Image restoration method and system

Abstract

The present invention provides an image restoration method and system, belonging to the field of artificial intelligence technology. The image restoration method comprises: determining the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center; determining the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center; determining the target offset of the pixel point according to the intermediate membership, and restoring the original image by the target offset of the pixel point. The present invention can improve the clarity and accuracy of the image.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an image restoration method and system. Background Art

[0002] In image acquisition scenarios, images are often damaged by non-uniform fields containing slowly changing grayscales. Non-uniform fields cause grayscales in the same part to change, thus affecting the clarity and accuracy of the image. Existing technologies use spatial information of images and algorithms with the function of correcting offset fields to repair images. However, the repaired images have high-frequency components, making it difficult to clearly display the images. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide an image restoration method and system to accurately restore images and improve the clarity and accuracy of images.

[0004] In order to achieve the above object, an embodiment of the present invention provides an image restoration method, comprising:

[0005] Determine the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center;

[0006] Determine the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center;

[0007] The target offset of the pixel is determined according to the intermediate membership, and the original image is restored by the target offset of the pixel.

[0008] The embodiment of the present invention also provides an image restoration system, including:

[0009] A distance function determination unit, used to determine the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center;

[0010] An intermediate membership determination unit, used to determine the intermediate membership of the pixel point with respect to the initial cluster center according to a distance function from the pixel point to the initial cluster center;

[0011] The image restoration unit is used to determine the target offset of the pixel points according to the intermediate membership, and to restore the original image by the target offset of the pixel points.

[0012] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the steps of the image restoration method are implemented when the processor executes the computer program.

[0013] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the image restoration method are implemented.

[0014] The image restoration method and system of the embodiment of the present invention first determine the distance function according to the pixels of the original image, the initial offset of the pixels, the initial cluster center and the initial membership, then determine the intermediate membership according to the distance function, and then determine the target offset of the pixels according to the intermediate membership. The original image is accurately restored by the target offset of the pixels, which can improve the clarity and accuracy of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, 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.

[0016] Figure 1 is a flow chart of an image restoration method according to an embodiment of the present invention;

[0017] Figure 2 is a flow chart of an image restoration method according to another embodiment of the present invention;

[0018] Figure 3 is a flow chart of determining a target offset in one embodiment of the present invention;

[0019] Figure 4 is a flow chart of determining a target offset in another embodiment of the present invention;

[0020] Figure 5 It is the bill image that is not affected by the uneven illumination field;

[0021] Figure 6 It is the bill image affected by the non-uniform illumination field;

[0022] Figure 7 The bill image is restored by the image restoration method of the present invention;

[0023] Figure 8 is a structural block diagram of an image restoration system in an embodiment of the present invention;

[0024] Fig. 9 It is a structural block diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0027] In view of the fact that it is difficult to clearly display an image in the prior art, an embodiment of the present invention provides an image restoration method, which can accurately restore an image and improve the clarity and accuracy of the image. The present invention is described in detail below in conjunction with the accompanying drawings.

[0028] An image subjected to inhomogeneous illumination can be equivalent to an image that is not actually contaminated by the inhomogeneous field multiplied by a slowly varying inhomogeneous field in the spatial domain:

[0029] Z i =X i G i , i∈{i,2,3,...,N};

[0030] In the formula, Z i is the value of the observed image at the i-th point, X i is the value of the image at the i-th point that is not contaminated by the uniform field, G i The value of the multiplicative inhomogeneous field at the i-th point, N is the total number of pixels in the image. For ease of calculation, take the logarithm on both sides of the above formula to obtain:

[0031] z i =x i +β i , i∈{i,2,3,...,N};

[0032] In the formula, z i is the logarithmic value of the observed image at the i-th point, x i is the logarithmic value of the image at the i-th point that is not contaminated by the uniform field, β i is the logarithm of the multiplicative inhomogeneous field at the i-th point. The purpose of grayscale image correction and restoration is to correct the grayscale image from z i Lieutenant General x i Recover.

[0033] For example, when collecting images of counter document information, invoice contract information, and physical bills at a bank, the collected images are unclear due to factors such as uneven lighting and reflection. The present invention can effectively remove the impact of uneven lighting on the image through the optimized FCM (mean clustering) algorithm, thereby achieving the purpose of image restoration.

[0034] Figure 1 is a flow chart of an image restoration method in one embodiment of the present invention. Figure 2 FIG. 1 is a flow chart of an image restoration method according to another embodiment of the present invention. Figure 1 and Figure 2 As shown, the image restoration method includes:

[0035] S101: determining a distance function from a pixel point to an initial cluster center according to a pixel point of an original image, an initial offset of the pixel point, an initial cluster center, and an initial membership degree of a neighborhood point corresponding to the pixel point with respect to the initial cluster center.

[0036] The initial offset of the pixel point may be a smaller value, such as 0.01.

[0037] In one embodiment, before executing S101, the method further includes:

[0038] Obtain an initial image; remove the background area of ​​the initial image to obtain the original image.

[0039] In specific implementation, a morphological method can be used to remove the background area and the area of ​​no interest of the initial image to obtain the original image.

[0040] In one embodiment, it is assumed that a finite set Y = {y1, y2, y3, ..., y N} is a set of N pixels, c is the number of predetermined categories, y i is the i-th pixel, v j is the jth cluster center, u j (y i ) is the membership of the i-th pixel to the j-th cluster center, then the clustering criterion function is defined by the membership function as:

[0041]

[0042] Among them, J(U,V) is the clustering criterion function, ||y i -v j || is y i to v j (j=1,2,3,...,c) square of the Euclidean distance; b is the fuzzy weighted power index, which is a parameter that can control the fuzziness of the clustering result; U is the fuzzy matrix of Y, and satisfies u j (y i)∈[0,1] and V is the set of cluster centers of Y.

[0043] The present invention improves the FCM algorithm and incorporates spatial relationship constraints: the purpose of the FCM clustering algorithm is to obtain V and U that minimize the criterion function. The algorithm defines each pixel y i About cluster center v j The membership degree u j (y i )for:

[0044]

[0045] Among them, v m is the mth cluster center.

[0046] according to and The following formula can be obtained to determine the distance function from the pixel to the initial cluster center:

[0047]

[0048]

[0049] Among them, d 2 (y i ,v j ) is the i-th pixel y i To the jth initial cluster center v j The distance function, y i is the i-th pixel, β i is the initial offset of the i-th pixel, v j is the jth initial cluster center, λ is the first adjustment space constraint relationship coefficient, λ∈[0,1], ξ is the second adjustment space constraint relationship coefficient, ξ∈[0,1], u j (y k ) is the kth area point y corresponding to the i-th pixel k Regarding the initial membership of the jth initial cluster center, G ik is the absolute value of the grayscale difference between the i-th pixel and the corresponding k-th field point, P ik is the neighborhood distance between the ith pixel and the corresponding kth neighborhood point, s is the value of y i The center point and y i The corresponding field point y k The number of 2 (y i ,v m ) is the i-th pixel y i To the mth initial cluster center v m The distance function, vm is the mth initial cluster center, u m (y k ) is the kth area point y corresponding to the i-th pixel k The initial membership of the mth initial cluster center.

[0050] It can be seen from this that the clustering fuzzy matrix is ​​not only determined by the grayscale of the pixel itself, but also affected by pixels in other areas.

[0051] S102: Determine the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center.

[0052] In one embodiment, the intermediate membership degree may be determined by the following formula:

[0053]

[0054] Among them, u j (y i ) is the i-th pixel y i Regarding the intermediate membership of the jth initial cluster center, b is the fuzzy weighted power exponent.

[0055] S103: Determine a target offset of the pixel point according to the intermediate membership degree, and repair the original image by the target offset of the pixel point.

[0056] In a specific implementation, the corresponding pixel points of the original image can be restored according to the target offset (non-uniform field) of the pixel points to remove the illumination non-uniform field and repair the original image.

[0057] Figure 3 FIG. 1 is a flow chart of determining a target offset in one embodiment of the present invention. Figure 3 As shown, determining the target offset of the pixel point according to the intermediate membership includes:

[0058] S201: Determine the intermediate offset of the pixel point according to the intermediate membership and the initial cluster center.

[0059] In one embodiment, the middle offset of the pixel point can be determined by the following formula:

[0060]

[0061] Among them, β' i is the middle offset of the i-th pixel.

[0062] S202: Determine a target offset of the pixel point according to the intermediate offset.

[0063] Figure 4 FIG. 1 is a flow chart of determining a target offset in another embodiment of the present invention. Figure 4 As shown, S202 also includes:

[0064] S301: Determine an intermediate cluster center according to the intermediate offset and the intermediate membership.

[0065] In one embodiment, the intermediate cluster center may be determined by the following formula:

[0066]

[0067] Among them, v' j is the jth intermediate cluster center, (u(y i )) max is the i-th pixel y i The maximum value of the intermediate membership about the initial cluster centers.

[0068] S302: Determine a target offset of the pixel point through the intermediate membership, the intermediate offset and the intermediate cluster center.

[0069] In one embodiment, S302 further includes:

[0070] 1. After replacing the initial membership, initial offset and initial cluster center with the intermediate membership, intermediate offset and intermediate cluster center respectively, corresponding iterative calculations are performed.

[0071] Among them, the domain point corresponding to the pixel point is also one of the pixel points of the original image. Therefore, the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center can be replaced according to the intermediate membership of the pixel point with respect to the initial cluster center.

[0072] 2. When the current number of iterations is equal to the preset number of iterations, the intermediate offset obtained by the current iteration calculation is used as the target offset of the pixel point.

[0073] Figure 1 The execution subject of the image restoration method shown can be a computer. Figure 1 As can be seen from the process shown, the image restoration method of the embodiment of the present invention first determines the distance function based on the pixels of the original image, the initial offset of the pixels, the initial clustering center and the initial membership, then determines the intermediate membership based on the distance function, and then determines the target offset of the pixels based on the intermediate membership. The original image is accurately restored by the target offset of the pixels, which can improve the clarity and accuracy of the image.

[0074] The specific process of the embodiment of the present invention is as follows:

[0075] 1. Get the initial image, remove the background area and the area of ​​no interest of the initial image, and get the original image.

[0076] 2. Determine the distance function from the pixel point to the initial cluster center according to the pixel point, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center.

[0077] 3. Determine the intermediate membership of the pixel point with respect to the initial cluster center based on the distance function from the pixel point to the initial cluster center.

[0078] 4. Determine the intermediate offset of the pixel point based on the intermediate membership and the initial cluster center.

[0079] 5. Determine the intermediate cluster center based on the intermediate offset and intermediate membership.

[0080] 6. Determine whether the current number of iterations is equal to the preset number of iterations.

[0081] 7. When the current number of iterations is equal to the preset number of iterations, the intermediate offset calculated by the current iteration is used as the target offset of the pixel point. Otherwise, the initial membership is replaced by the intermediate membership, the initial offset is replaced by the intermediate offset, and the initial cluster center is replaced by the intermediate cluster center, and return to step 2.

[0082] 8. Repair the original image according to the target offset of the pixels.

[0083] Figure 5 It is the bill image that is not affected by the uneven illumination field. Figure 6 It is a bill image affected by the uneven illumination field. Figure 7 The bill image is restored by the image restoration method of the present invention. Figure 5-Figure 7 As shown, the restored bill image can be recognized by the human eye, so it can be archived and saved as image information, achieving the purpose of image restoration.

[0084] In summary, the present invention proposes a method for repairing an image with uneven illumination, which not only introduces an offset field function, but also introduces the spatial domain information of the image into the FCM clustering criterion function based on grayscale information, which can accurately repair the original image and improve the clarity and accuracy of the image.

[0085] Based on the same inventive concept, an embodiment of the present invention further provides an image restoration system. Since the principle of solving the problem by the system is similar to that of the image restoration method, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be repeated.

[0086] Figure 8 is a structural block diagram of an image restoration system in an embodiment of the present invention. Figure 8 As shown, the image restoration system includes:

[0087] A distance function determination unit, used to determine the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center;

[0088] An intermediate membership determination unit, used to determine the intermediate membership of the pixel point with respect to the initial cluster center according to a distance function from the pixel point to the initial cluster center;

[0089] The image restoration unit is used to determine the target offset of the pixel points according to the intermediate membership, and to restore the original image by the target offset of the pixel points.

[0090] In one embodiment, the image restoration unit is specifically used for:

[0091] Determine the intermediate offset of the pixel point according to the intermediate membership and the initial cluster center;

[0092] The target offset of the pixel is determined based on the intermediate offset.

[0093] In one embodiment, the image restoration unit is further configured to:

[0094] Determine the intermediate cluster center according to the intermediate offset and intermediate membership;

[0095] The target offset of the pixel point is determined by the intermediate membership, intermediate offset and intermediate cluster center.

[0096] In one embodiment, the image restoration unit is further configured to:

[0097] After replacing the initial membership, initial offset and initial cluster center respectively with the intermediate membership, intermediate offset and intermediate cluster center, corresponding iterative calculation is performed;

[0098] When the current number of iterations is equal to the preset number of iterations, the intermediate offset obtained by the current iteration calculation is used as the target offset of the pixel point.

[0099] In summary, the image restoration system of the embodiment of the present invention first determines the distance function based on the pixels of the original image, the initial offset of the pixels, the initial cluster centers and the initial membership of the domain points corresponding to the pixels with respect to the initial cluster centers, and then determines the intermediate membership based on the distance function, and then determines the target offset of the pixels based on the intermediate membership. The original image can be accurately restored by the target offset of the pixels, which can improve the clarity and accuracy of the image.

[0100] The embodiment of the present invention also provides a specific implementation of a computer device capable of implementing all the steps of the image restoration method in the above embodiment. Fig. 9 is a block diagram of a computer device in an embodiment of the present invention, see Fig. 9 , the computer equipment specifically includes the following contents:

[0101] Processor (processor) 901 and memory (memory) 902 .

[0102] The processor 901 is used to call the computer program in the memory 902. When the processor executes the computer program, all the steps in the image restoration method in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0103] Determine the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center;

[0104] Determine the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center;

[0105] The target offset of the pixel is determined according to the intermediate membership, and the original image is restored by the target offset of the pixel.

[0106] In summary, the computer device of the embodiment of the present invention first determines the distance function based on the pixels of the original image, the initial offset of the pixels, the initial cluster center and the initial membership of the domain points corresponding to the pixels with respect to the initial cluster center, and then determines the intermediate membership based on the distance function, and then determines the target offset of the pixels based on the intermediate membership. The original image can be accurately repaired by the target offset of the pixels, which can improve the clarity and accuracy of the image.

[0107] The embodiment of the present invention also provides a computer-readable storage medium capable of implementing all the steps of the image restoration method in the above embodiment. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all the steps of the image restoration method in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0108] Determine the distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center;

[0109] Determine the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center;

[0110] The target offset of the pixel is determined according to the intermediate membership, and the original image is restored by the target offset of the pixel.

[0111] In summary, the computer-readable storage medium of an embodiment of the present invention first determines a distance function based on the pixels of the original image, the initial offset of the pixels, the initial cluster centers, and the initial memberships of the domain points corresponding to the pixels with respect to the initial cluster centers, and then determines the intermediate membership based on the distance function, and then determines the target offset of the pixels based on the intermediate membership. The original image can be accurately repaired by the target offset of the pixels, thereby improving the clarity and accuracy of the image.

[0112] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0113] Those skilled in the art may also understand that the various illustrative logical blocks, units, and steps listed in the embodiments of the present invention may be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly demonstrate the interchangeability of hardware and software, the various illustrative components, units, and steps described above have generally described their functions. Whether such functions are implemented by hardware or software depends on the specific application and the design requirements of the entire system. Those skilled in the art may use various methods to implement the described functions for each specific application, but such implementation should not be understood as exceeding the scope of protection of the embodiments of the present invention.

[0114] The various illustrative logic blocks, or units, or devices described in the embodiments of the present invention can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of the above. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0115] The steps of the method or algorithm described in the embodiments of the present invention can be directly embedded in hardware, a software module executed by a processor, or a combination of the two. The software module can be stored in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM, or other storage media of any form in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and the storage medium can be arranged in an ASIC, and the ASIC can be arranged in a user terminal. Optionally, the processor and the storage medium can also be arranged in different components in the user terminal.

[0116] In one or more exemplary designs, the above functions described in the embodiments of the present invention can be implemented in hardware, software, firmware or any combination of the three. If implemented in software, these functions can be stored on a computer-readable medium, or transmitted in the form of one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media that facilitate the transfer of computer programs from one place to another. The storage medium can be any available medium that can be accessed by any general or special computer. For example, such computer-readable media can include but are not limited to RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store program codes in the form of instructions or data structures and other forms that can be read by general or special computers, or general or special processors. In addition, any connection can be appropriately defined as a computer-readable medium, for example, if the software is transmitted from a website site, server or other remote resource through a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) or wirelessly, such as infrared, wireless and microwave, it is also included in the defined computer-readable medium. The disk and disc include compact disk, laser disk, optical disk, DVD, floppy disk and blue-ray disk. Disks usually copy data magnetically, while discs usually copy data optically with lasers. The above combination can also be included in computer readable media.

Claims

1. An image restoration method, characterized in that: include: Determine a distance function from the pixel point to the initial cluster center according to the pixel point of the original image, the initial offset of the pixel point, the initial cluster center and the initial membership of the domain point corresponding to the pixel point with respect to the initial cluster center; Determining the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center; Determine a target offset of a pixel point according to the intermediate membership, and repair the original image by using the target offset of the pixel point; Determining the target offset of the pixel point according to the intermediate membership degree includes: Determining an intermediate offset of a pixel point according to the intermediate membership and the initial cluster center; Determine a target offset of the pixel point according to the intermediate offset; Determining the target offset of the pixel point according to the intermediate offset also includes: Determining an intermediate cluster center according to the intermediate offset and the intermediate membership; Determining a target offset of a pixel point by using the intermediate membership, the intermediate offset and the intermediate cluster center; The intermediate cluster center is determined by the following formula: Among them, v' j is the jth intermediate cluster center, (u(yi))max is the i-th pixel y i The maximum value of the intermediate membership about the initial cluster centers.

2. The image restoration method according to claim 1, characterized in that: Determining the target offset of the pixel point by using the intermediate membership, the intermediate offset and the intermediate cluster center also includes: After replacing the initial membership degree, the initial offset and the initial cluster center respectively with the intermediate membership degree, the intermediate offset and the intermediate cluster center, corresponding iterative calculation is performed; When the current number of iterations is equal to the preset number of iterations, the intermediate offset obtained by the current iteration calculation is used as the target offset of the pixel point.

3. An image restoration system, characterized in that: include: A distance function determination unit, configured to determine a distance function from a pixel point to an initial cluster center according to a pixel point of an original image, an initial offset of the pixel point, an initial cluster center, and an initial membership degree of a domain point corresponding to the pixel point with respect to the initial cluster center; An intermediate membership determination unit, used to determine the intermediate membership of the pixel point with respect to the initial cluster center according to the distance function from the pixel point to the initial cluster center; An image restoration unit, configured to determine a target offset of a pixel point according to the intermediate membership degree, and restore the original image by using the target offset of the pixel point; The image restoration unit is specifically used for: Determining an intermediate offset of a pixel point according to the intermediate membership and the initial cluster center; Determine a target offset of the pixel point according to the intermediate offset; The image restoration unit is also used for: Determining an intermediate cluster center according to the intermediate offset and the intermediate membership; Determining a target offset of a pixel point by using the intermediate membership, the intermediate offset and the intermediate cluster center; The intermediate cluster center is determined by the following formula: Among them, v' j is the jth intermediate cluster center, (u(yi))max is the i-th pixel y i The maximum value of the intermediate membership about the initial cluster centers.

4. The image restoration system according to claim 3, characterized in that: The image restoration unit is also used for: After replacing the initial membership degree, the initial offset and the initial cluster center respectively with the intermediate membership degree, the intermediate offset and the intermediate cluster center, corresponding iterative calculation is performed; When the current number of iterations is equal to the preset number of iterations, the intermediate offset obtained by the current iteration calculation is used as the target offset of the pixel point.

5. A computer device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the image restoration method according to any one of claims 1 to 2 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the image restoration method according to any one of claims 1 to 2 are implemented.

Images