Image processing method, device, medium and equipment

By calculating the gradient vector direction and gradient direction difference of pixel points surrounding the exposure image, distinguishing and removing moving type pixel points in dynamic scenes, the ghosting problem in high dynamic range image fusion in the prior art is solved, and image quality is improved.

CN114037643BActive Publication Date: 2024-12-03CHENGDU LIGHT COLLECTOR TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111337269.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-12
Publication Date
2024-12-03
Estimated Expiration
2041-11-12

AI Technical Summary

Technical Problem

The existing multi-exposure image fusion algorithm is difficult to effectively handle dynamic scenes in high dynamic range scenes, resulting in underexposed or overexposed, especially when new objects or scene changes appear during shooting intervals, ghosts are generated and image quality is affected.

Method used

By calculating the gradient vector direction and gradient direction difference of pixel points surrounding the exposure image, distinguish between motion type and non-motion type pixel points, remove motion type pixel points for image fusion, and improve image quality.

Benefits of technology

Effectively remove messy parts in the image, improve image fusion quality, and obtain clearer high dynamic range images.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114037643B_ABST
    Figure CN114037643B_ABST
Patent Text Reader

Abstract

The present invention discloses an image processing method, device, medium and equipment for image fusion of multiple images to generate a merged image with clear imaging. The image processing method includes: obtaining a set of bracketed exposure images of the same scene, calculating the gradient vector direction of each pixel point of each bracketed exposure image; for the Lth pixel point of the kth bracketed exposure image among N bracketed exposure images, performing the following processing: according to the gradient vector direction of the Lth pixel point of N-1 bracketed exposure images other than the kth bracketed exposure image, and the gradient vector direction of the Lth pixel point of the kth bracketed exposure image, calculating the gradient direction difference of the Lth pixel point, and obtaining the gradient direction differences of N-1 Lth pixel points; and determining the type of the Lth pixel point of the kth bracketed exposure image; performing image fusion according to the type of each pixel point of each bracketed exposure image, and the fused image does not include pixels of motion type.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an image processing method, device, medium and equipment. Background Art

[0002] At present, the dynamic range of photographic and display equipment is much lower than the dynamic range of real scenes. In high dynamic range scenes, photos taken with digital cameras are often underexposed or overexposed. Therefore, it is necessary to perform image fusion processing on a set of bracketed exposure images to obtain a high dynamic range image. Considering that most of the scenes shot in reality are dynamic scenes, for example, when taking multiple bracketed exposure images, new moving objects may appear in the scene between shots. For example, leaves in the scene may also move due to wind. However, most of the existing multi-exposure image fusion algorithms are only for image fusion in static scenes. In the high dynamic range image fusion process, because there is a time difference between the exposure frame images, in some landscape images, there will be ghost images of moving objects, which will affect the image quality.

[0003] Therefore, how to fuse images in dynamic scenes has become an urgent problem that needs to be solved in the industry. Summary of the invention

[0004] The embodiments of the present invention provide an image processing method, apparatus, medium and device for removing cluttered parts in an image during an image fusion process to improve the quality of a fused image.

[0005] In a first aspect, the present invention provides an image processing method, the method comprising:

[0006] Acquire a bracketed exposure image set of the same scene, where the bracketed exposure image set includes N bracketed exposure images, and any bracketed exposure image of the N bracketed exposure images includes M pixels, where N and M are positive integers;

[0007] Calculate the gradient vector direction of each pixel point in each bracketed exposure image;

[0008] For the Lth pixel point of a kth bracketed exposure image among N bracketed exposure images, where the kth bracketed exposure image is any bracketed exposure image and the Lth pixel point is any one of the M pixels, the following processing is performed: according to the gradient vector direction of the Lth pixel point of N-1 bracketed exposure images other than the kth bracketed exposure image and the gradient vector direction of the Lth pixel point of the kth bracketed exposure image, a gradient direction difference of the Lth pixel point is calculated to obtain the gradient direction difference of the N-1 Lth pixel points; according to the gradient direction difference of the N-1 Lth pixel points, a type of the Lth pixel point of the kth bracketed exposure image is determined, where k and L are positive integers;

[0009] According to the types of each pixel point of each bracketed exposure image calculated according to the above method, image fusion is performed on the N bracketed exposure images, wherein the fused image does not include pixel points of the motion type.

[0010] The beneficial effect of the image processing method provided by the present invention is that: in the above method, the gradient vector direction and the gradient direction difference of the pixel point are used to calculate the type of each pixel point, which helps to distinguish the motion type pixels and non-motion type pixels in the image, and then distinguish the moving target and the image background, so that the motion type pixels can be removed as needed during the image fusion process to obtain a clearer image.

[0011] In a possible implementation, determining the type of the Lth pixel of the kth bracketed exposure image according to the gradient direction differences of the N-1 Lth pixel points includes:

[0012] According to the gradient direction differences of the N-1 L-th pixel points, the category score of the L-th pixel point in the k-th bracketed exposure image is calculated, wherein the larger the value of the gradient direction difference of the L-th pixel point is, the stronger the gradient direction change of the L-th pixel point is, and the larger the category score of the L-th pixel point is.

[0013] When the category score is greater than a set threshold, determining that the type of the Lth pixel point of the kth bracketed exposure image is a motion type;

[0014] When the category score is less than or equal to the set threshold, it is determined that the type of the Lth pixel point of the kth bracketed exposure image is a non-motion type.

[0015] In this implementation, it is considered that the gradient direction of non-motion type pixels in the image is basically unaffected by the change in exposure, while the gradient direction of motion type pixels changes dramatically. Therefore, the motion type pixels are identified by calculating the change in gradient direction, and the motion type pixels can be determined more accurately with the help of the gradient change between adjacent pixels.

[0016] In a possible implementation, the category score of the Lth pixel satisfies the following formula:

[0017]

[0018] Among them, D k (x, y) is the category score of the Lth pixel, d i,k (x, y) is the gradient direction difference between the gradient vector direction of the k-th bracketed exposure image and the gradient vector direction of the i-th bracketed exposure image, and the value of i ranges from 1 to N, and i is not equal to k.

[0019] In a possible implementation, the gradient vector direction of each pixel point of each bracketed exposure image is calculated, specifically including the following method:

[0020] For the gradient vector direction of the pixel at the position (x, y) of the i-th bracketed exposure image in N bracketed exposure images, where the value of i ranges from 1 to N, perform the following processing:

[0021] Calculate the horizontal gradient vector value and the vertical gradient vector value of the pixel at the (x, y) position of the i-th image, and calculate the gradient vector direction of the pixel at the (x, y) position according to the horizontal gradient vector value and the vertical gradient vector value. The gradient vector direction of the pixel at the (x, y) position satisfies the following formula:

[0022]

[0023] Among them, O i (x,y) is the gradient vector direction of the pixel at the (x,y) position, T iy (x,y) is the horizontal gradient vector value of the pixel at the (x,y) position, T ix (x,y) is the vertical gradient vector value of the pixel at the (x,y) position.

[0024] In a possible implementation, according to the type of each pixel point of each bracketed exposure image calculated according to the above method, image fusion is performed on N bracketed exposure images, specifically including the following methods:

[0025] Calculate the processing function corresponding to the pixel point of the motion type;

[0026] In the image fusion process of each bracketed exposure image, according to the type of each pixel point of each bracketed exposure image calculated according to the above method, the processing function is used to perform mask processing on the pixel points of the motion type, so that the fused image does not include the pixel points of the motion type;

[0027] The processing function satisfies the following formula:

[0028]

[0029] Among them, R' k (x, y) is the processing value of the pixel at the position (x, y), α is the judgment threshold, which needs to be set according to the actual scene, D k (x,y) is the category score of the Lth pixel.

[0030] In a second aspect, the present application also provides an image processing device, which includes modules / units for executing any possible design method of the first aspect. These modules / units can be implemented by hardware, or by executing corresponding software implementations by hardware.

[0031] In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory, wherein the memory is used to store one or more programs; when the one or more programs stored in the memory are executed by the processor, the electronic device can implement any possible design method of the first aspect.

[0032] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a program, and when the program runs on an electronic device, the electronic device executes any possible design method of the first aspect above.

[0033] In a fifth aspect, an embodiment of the present application further provides a method comprising a program product, which, when executed on a terminal, enables the electronic device to execute any possible design of the first aspect.

[0034] For the beneficial effects of the second to fifth aspects, reference may be made to the description of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A schematic diagram of a communication system architecture provided by an embodiment of the present invention;

[0037] Figure 2 A schematic diagram of a flow chart of an image processing method provided by an embodiment of the present invention;

[0038] Figure 3 A schematic diagram of an application scenario provided by an embodiment of the present invention;

[0039] Figure 4 A schematic diagram of an image processing device provided by an embodiment of the present invention;

[0040] Figure 5 A schematic diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to improve the shortcomings of existing bracketed exposure images during image fusion, the present invention provides an image processing method, which can use the gradient vector direction and gradient direction difference of the pixel point to calculate the type of each pixel point, which helps to distinguish between motion type pixels and non-motion type pixels in the image, and then distinguish between moving targets and image backgrounds, so that the motion type pixels can be removed as needed during the image fusion process to obtain a clearer image.

[0042] Some of the terms used in the embodiments of the present application are explained below to facilitate understanding by those skilled in the art.

[0043] 1. Exposure bracketing

[0044] Using bracketing exposure in the imaging stage can obtain multiple frames of images with multiple levels of different exposure and containing real colors, and the multiple frames of images can constitute a bracketing exposure image set.

[0045] 2. High-Dynamic Range (HDR) images

[0046] Compared with ordinary images, HDR images can provide more dynamic range and image details. According to low-dynamic range (LDR) images with different exposure times, the final HDR image is synthesized using the LDR image with the best details corresponding to each exposure time, so as to better reflect the visual scene in the real environment.

[0047] In the description of the embodiments of the present application, the terms used in the following embodiments are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and the appended claims of the present application, the singular expressions "a", "said", "above", "the" and "this" are intended to also include such expressions as "one or more", unless there is a clear contrary indication in the context. It should also be understood that in the following embodiments of the present application, "at least one", "one or more" refer to one or more (including two). The term "and / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist; for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are a kind of "or" relationship.

[0048] References to "one embodiment" or "some embodiments" etc. described in this specification mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Thus, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways. The term "connection" includes direct connection and indirect connection, unless otherwise specified. "First" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated.

[0049] In the embodiments of the present application, the words "exemplarily" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplarily" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplarily" or "for example" is intended to present related concepts in a specific way.

[0050] The image processing method provided in the embodiments of the present application can be applied to Figure 1 The application scenario shown includes a server 100 and a terminal device 200.

[0051] In one possible design, the server 100 is used to obtain a set of bracketed exposure images of the same scene from the terminal device 200, and then calculate the gradient vector direction and gradient direction difference of the pixel points, and use the gradient vector direction and gradient direction difference of the pixel points to calculate the type of each pixel point, so as to distinguish between motion type pixels and non-motion type pixels in the image, and remove the motion type pixels during the image fusion process as needed to obtain a clearer image, and then send the fused clear image to the terminal device 200.

[0052] In another possible design, the terminal device 200 obtains a set of bracketed exposure images of the same scene, and then calculates the gradient vector direction and gradient direction difference of the pixel points, and uses the gradient vector direction and gradient direction difference of the pixel points to calculate the type of each pixel point, thereby distinguishing between motion type pixels and non-motion type pixels in the image, and removing the motion type pixels during the image fusion process as needed to obtain a clearer image.

[0053] The server 100 and the terminal device 200 may be connected via a wireless network, and the terminal device 200 may be a terminal device with an image sensor, such as a smart phone, a tablet computer, etc. The server 100 may be a single server, or a server cluster or a cloud computing center consisting of several servers.

[0054] based on Figure 1 The application scenario diagram shown in the figure, the embodiment of the present application provides a process of an image processing method, such as Figure 2 As shown, the process of the method can be executed by the terminal device 200, and the method includes the following steps:

[0055] S201, obtaining a bracketed exposure image set of the same scene, where the bracketed exposure image set includes N bracketed exposure images.

[0056] Any of the N bracketed exposure images includes M pixels, and N and M are positive integers.

[0057] S202, calculating the direction of the gradient vector of each pixel point of each bracketed exposure image.

[0058] In S202, in a possible implementation manner, for the gradient vector direction of the pixel point at the position (x, y) of the i-th bracketed exposure image among the N bracketed exposure images, where the value of i ranges from 1 to N, the following processing is performed:

[0059] Calculate the horizontal gradient vector value and the vertical gradient vector value of the pixel point at the (x, y) position of the i-th image, and calculate the gradient vector direction of the pixel point at the (x, y) position according to the horizontal gradient vector value and the vertical gradient vector value. The gradient vector direction of the pixel point at the (x, y) position satisfies the following formula 1:

[0060]

[0061] Among them, O i (x,y) is the gradient vector direction of the pixel at the (x,y) position, T iy (x,y) is the horizontal gradient vector value of the pixel at the (x,y) position, T ix (x,y) is the vertical gradient vector value of the pixel at the (x,y) position.

[0062] S203, performing the following processing for the Lth pixel point of the kth bracketed exposure image among the N bracketed exposure images: calculating the gradient direction difference of the Lth pixel point based on the gradient vector direction of the Lth pixel point of N-1 bracketed exposure images other than the kth bracketed exposure image and the gradient vector direction of the Lth pixel point of the kth bracketed exposure image, and obtaining the gradient direction differences of N-1 Lth pixels; determining the type of the Lth pixel point of the kth bracketed exposure image based on the gradient direction differences of the N-1 Lth pixels.

[0063] The kth bracketed exposure image is any bracketed exposure image, and the Lth pixel point is any one of the M pixel points.

[0064] In the above S203, in a possible implementation manner, determining the type of the Lth pixel of the kth bracketed exposure image according to the gradient direction difference of the N-1 Lth pixel, specifically includes:

[0065] According to the gradient direction difference of the N-1 L-th pixel points, the category score of the L-th pixel point of the k-th bracketed exposure image is calculated; when the category score is greater than a set threshold, the type of the L-th pixel point of the k-th bracketed exposure image is determined to be a motion type; when the category score is less than or equal to the set threshold, the type of the L-th pixel point of the k-th bracketed exposure image is determined to be a non-motion type.

[0066] Among them, the category score of the Lth pixel satisfies the following formula 2:

[0067]

[0068] d i,k (x,y)=|O i (x,y)-O k (x,y)|..........Formula 3

[0069] Among them, D k (x, y) is the category score of the Lth pixel, d i,k (x, y) is the gradient vector direction O of the kth bracketed exposure image k (x, y) and the gradient vector direction O of the i-th bracketed exposure image i The gradient direction difference between (x, y), the value of i ranges from 1 to N, and i is not equal to k.

[0070] It can be seen from the above formulas 2 and 3 that the larger the value of the gradient direction difference of the Lth pixel point, the stronger the gradient direction change of the Lth pixel point, and the larger the category score of the Lth pixel point.

[0071] S204, performing image fusion on the N bracketed exposure images according to the types of the pixels of the bracketed exposure images calculated according to the above method, wherein the fused images do not include pixels of motion type.

[0072] In other words, in the process of image fusion of N bracketed exposure images, the pixels of the motion type are removed, or the pixels of the motion type are covered with the generated mask, so as to obtain a clear image.

[0073] In S204, in a possible implementation manner, a processing function corresponding to a pixel point of the motion type may be calculated first; and then, in the image fusion process of each bracketed exposure image, according to the type of each pixel point of each bracketed exposure image calculated according to the above method, the processing function is used to perform mask processing on the pixel point of the motion type, so that the fused image does not include the pixel point of the motion type;

[0074] The processing function satisfies the following formula:

[0075]

[0076] Among them, R' k (x, y) is the processing value of the pixel at the position (x, y), α is the judgment threshold, which needs to be set according to the actual scene, D k (x,y) is the category score of the Lth pixel.

[0077] It can be seen that after the motion type pixel is processed by the processing function, the value of the motion type pixel is 0, and after the non-motion type pixel is processed by the processing function, the value of the non-motion type pixel is 1.

[0078] For example, Figure 3 As shown, a set of bracketed exposure images includes a candidate bracketed exposure image 301 and its adjacent bracketed exposure image 302 and adjacent bracketed exposure image 303. First, the pixel values ​​of the candidate bracketed exposure image 301, its adjacent bracketed exposure image 302 and adjacent bracketed exposure image 303 are quantized to the range of [0,1]. Then, the gradient vector direction and gradient direction difference of each pixel point of each bracketed exposure image are calculated. For each pixel point, the category score of the pixel point is obtained for the two gradient direction differences corresponding to the pixel point, so as to determine the type of the pixel point. The type of the pixel point may include motion type and non-motion type. Based on the type of each pixel point, the content in the image may be distinguished into target content, response content or background content. The target content is the area where the moving object is located. The response content indicates that there is a moving object in this area in the adjacent image of the processed image. The background content indicates that there is no moving object in this area in the candidate image and its adjacent image set. Figure 3As shown, the target content is a, the response content is b, and the background content is the content of the adjacent bracketed exposure image 302 except the target content a. Based on the type of pixel points, the three bracketed exposure images are fused to obtain a fused image 304. It can be seen from the fused image 304 that the target content a and the response content b containing the pixels of the motion type are removed, and the fused image is a clear image.

[0079] In some embodiments of the present application, the present application embodiment discloses an image processing device, such as Figure 4 As shown, the device is used to implement the methods recorded in the above method embodiments, and includes: an acquisition unit 401, a calculation unit 402, a determination unit 403 and a processing unit 404. The acquisition unit 401 is used to acquire a bracketed exposure image set about the same scene, the bracketed exposure image set includes N bracketed exposure images, any bracketed exposure image of the N bracketed exposure images includes M pixels, and N and M are positive integers;

[0080] A calculation unit 402 is used to calculate the gradient vector direction of each pixel point of each bracketed exposure image;

[0081] The determining unit 403 is configured to perform the following processing on an L-th pixel point of a k-th bracketed exposure image among N bracketed exposure images, where the k-th bracketed exposure image is any bracketed exposure image and the L-th pixel point is any one of the M pixels points: calculating a gradient direction difference of the L-th pixel point according to a gradient vector direction of the L-th pixel point of N-1 bracketed exposure images other than the k-th bracketed exposure image and a gradient vector direction of the L-th pixel point of the k-th bracketed exposure image, to obtain a gradient direction difference of N-1 L-th pixel points; determining a type of the L-th pixel point of the k-th bracketed exposure image according to the gradient direction difference of the N-1 L-th pixel points, where k and L are positive integers;

[0082] The processing unit 404 is used to perform image fusion according to the type of each pixel point of each bracketed exposure image, wherein the fused image does not include pixels of the motion type. All relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here.

[0083] In other embodiments of the present application, the present application embodiment discloses a terminal device, such as Figure 5As shown, the device may include: one or more processors 501; a memory 502; a display 503; one or more application programs (not shown); and one or more programs 504, and the above-mentioned components may be connected via one or more communication buses 505. The one or more programs 504 are stored in the above-mentioned memory 502 and are configured to be executed by the one or more processors 501, and the one or more programs 504 include instructions.

[0084] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0085] Each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0086] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as flash memory, mobile hard disk, read-only memory, random access memory, disk or optical disk.

[0087] The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto, and any changes or replacements within the technical scope disclosed in the embodiment of the present application should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.

Claims

1. An image processing method, characterized in that: The method includes: Acquire a bracketed exposure image set of the same scene, where the bracketed exposure image set includes N bracketed exposure images, and any bracketed exposure image of the N bracketed exposure images includes M pixels, where N and M are positive integers; Calculate the gradient vector direction of each pixel point in each bracketed exposure image; For the Lth pixel point of a kth bracketed exposure image among N bracketed exposure images, where the kth bracketed exposure image is any bracketed exposure image and the Lth pixel point is any one of the M pixels, the following processing is performed: according to the gradient vector direction of the Lth pixel point of N-1 bracketed exposure images other than the kth bracketed exposure image and the gradient vector direction of the Lth pixel point of the kth bracketed exposure image, a gradient direction difference of the Lth pixel point is calculated to obtain the gradient direction difference of the N-1 Lth pixel points; according to the gradient direction difference of the N-1 Lth pixel points, a type of the Lth pixel point of the kth bracketed exposure image is determined, where k and L are positive integers; According to the types of each pixel point of each bracketed exposure image calculated according to the above method, the N bracketed exposure images are image fused, wherein the fused image does not include pixels of the motion type, including: Calculate the processing function corresponding to the pixel point of the motion type; In the image fusion process of each bracketed exposure image, according to the type of each pixel point of each bracketed exposure image calculated according to the above method, the processing function is used to perform mask processing on the pixel points of the motion type, so that the fused image does not include the pixel points of the motion type; The processing function satisfies the following formula: Among them, R′ k (x, y) is the processing value of the pixel at the position (x, y), α is the judgment threshold, which needs to be set according to the actual scene, D k (x,y) is the category score of the Lth pixel.

2. The method according to claim 1, characterized in that Determining the type of the Lth pixel of the kth bracketed exposure image according to the gradient direction difference of the N-1th Lth pixel, including: According to the gradient direction differences of the N-1 L-th pixel points, the category score of the L-th pixel point in the k-th bracketed exposure image is calculated, wherein the larger the value of the gradient direction difference of the L-th pixel point is, the stronger the gradient direction change of the L-th pixel point is, and the larger the category score of the L-th pixel point is. When the category score is greater than a set threshold, determining that the type of the Lth pixel point of the kth bracketed exposure image is a motion type; When the category score is less than or equal to the set threshold, it is determined that the type of the Lth pixel point of the kth bracketed exposure image is a non-motion type.

3. The method according to claim 2, characterized in that The category score of the Lth pixel satisfies the following formula: Among them, D k (x, y) is the category score of the Lth pixel, d i,k (x, y) is the gradient direction difference between the gradient vector direction of the k-th bracketed exposure image and the gradient vector direction of the i-th bracketed exposure image, and the value of i ranges from 1 to N, and i is not equal to k.

4. The method according to any one of claims 1 to 3, characterized in that: The step of calculating the gradient vector direction of each pixel point of each bracketed exposure image includes: For the gradient vector direction of the pixel at the position (x, y) of the i-th bracketed exposure image in N bracketed exposure images, where the value of i ranges from 1 to N, perform the following processing: Calculate the horizontal gradient vector value and the vertical gradient vector value of the pixel at the (x, y) position of the i-th image, and calculate the gradient vector direction of the pixel at the (x, y) position according to the horizontal gradient vector value and the vertical gradient vector value. The gradient vector direction of the pixel at the (x, y) position satisfies the following formula: Among them, O i (x,y) is the gradient vector direction of the pixel at the (x,y) position, T iy (x,y) is the horizontal gradient vector value of the pixel at the (x,y) position, T ix (x,y) is the vertical gradient vector value of the pixel at the (x,y) position.

5. An image processing device, used to implement the image processing method according to any one of claims 1 to 4, characterized in that: The device includes: An acquisition unit is used to acquire a bracketed exposure image set of the same scene, wherein the bracketed exposure image set includes N bracketed exposure images, and any bracketed exposure image of the N bracketed exposure images includes M pixels, where N and M are positive integers; A calculation unit, used for calculating the gradient vector direction of each pixel point of each bracketed exposure image; a determining unit, configured to perform the following processing on an L-th pixel point of a k-th bracketed exposure image among N bracketed exposure images, where the k-th bracketed exposure image is any bracketed exposure image, and the L-th pixel point is any one of the M pixels points: calculating a gradient direction difference of the L-th pixel point according to a gradient vector direction of the L-th pixel point of N-1 bracketed exposure images other than the k-th bracketed exposure image and a gradient vector direction of the L-th pixel point of the k-th bracketed exposure image, to obtain a gradient direction difference of N-1 L-th pixels points; and determining a type of the L-th pixel point of the k-th bracketed exposure image according to the gradient direction difference of the N-1 L-th pixels points, where k and L are positive integers; The processing unit is used to perform image fusion according to the type of each pixel point of each bracketed exposure image, wherein the fused image does not include pixel points of motion type.

6. The device according to claim 5, characterized in that The determining unit determines the type of the Lth pixel point of the kth bracketed exposure image according to the gradient direction difference of the N-1th Lth pixel point, and is specifically used to: According to the gradient direction differences of the N-1 L-th pixel points, the category score of the L-th pixel point in the k-th bracketed exposure image is calculated, wherein the larger the value of the gradient direction difference of the L-th pixel point is, the stronger the gradient direction change of the L-th pixel point is, and the larger the category score of the L-th pixel point is. When the category score is greater than a set threshold, determining that the type of the Lth pixel point of the kth bracketed exposure image is a motion type; When the category score is less than or equal to the set threshold, it is determined that the type of the Lth pixel point of the kth bracketed exposure image is a non-motion type.

7. The device according to claim 6, characterized in that The category score of the Lth pixel satisfies the following formula: Among them, D k (x, y) is the category score of the Lth pixel, d i,k (x, y) is the gradient direction difference between the gradient vector direction of the k-th bracketed exposure image and the gradient vector direction of the i-th bracketed exposure image, and the value of i ranges from 1 to N, and i is not equal to k.

8. The device according to any one of claims 5 to 6, characterized in that When calculating the gradient vector direction of each pixel point of each bracketed exposure image, the calculation unit is specifically used to: For the gradient vector direction of the pixel at the position (x, y) of the i-th bracketed exposure image in N bracketed exposure images, where the value of i ranges from 1 to N, perform the following processing: Calculate the horizontal gradient vector value and the vertical gradient vector value of the pixel at the (x, y) position of the i-th image, and calculate the gradient vector direction of the pixel at the (x, y) position according to the horizontal gradient vector value and the vertical gradient vector value. The gradient vector direction of the pixel at the (x, y) position satisfies the following formula: Among them, O i (x,y) is the gradient vector direction of the pixel at the (x,y) position, T iy (x,y) is the horizontal gradient vector value of the pixel at the (x,y) position, T ix (x,y) is the vertical gradient vector value of the pixel at the (x,y) position.

9. The device according to any one of claims 5 to 6, characterized in that: Image fusion is performed according to the type of each pixel point of each bracketed exposure image, wherein the fused image does not include pixels of the motion type, including: Calculate the processing function corresponding to the pixel point of the motion type; In the image fusion process of each bracketed exposure image, the processing function is used to perform mask processing on the pixels of the motion type, so that the fused image does not include the pixels of the motion type; The processing function satisfies the following formula: Among them, R′ k (x, y) is the processing value of the pixel at the position (x, y), α is the judgment threshold, which needs to be set according to the actual scene, D k (x,y) is the category score of the Lth pixel.

10. A readable storage medium, wherein a program is stored in the readable storage medium, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

11. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores a program executable on the processor, and when the program is executed by the processor, the electronic device implements the method according to any one of claims 1 to 4.