Image color balance method, system and device based on multiple groups of fisheye lenses and medium

Abstract

The invention discloses an image color balance method, system and device based on multiple groups of fisheye lenses and a medium, and the method comprises the steps: converting RGB color data in an RGB image into LAB color space data, and obtaining an image represented by an LAB color space; performing region division on the image represented by the LAB color space, and clustering effective pixel points in each region according to the A value and the B value in the LAB color space data to obtain a clustering result; and based on the clustering result, calculating the error between the average value of the L values of the pixel point set of the target image in the LAB color space and the average value of the L values of the pixel point set of the to-be-balanced image in the LAB color space, and adjusting the exposure value of the image according to the error. According to the method, the RGB image is converted into the LAB color space, and the image in the LAB color space is subjected to region division and color clustering by using the clustering algorithm, so that the color difference between different lenses is accurately identified and quantized, the exposure value is accurately adjusted, and the balance of image colors is realized.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an image color balancing method, system, device and medium based on multiple groups of fisheye lenses. Background Art

[0002] In the field of digital image processing, especially in applications involving multi-camera systems (such as panoramic cameras and multi-view surveillance systems), it is crucial to ensure that images captured by different lenses are consistent in color. Fisheye lenses are commonly used components in such systems due to their ultra-wide-angle field of view, but their unique imaging characteristics and individual differences between different lenses often lead to differences in the color of the captured images, including uneven brightness and hue shift. To improve this situation, automatic white balance and color correction software are widely used in image processing processes. These software are designed to automatically adjust the color temperature and color balance of images through algorithms to achieve visual consistency and naturalness. However, when dealing with the color balance of multiple fisheye lenses, the existing automatic white balance and color correction software often uses different technical routes and parameter settings because the automatic white balance and color correction algorithms of the same software manufacturer or different versions often use different technical routes and parameter settings, resulting in different correction results even when processing the same set of images, leading to consistency issues; secondly, for scenes with complex lighting conditions or requiring fine color adjustments, the automatic correction algorithm may not be able to accurately capture and correct all color differences; in addition, most of the existing automatic white balance and color correction software is designed for general images, lacking an in-depth understanding of the characteristics of multiple fisheye lenses and targeted optimization. The above problems need to be solved. Summary of the invention

[0003] The main purpose of this application is to overcome the shortcomings and deficiencies of the prior art and to provide an image color balancing method, system, device and medium based on multiple groups of fisheye lenses. By converting RGB images into LAB color space and using a clustering algorithm to perform region division and color clustering on the image in the LAB color space, the color differences between different lenses can be accurately identified and quantified, and the exposure value can be accurately adjusted, thereby achieving image color balance.

[0004] In order to achieve the above objectives, this application adopts the following technical solutions:

[0005] In a first aspect, the present application provides an image color balancing method based on multiple groups of fisheye lenses, comprising the following steps:

[0006] Acquire an RGB image, convert RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space;

[0007] Divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result;

[0008] Based on the clustering result, the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space is calculated, and the exposure value of the image is adjusted according to the error.

[0009] As a preferred technical solution, the step of converting the RGB color data in the RGB image into LAB color space data to obtain an image represented in the LAB color space includes:

[0010] Normalizing the RGB color data to obtain a normalized RGB value;

[0011] Converting the normalized RGB values ​​into linear RGB values;

[0012] Convert the linear RGB value to a value in the XYZ color space;

[0013] The value of the XYZ color space is normalized to the XYZ value of the reference white point, and the LAB color space data is calculated according to the normalized XYZ value of the reference white point.

[0014] As a preferred technical solution, the LAB color space data includes L value, A value and B value;

[0015] The L value refers to the brightness of the image;

[0016] The A value refers to the value of the red and green components, indicating a change from green to red;

[0017] The B value refers to the value of the yellow-blue component, indicating the change from blue to yellow.

[0018] As a preferred technical solution, the region division of the image represented by the LAB color space is to divide the image represented by the LAB color space into preset partial regions; wherein each region partially occupies 90° of the space.

[0019] As a preferred technical solution, clustering the valid pixels in each area according to the A value and the B value in the LAB color space data to obtain a clustering result includes:

[0020] Determine the minimum density number and ε radius;

[0021] The A value and B value of the pixel point in the image are regarded as points in the data set, and all the points in the data set are marked as unvisited;

[0022] Randomly select an unvisited point from the data set and mark the selected point as visited;

[0023] Find all neighboring points of the selected point within the ε radius;

[0024] If the number of the neighborhood points is greater than or equal to the minimum density number, the selected point is regarded as a core point, a new cluster is created, and the neighborhood points are added to the new cluster;

[0025] Determine whether each neighboring point of the core point is a core point. If it is a core point, add the neighboring point to the current cluster and mark it as visited;

[0026] If the number of the neighborhood points is less than the minimum density number, marking the selected point as a noise point;

[0027] Repeat the above steps for all unvisited points until all points have been processed.

[0028] As a preferred technical solution, the calculation of the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space based on the clustering result includes:

[0029] Loop through all clustering results, and calculate the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space of each clustering result set;

[0030] According to the average L value of the pixel point set of the target image of each clustering result set in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space, calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space;

[0031] The error calculation formula between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space is:

[0032] T = T + (meanL1-meanL2) * (the number of cluster points in each clustering result / the number of all pixels);

[0033] Among them, meanL1 is the average L value of the pixel point set of the target image in the LAB color space; meanL2 is the average L value of the pixel point set of the image to be balanced in the LAB color space; the initial value of T is 0.

[0034] As a preferred technical solution, the step of adjusting the exposure value of the image according to the error includes:

[0035] According to the calculated error, the exposure value of the image to be balanced is adjusted until the change in the error is less than a preset threshold.

[0036] In a second aspect, the present application provides an image color balancing system based on multiple groups of fisheye lenses, which is applied to the image color balancing method based on multiple groups of fisheye lenses, and includes a conversion module, a clustering module, and an adjustment module;

[0037] The conversion module is used to obtain an RGB image, convert the RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space;

[0038] The clustering module is used to divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result;

[0039] The adjustment module is used to calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space based on the clustering result, and adjust the exposure value of the image according to the error.

[0040] In a third aspect, the present application provides an electronic device, the electronic device comprising:

[0041] at least one processor; and,

[0042] a memory communicatively connected to the at least one processor; wherein,

[0043] The memory stores computer program instructions that can be executed by the at least one processor, and the computer program instructions are executed by the at least one processor to enable the at least one processor to perform the image color balancing method based on multiple groups of fisheye lenses.

[0044] In a fourth aspect, the present application provides a computer-readable storage medium storing a program, which, when executed by a processor, implements the image color balancing method based on multiple groups of fisheye lenses.

[0045] In summary, compared with the prior art, the effective effects brought about by the technical solution provided by this application include at least:

[0046] This application proposes an image color balancing method based on multiple groups of fisheye lenses. This application uses LAB color space for color analysis, and refines the image into multiple local areas through regional division, and performs fine clustering based on A value and B value in LAB color space, which can more accurately identify and quantify the color characteristics of pixels in each area, especially the change in brightness (L value), so as to achieve precise adjustment of exposure value; not only eliminates problems such as uneven brightness and hue shift, but also improves the accuracy and visual effect of image color correction. In addition, through the color conversion and clustering analysis process in this application, it can also ensure that the images captured by different lenses follow the unified algorithm logic and parameter settings during the color correction process, thereby effectively solving the problem of inconsistent correction results caused by algorithm differences between different software or versions; this enables images taken by different fisheye lenses in a multi-camera system to present a more unified and coordinated color performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 A flowchart of an image color balancing method based on multiple groups of fisheye lenses provided in one embodiment of the present application;

[0049] Figure 2 A block diagram of an image color balancing system based on multiple groups of fisheye lenses provided in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0051] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0052] Example:

[0053] See also Figure 1 In one embodiment of the present application, a method for image color balancing based on multiple groups of fisheye lenses is provided, comprising the following steps:

[0054] S1. Acquire an RGB image, convert RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space.

[0055] RGB color data represents the color information of each pixel in the image, which is composed of the intensity values ​​of the three color channels: red, green, and blue. The value range of the three color channels is 0 to 255.

[0056] LAB color space data includes L value, A value and B value; among them, L value refers to the brightness of the image; A value refers to the value of the red and green components, indicating the change from green to red; B value refers to the value of the yellow and blue components, indicating the change from blue to yellow.

[0057] Convert the RGB color data in the RGB image into LAB color space data to obtain an image represented in the LAB color space, including:

[0058] S11, normalizing the RGB color data to obtain a normalized RGB value;

[0059] Furthermore, it is necessary to normalize the value of each color channel (red, green, blue) in the RGB image from the range of [0,255] to the range of [0,1]; this is achieved by dividing the value of each channel by 255, and the normalization formula is:

[0060]

[0061] Among them, R s The original red channel value; G s The original green channel value; B s Represents the original blue channel value; R′ s represents the normalized red channel value, G′ s Normalized green channel value, B′ s Normalized blue channel value.

[0062] The normalized RGB value is converted into a linear RGB value, and the specific formula is:

[0063]

[0064] The linear RGB value is converted into a value in the XYZ color space, and the specific formula is:

[0065]

[0066] Specifically, the XYZ color space is a color space based on the human eye's perception of color; the X value represents the relative brightness of red light stimulating vision, the Y value represents the relative brightness of green light stimulating vision, and the Z value represents the relative brightness of blue light stimulating vision.

[0067] Normalizing the value of the XYZ color space to a reference white point XYZ value, and calculating the LAB color space data according to the normalized reference white point XYZ value;

[0068] The XYZ values ​​of the XYZ color space normalized to the reference white point are:

[0069] X n =95.047, Y n =100.000, Z n =108.883

[0070] The LAB color space data is calculated based on the normalized reference white point value, and the formula is as follows:

[0071] First calculate the middle value:

[0072]

[0073] Among them, fx, fy, and fz are the intermediate values ​​of the three color components X, Y, and Z respectively calculated by the f(t) function; t is an intermediate variable for further mathematical transformation of the X, Y, and Z values ​​in the XYZ color space to obtain the corresponding values ​​in the LAB color space.

[0074] The formula for calculating the LAB value is:

[0075] L=116×f Y -16;

[0076] A=500×(f X -f Y );

[0077] B=200×(f Y -f Z );

[0078] S2. Divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result.

[0079] Furthermore, in this embodiment, the image represented by the LAB color space is divided into four parts, and each area occupies 90° of the space; for example, the upper part is 45° to 135°, the lower part is -45° to -135°, the left is 135° to -135°, and the right is -45° to 45°; such area division can achieve better results for subsequent clustering.

[0080] Furthermore, DBSCAN clustering is performed on the valid pixels in each area according to the A value and the B value in the LAB color space data to obtain a clustering result, including:

[0081] S21, determine the minimum density number and ε radius;

[0082] Specifically, the minimum density is set to 5% of the total pixels of the image; the ε radius parameter can be adaptively adjusted, and the steps include:

[0083] Traverse all points in the area and calculate the average point spacing of the minimum density points near each point;

[0084] Find the average value of all average point distances ε mi and mean square error sigma;

[0085] Set the ε radius to: ε mi +sigma.

[0086] S22, treating the A value and the B value of the pixel point in the image as the point in the data set, and marking all the points in the data set as unvisited;

[0087] S23, randomly selecting an unvisited point from the data set, and marking the selected point as visited;

[0088] S24, finding all neighboring points of the selected point within the ε radius;

[0089] S25. If the number of the neighborhood points is greater than or equal to the minimum density number, the selected point is regarded as a core point, a new cluster is created, and the neighborhood points are added to the new cluster;

[0090] S26, determining whether each neighboring point of the core point is a core point, and if so, adding the neighboring point to the current cluster and marking it as visited;

[0091] S27, if the number of the neighborhood points is less than the minimum density number, marking the selected point as a noise point;

[0092] S28. Repeat the above steps for all unvisited points until all points have been processed.

[0093] S3. Based on the clustering result, calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space, and adjust the exposure value of the image according to the error.

[0094] Further, all clustering results are looped, and the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space of each clustering result set are calculated;

[0095] According to the average L value of the pixel point set of the target image of each clustering result set in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space, calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space;

[0096] The error calculation formula between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space is:

[0097] T = T + (meanL1-meanL2) * (the number of cluster points in each clustering result / the number of all pixels);

[0098] Among them, meanL1 is the average L value of the pixel point set of the target image in the LAB color space; meanL2 is the average L value of the pixel point set of the image to be balanced in the LAB color space; the initial value of T is 0.

[0099] Finally, according to the calculated error (i.e., the result of weighted average of L value differences in all corresponding areas, which represents the difference in brightness between the current image to be balanced and the target image), the exposure value of the image to be balanced is adjusted until the change in the error is less than a preset threshold; wherein the preset threshold is set based on experience.

[0100] It should be noted that, for the sake of convenience of description, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously.

[0101] Based on the same idea as the image color balancing method based on multiple groups of fisheye lenses in the above-mentioned embodiment, the present application also provides an image color balancing system based on multiple groups of fisheye lenses, which can be used to execute the above-mentioned image color balancing method based on multiple groups of fisheye lenses. For ease of explanation, the structural schematic diagram of the embodiment of the image color balancing system based on multiple groups of fisheye lenses only shows the parts related to the embodiment of the present application. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the system, and may include more or fewer components than shown in the diagram, or combine certain components, or arrange the components differently.

[0102] See also Figure 2 In another embodiment of the present application, there is provided an image color balancing system based on multiple groups of fisheye lenses, the system comprising a conversion module 101, a clustering module 102 and an adjustment module 103;

[0103] The conversion module 101 is used to obtain an RGB image, convert the RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space;

[0104] The clustering module 102 is used to divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result;

[0105] The adjustment module 103 is used to calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space based on the clustering result, and adjust the exposure value of the image according to the error.

[0106] It should be noted that the image color balancing system based on multiple groups of fisheye lenses of the present application corresponds one to one with the image color balancing method based on multiple groups of fisheye lenses of the present application. The technical features and beneficial effects described in the above-mentioned embodiment of the image color balancing method based on multiple groups of fisheye lenses are applicable to the embodiment of the image color balancing system based on multiple groups of fisheye lenses. For specific contents, please refer to the description in the embodiment of the method of the present application, which will not be repeated here. This is hereby declared.

[0107] In addition, in the implementation of the image color balancing system based on multiple groups of fisheye lenses in the above-mentioned embodiment, the logical division of each program module is only an example. In actual applications, the above-mentioned functions can be assigned to different program modules as needed, for example, for the configuration requirements of the corresponding hardware or the convenience of software implementation. That is, the internal structure of the image color balancing system based on multiple groups of fisheye lenses is divided into different program modules to complete all or part of the functions described above.

[0108] In another embodiment, an electronic device for implementing an image color balancing method based on multiple groups of fisheye lenses is provided, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor; when the processor executes the computer program, the image color balancing method based on multiple groups of fisheye lenses of any embodiment of the present application is implemented.

[0109] Exemplarily, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present application. The one or more module elements may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the device.

[0110] The device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The device may include, but is not limited to, a processor and a memory.

[0111] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0112] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; in addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0113] Accordingly, the present application also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the image color balancing method based on multiple groups of fisheye lenses as described in any one of the above embodiments.

[0114] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0115] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The above embodiments are preferred implementation modes of the present application, but the implementation modes of the present application are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present application should be equivalent replacement methods and are included in the protection scope of the present application.

Claims

1. An image color balancing method based on multiple groups of fisheye lenses, characterized in that: The steps include: Acquire an RGB image, convert RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space; Divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result; Based on the clustering result, the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space is calculated, and the exposure value of the image is adjusted according to the error.

2. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The step of converting the RGB color data in the RGB image into LAB color space data to obtain an image represented in the LAB color space includes: Normalizing the RGB color data to obtain a normalized RGB value; Converting the normalized RGB values ​​into linear RGB values; Convert the linear RGB value to a value in the XYZ color space; The value of the XYZ color space is normalized to the XYZ value of the reference white point, and the LAB color space data is calculated according to the normalized XYZ value of the reference white point.

3. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The LAB color space data includes L value, A value and B value; The L value refers to the brightness of the image; The A value refers to the value of the red and green components, indicating a change from green to red; The B value refers to the value of the yellow-blue component, indicating the change from blue to yellow.

4. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The region division of the image represented by the LAB color space is to divide the image represented by the LAB color space into preset partial regions; wherein each region portion occupies 90° of the space.

5. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The clustering of the valid pixels in each area according to the A value and the B value in the LAB color space data to obtain a clustering result includes: Determine the minimum density number and ε radius; The A value and B value of the pixel point in the image are regarded as points in the data set, and all the points in the data set are marked as unvisited; Randomly select an unvisited point from the data set and mark the selected point as visited; Find all neighboring points of the selected point within the ε radius; If the number of the neighborhood points is greater than or equal to the minimum density number, the selected point is regarded as a core point, a new cluster is created, and the neighborhood points are added to the new cluster; Determine whether each neighboring point of the core point is a core point. If it is a core point, add the neighboring point to the current cluster and mark it as visited; If the number of the neighborhood points is less than the minimum density number, marking the selected point as a noise point; Repeat the above steps for all unvisited points until all points have been processed.

6. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The calculating, based on the clustering result, an error between an average value of L values ​​of a pixel point set of the target image in the LAB color space and an average value of L values ​​of a pixel point set of the image to be balanced in the LAB color space comprises: Loop through all clustering results, and calculate the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space of each clustering result set; According to the average L value of the pixel point set of the target image of each clustering result set in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space, calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space; The error calculation formula between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space is: T = T + (meanL1-meanL2) * (the number of cluster points in each clustering result / the number of all pixels); Among them, meanL1 is the average L value of the pixel point set of the target image in the LAB color space; meanL2 is the average L value of the pixel point set of the image to be balanced in the LAB color space; the initial value of T is 0.

7. The image color balancing method based on multiple groups of fisheye lenses according to claim 1, characterized in that: The adjusting the exposure value of the image according to the error comprises: According to the calculated error, the exposure value of the image to be balanced is adjusted until the change in the error is less than a preset threshold.

8. An image color balancing system based on multiple groups of fisheye lenses, characterized in that: An image color balancing method based on multiple groups of fisheye lenses applied to any one of claims 1-7, comprising a conversion module, a clustering module and an adjustment module; The conversion module is used to obtain an RGB image, convert the RGB color data in the RGB image into LAB color space data, and obtain an image represented in the LAB color space; The clustering module is used to divide the image represented by the LAB color space into regions, and cluster the valid pixels in each region according to the A value and the B value in the LAB color space data to obtain a clustering result; The adjustment module is used to calculate the error between the average L value of the pixel point set of the target image in the LAB color space and the average L value of the pixel point set of the image to be balanced in the LAB color space based on the clustering result, and adjust the exposure value of the image according to the error.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores computer program instructions that can be executed by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the image color balancing method based on multiple groups of fisheye lenses as described in any one of claims 1-7.

10. A computer-readable storage medium storing a program, characterized in that: When the program is executed by a processor, the image color balancing method based on multiple groups of fisheye lenses described in any one of claims 1 to 7 is implemented.

Images