a parallel energy equation; alternately and parallelly optimizes the parallel energy equation to obtain color mapping function parameters and the auxiliary parameters; corrects the colors of the images to be corrected according to the color mapping function to obtain a plurality of images with consistent colors. Through the above method, matching colors of matching points or matching areas are extracted between multiple images to be corrected, and an auxiliary parameter is preset to express the same target color corresponding to the matching colors of different images after correction, so that the colors of different images tend to be consistent after correction; and the auxiliary parameters are paired with the corresponding matching colors in different images, so as to obtain a color correspondence relationship corresponding only to each image to be corrected, providing a parallel basis for parallel correction of multiple images; according to the color correspondence relationship, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a global optimization total energy equation including a color consistency term, a quality term and a regularization term, and the color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation; the parallelizable energy equation is used to alternately optimize the unknown auxiliary parameters and color mapping function parameters, and the auxiliary parameters and color mapping function parameters corresponding to each image to be corrected are solved, and the solved color mapping function parameters are substituted into the color mapping function to perform parallel color correction on each image to be corrected, thereby improving the color correction efficiency of multiple images and solving the technical problem that the existing global color consistency correction method is difficult to be parallelized. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 Schematic diagram of the hardware structure of the image parallel color correction device involved in the embodiment of the present invention;

[0010] Figure 2 FIG. 4 is a flow chart of a first embodiment of a method for parallel color correction of images according to the present invention.

[0011] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0012] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0013] The image parallel color correction method involved in the embodiment of the present invention is mainly applied to an image parallel color correction device, which may be a device with display and processing functions such as a PC, a portable computer, or a mobile terminal.

[0014] Reference Figure 1 , Figure 1 The hardware structure diagram of the image parallel color correction device involved in the embodiment of the present invention is shown in FIG. In the embodiment of the present invention, the image parallel color correction device may include a processor 1001 (e.g., a CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard); the network interface 1004 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface); the memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory, and the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0015] Those skilled in the art will understand that Figure 1 The hardware structure shown in the figure does not constitute a limitation on the image parallel color correction device, and may include more or less components than those shown in the figure, or combine certain components, or arrange the components differently.

[0016] Continue to refer to Figure 1 , Figure 1 The memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, and an image parallel color correction program.

[0017] exist Figure 1In the embodiment, the network communication module is mainly used to connect to the server and perform data communication with the server; and the processor 1001 can call the image parallel color correction program stored in the memory 1005 and execute the image parallel color correction method provided by the embodiment of the present invention.

[0018] An embodiment of the present invention provides an image parallel color correction method.

[0019] Reference Figure 2 , Figure 2 FIG. 4 is a flow chart of a first embodiment of a method for parallel color correction of images according to the present invention.

[0020] In this embodiment, the image parallel color correction method includes the following steps:

[0021] Step S10, obtaining at least two images to be corrected, presetting a color mapping function for each image, and extracting matching colors corresponding to matching points or matching areas between the images to be corrected;

[0022] In this embodiment, a set of images to be corrected with color differences is obtained. Wherein, N represents the number of images, which is a positive integer, and N ≥ 2, that is, there are at least two images to be corrected; I i represents the i-th image to be corrected.

[0023] In this embodiment, N color mapping functions f1, f2, ..., f are preset for N images to be corrected. N . Treat the rectified image The color mapping function is Expression Image I i The mathematical transformation between the colors before and after correction, For image I i The color mapping function parameters, M is a positive integer, the color mapping function parameter value is unknown, and is subsequently obtained by minimizing the parallelizable energy equation. Thus, for image I i For any pixel color x, the corrected color can be expressed as:

[0024] x′=f i (x) (1)

[0025] Preset the three-channel linear color mapping function, and then get the color mapping function expression:

[0026]

[0027] Where x = [BGR] T , represents the three-channel color value (blue, green, red); v k (k=1,2,…,12) is the image I to be found i The corresponding color mapping function parameters.

[0028] It can be understood that the image parallel color correction method of the present application is very flexible and supports the preset of any color mapping function. In addition to the above-mentioned three-channel linear color mapping function, the Gamma color mapping function, spline color mapping function, etc. can also be preset to obtain a more complex and flexible expression and obtain higher color correction accuracy.

[0029] In this embodiment, matching point pairs or matching areas of N images to be corrected can be obtained by feature point extraction and matching (such as SIFT feature extraction and matching), image geometric correction, and the like.

[0030] Furthermore, the matching color can be obtained by directly extracting the pixel color of the matching point, extracting the color statistics of the matching area (such as the colors of the same percentage of the regional color cumulative histogram), and the like.

[0031] Step S20, presetting auxiliary parameters as the same target color corresponding to at least two of the matching colors after correction, and pairing the auxiliary parameters with the corresponding matching colors one by one to obtain a color correspondence relationship;

[0032] In this embodiment, the same target color that multiple matching colors approach after correction is used as an auxiliary parameter. The auxiliary parameter is an unknown parameter and is subsequently obtained by minimizing the parallel energy equation. After the matching color in the image to be corrected is corrected by a preset mapping function, the corrected matching color is paired with the corresponding auxiliary parameter to construct a color correspondence. For example, assuming c i , c j , c k is image I i , I j , I k After correction, all three colors are close to the target color c, so c and f i (c i ), c and f j (c j ), c and f j (c k ), forming three color correspondences.

[0033] In this embodiment, in order to support parallelism, the present invention divides all auxiliary parameters into N groups according to the gradient maximization principle. The i-th group of auxiliary parameters is in image I i The corresponding color image gradient on is the largest. For example, assuming the matching color c i , c j , c k The gradient g on the corresponding image i , g j , g k Satisfy the relationship g i >j > k , then c is divided into the i-th group, corresponding to image I i .

[0034] Furthermore, the color correspondences are divided into N groups, so that a single group of color correspondences is only related to the unknown parameters corresponding to a single image, thereby supporting the subsequent construction of parallel energy equations.

[0035] In a specific embodiment, if the auxiliary parameters are regarded as known numbers, each color correspondence is only related to the unknown color mapping function parameters of a single image, so all color correspondences can be divided into N groups, each group is only related to the unknown auxiliary parameters corresponding to a single image. i (c i ) is only related to the color of image I i The unknown parameters Related, divided into group i, c and f j (c j ) is only related to the color of image I j The unknown parameters Related, divided into group j, c and f j (c k ) is only related to the color of image I k The unknown parameters Related, divided into group k.

[0036] Alternatively, if the color mapping function parameters are considered as known numbers, each set of color correspondences is only related to a single unknown auxiliary parameter. i (c i ), c and f j (c j ), c and f j (c k ) are only related to c. According to the above auxiliary parameter grouping, all color correspondences can be further divided into N groups, each of which is only related to the unknown auxiliary parameters corresponding to a single image. For example, if c is grouped as i, corresponding to image I i , then c and f i (c i ), c and f j (c j ), c and f j (c k ) are evenly divided into the i-th group, corresponding to image I i .

[0037] In this embodiment, compared with the color correspondence construction method in which matching colors between images form a correspondence relationship between each other in the existing method, in the color correspondence constructed by the present application, a set of color correspondence relationships is only related to the relevant unknown parameters corresponding to a single image and is not affected by other images; thereby, the parallel energy equation constructed based on the color correspondence relationship can be decomposed into N independent parts, supporting parallel solution of color mapping function parameters, and the number of parallelism is equal to the number of images N involved in constructing the parallel energy equation.

[0038] Step S30, constructing a parallel energy equation for multiple images based on the image to be corrected, the color correspondence, the corresponding color consistency criterion, and the image quality lossless criterion;

[0039] Furthermore, the step S30 specifically includes:

[0040] According to the color correspondence, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a global optimization total energy equation including a color consistency term, a quality term and a regularization term. The color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation.

[0041] In this embodiment, the energy equations that can be run in parallel include: the color consistency term E c , which expresses the color difference of the corresponding colors after correction; the quality term E q , expressing the loss of image quality after correction; regular term E r , which expresses the degree of deviation of the corrected image color relative to the original image color.

[0042] Furthermore, the color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, which specifically includes:

[0043] Accumulating the color differences corresponding to the colors related to the image to be corrected, and obtaining the color consistency loss related to the color mapping function and the auxiliary parameters as a sub-energy equation related to the single image to be corrected;

[0044] The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as the parallelized color consistency term of the parallelizable energy equation.

[0045] In this embodiment, the color consistency E c The term is defined as the cumulative color difference corresponding to all colors. The smaller the value, the more consistent the colors of multiple images.

[0046] In this embodiment, in order to make the energy equation parallelizable, the color consistency term E cIt can be written as the sum of N independent sub-energy equations, each of which is only related to the unknown parameters corresponding to a single image:

[0047]

[0048] in, represents the color difference set corresponding to the i-th group of colors, ρ represents the loss function, and the L2 loss function ρ(r) = r is usually used. 2 If the matching relationship is inaccurate, resulting in unreliable color correspondence and many gross errors, the L1 loss function or Huber loss function can also be used.

[0049] Furthermore, the color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, which specifically also includes:

[0050] The sum of the image quality loss related to the auxiliary parameters and the image quality loss related to the color mapping function corresponding to the single image to be corrected is used as the sub-energy equation related to the single image to be corrected;

[0051] The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as the parallelization quality term of the parallelizable energy equation.

[0052] In this embodiment, the quality term E q It is defined as the sum of the image quality loss related to the auxiliary parameters and the image quality loss related to the color mapping function parameters. It is used to ensure that the quality of the corrected image is not degraded and to avoid achieving color consistency among multiple images at the expense of image quality.

[0053] In this embodiment, in order to make the energy equation parallelizable, E q It can be written as the sum of N independent sub-energy equations, each of which is only related to the unknown parameters corresponding to a single image:

[0054]

[0055] in, represents the quality loss corresponding to the i-th group of auxiliary parameters, Represents image I i Color mapping function f i The quality loss corresponding to the parameters.

[0056] Further, accumulating the difference between the gradients of the auxiliary parameters corresponding to the image to be corrected and the gradients of the matching colors on the image to be corrected, to obtain the image quality loss related to the auxiliary parameters;

[0057] Specifically, It is defined as the cumulative gradient between the i-th group of auxiliary parameters and their gradients in image I i Difference in gradients of matching colors on :

[0058]

[0059] in, is the set of auxiliary parameters of the i-th group, I i (c1) represents the auxiliary parameter c1 in image I i Matching colors on I i (c2) represents the auxiliary parameter c2 in image I i Matching colors on.

[0060] Further, accumulating the difference in image color gradient before and after correction of the image to be corrected to obtain the image quality loss related to the color mapping function;

[0061] Specifically, Represents image I i Color mapping function f i The quality loss corresponding to the parameter is defined as the difference between the cumulative image gradient before and after correction:

[0062]

[0063] in, For image I i The set of sampled pixels, if sampled pixel by pixel, represents all pixels of the whole image; Represents pixel p in image I i The four neighborhoods on .

[0064] It can be understood that the image parallel color correction method of the present application is very flexible and supports various ways of defining You can also customize other more complex To better ensure image quality.

[0065] Furthermore, the color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, which specifically also includes:

[0066] Accumulating the color difference of the image to be corrected before and after correction to obtain a regularization term corresponding to the image to be corrected as a sub-energy equation related to the image to be corrected;

[0067] The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as a parallelized regularization term of the parallelizable energy equation.

[0068] In this embodiment, the regularization term E rIt is defined as the cumulative color difference between the corrected image and the original image. It is encouraged that the corrected image be close to the original image in color to ensure the validity and uniqueness of the solution.

[0069] In this embodiment, in order to make the energy equation parallelizable, E r Written as the sum of N independent sub-energy equations, each of which is only related to the color mapping function parameters corresponding to a single image:

[0070]

[0071] in, Represents a set of sampled colors, uniformly sampled from the color range [0,255].

[0072] Therefore, according to the criteria of corresponding color consistency and image quality loss after correction, the global color correction of multiple images to be corrected jointly constructs a parallel energy equation:

[0073] E=λ c E c +λ q E q +λ r E r (8)

[0074] Among them, λ c ,λ q and λ r To control the color consistency term E c 、Quality Item E q And the regularization term E r The coefficient of the balance of the three is a constant given in advance, and λ c ,λ q and λ r All are positive numbers.

[0075] Step S40, alternately and in parallel optimizing the parallelizable energy equations to obtain color mapping function parameters and the auxiliary parameters;

[0076] In this embodiment, image color correction requires obtaining the color mapping function parameters corresponding to each image to be corrected. However, in the parallel energy equation, the color mapping function parameters and auxiliary parameters are unknown variables. Therefore, through alternating optimization, the unknown variables in the parallel energy equation are jointly optimized and solved to obtain the color mapping function parameters and auxiliary parameters corresponding to each image to be corrected.

[0077] Furthermore, the step S40 specifically includes:

[0078] Fixing the auxiliary parameters, minimizing the parallelizable energy equation, and optimizing the color mapping function parameters in parallel;

[0079] Fixing the color mapping function parameters, minimizing the parallelizable energy equation, and optimizing the auxiliary parameters in parallel;

[0080] The above parallel optimization is performed alternately to obtain the color mapping function parameters and the auxiliary parameters corresponding to the plurality of images to be corrected.

[0081] In this embodiment, an alternating optimization method is used to solve the auxiliary parameters and the color mapping function parameters.

[0082] The first step is to initialize the color mapping function parameters.

[0083] The second step is to fix the parameters of the color mapping function and optimize the auxiliary parameters. The total energy equation removes the items that are not related to the auxiliary parameters, and the energy equation to be optimized is:

[0084]

[0085] It can be understood that since the above equation (9) is the sum of N mutually independent sub-energy equations, each sub-energy equation is only related to the auxiliary parameters corresponding to a single image, the sub-energy equations can be solved in parallel to obtain optimized auxiliary parameters.

[0086] The third step is to fix the auxiliary parameters and optimize the color mapping function parameters. The total energy equation removes the items that are not related to the auxiliary parameters, and the energy equation to be optimized is:

[0087]

[0088] It can be understood that since the above equation (10) is the sum of N independent sub-energy equations, each sub-energy equation is only related to the color mapping function parameters corresponding to a single image, the sub-energy equations can be solved in parallel to obtain the optimized color mapping function parameters.

[0089] Furthermore, the second and third steps are performed alternately until convergence or the maximum number of iterations defined by the user is reached, and the auxiliary parameters and color mapping function parameters of the optimal solution are obtained. Usually, convergence can be achieved within 5 iterations.

[0090] Step S50, correcting the color of the image to be corrected according to the color mapping function to obtain a plurality of images with consistent colors.

[0091] In this embodiment, since a color mapping function is preset for each image to be corrected, but the color mapping function parameters are unknown quantities, when the parameters of the preset color mapping function of each image to be corrected and the corresponding auxiliary parameters are obtained by solving the parallel energy equation, the color mapping function and the solved color mapping function parameters can be used to transform the color of the image to be corrected, complete the correction of the image to be corrected, and finally obtain multiple images with consistent colors.

[0092] In this embodiment, the optimized color mapping function parameters are substituted into the color mapping function to correct the image. Using the color mapping function f i Corrected Image I i .

[0093] In addition, an embodiment of the present invention further provides a computer-readable storage medium.

[0094] The computer-readable storage medium of the present invention stores an image parallel color correction program, wherein when the image parallel color correction program is executed by a processor, the steps of the above-mentioned image parallel color correction method are implemented.

[0095] The method implemented when the image parallel color correction program is executed can refer to the various embodiments of the image parallel color correction method of the present invention, and will not be described in detail here.

[0096] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0097] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0098] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0100] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for parallel color correction of images, characterized in that: The method comprises the following steps: Acquire at least two images to be corrected, preset a color mapping function for each image, and extract matching colors corresponding to matching points or matching areas between the images to be corrected; Preset auxiliary parameters as the same target color corresponding to at least two of the matching colors after correction, and pair the auxiliary parameters with the corresponding matching colors one by one to obtain a color correspondence relationship; According to the image to be corrected, the color correspondence, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a parallel energy equation; Alternately and in parallel optimize the parallelizable energy equations to obtain color mapping function parameters and the auxiliary parameters; Correcting the color of the image to be corrected according to the color mapping function to obtain a plurality of images with consistent colors; Wherein, according to the image to be corrected, the color correspondence, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a parallel energy equation, including: According to the color correspondence, the corresponding color consistency criterion and the image quality lossless criterion, multiple images jointly construct a global optimization total energy equation including a color consistency term, a quality term and a regularization term, wherein the color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation; The color consistency term, the quality term, and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, including: Accumulating the color differences corresponding to the colors related to the image to be corrected, and obtaining the color consistency loss related to the color mapping function and the auxiliary parameters as a sub-energy equation related to the single image to be corrected; The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as a parallelized color consistency term of the parallelizable energy equation; The color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, further comprising: The sum of the image quality loss related to the auxiliary parameters and the image quality loss related to the color mapping function corresponding to the single image to be corrected is used as the sub-energy equation related to the single image to be corrected; The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as a parallelization quality term of the parallelizable energy equation; The color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, further comprising: Accumulating the color difference of the image to be corrected before and after correction to obtain a regularization term corresponding to the image to be corrected as a sub-energy equation related to the image to be corrected; The sum of the mutually independent sub-energy equations corresponding to the multiple images to be corrected is used as a parallelized regularization term of the parallelizable energy equation.

2. The image parallel color correction method according to claim 1, characterized in that: The color consistency term, the quality term and the regularization term are all constructed in a parallelizable form to obtain the parallelizable energy equation, further comprising: Accumulating the difference between the gradients of the auxiliary parameters corresponding to the image to be corrected and the gradients of the matching colors on the image to be corrected, to obtain the image quality loss related to the auxiliary parameters; The image quality loss related to the color mapping function is obtained by accumulating the difference in image color gradient before and after the image to be corrected.

3. The image parallel color correction method according to claim 1 or 2, characterized in that: Alternating and parallel optimizing the parallelizable energy equation to obtain color mapping function parameters and the auxiliary parameters includes: Fixing the auxiliary parameters, minimizing the parallelizable energy equation, and optimizing the color mapping function parameters in parallel; Fixing the color mapping function parameters, minimizing the parallelizable energy equation, and optimizing the auxiliary parameters in parallel; The above parallel optimization is performed alternately to obtain the color mapping function parameters and the auxiliary parameters corresponding to the plurality of images to be corrected.

4. An image parallel color correction device, characterized in that: The image parallel color correction device includes a processor, a memory, and an image parallel color correction program stored in the memory and executable by the processor, wherein when the image parallel color correction program is executed by the processor, the steps of the image parallel color correction method according to any one of claims 1 to 3 are implemented.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an image parallel color correction program, wherein when the image parallel color correction program is executed by a processor, the steps of the image parallel color correction method according to any one of claims 1 to 3 are implemented.

Images