Light field camera correction method and device, electronic equipment and storage medium

Abstract

The invention provides a light field camera correction method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining a test image shot by a target light field camera in a test scene and a reference image generated by simulation according to the test image; determining a target correction matrix of the target light field camera according to the difference between the test image and the reference image; and correcting an initial image shot by the target light field camera in an actual scene based on the target correction matrix to obtain a corrected target image. According to the method, the reference image is constructed as the reference standard for testing image correction, the pixel offset condition of the light field camera can be accurately reflected, the method has the advantages of being efficient, accurate and high in robustness, the automation degree is high, the labor cost and the time cost of industrial application can be remarkably reduced, and the method is suitable for popularization and application. And a powerful guarantee is provided for wide application of the light field camera.

Description

Technical Field

[0001] The present application relates to the field of image processing technology. Specifically, it relates to a method, device, electronic device and storage medium for correcting a light field camera. Background Art

[0002] A light field camera is a camera that can record the direction and amount of light. It can not only capture the light amount information in the light field, but also record the direction information of the light, so as to achieve focusing on any depth of the scene.

[0003] Due to the optical imperfections of the main lens and microlenses of the light field camera and the deviation during the assembly process, the captured light field image cannot present an ideal state, specifically manifested as pixel offset on the image. Furthermore, it affects the conversion process of light field information from the macro-pixel space to the sub-aperture space and the epipolar plane space. In view of this problem, the existing technology usually only processes the light field image as the object. This processing method not only requires a large amount of manual participation, but also requires a large number of trial processes, with low correction efficiency and unsatisfactory results. Summary of the Invention

[0004] In view of this, the purpose of the present application is to provide a method, device, electronic device and storage medium for correcting a light field camera to overcome the problems in the existing technology.

[0005] In a first aspect, an embodiment of the present application provides a method for correcting a light field camera, the method includes:

[0006] Obtain a test image captured by a target light field camera in a test scene and a reference image simulated according to the test image;

[0007] Determine a target correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0008] Based on the target correction matrix, correct an initial image captured by the target light field camera in an actual scene to obtain a corrected target image.

[0009] In some technical solutions of the present application, the above determining the target correction matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0010] Based on a first method, determine a first correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0011] Based on a second method, determine a second correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0012] Based on the first correction matrix and the second correction matrix, obtain the target correction matrix of the target light field camera.

[0013] In some technical solutions of the present application, based on the first method, determining the first correction matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0014] Generate an initial correction matrix corresponding to the size of the test image according to the test image;

[0015] Adjust the initial correction matrix according to the difference between the test image and the reference image to obtain the first correction matrix.

[0016] In some technical solutions of the present application, after generating the initial correction matrix, the method further includes:

[0017] Shrink the initial correction matrix according to a preset magnification to obtain a processed correction matrix to be adjusted;

[0018] The adjusting the initial correction matrix according to the difference between the test image and the reference image to obtain the first correction matrix includes:

[0019] Obtain a to-be-corrected image according to the test image and the correction matrix to be adjusted;

[0020] Adjust the initial correction matrix according to the loss calculation result between the to-be-corrected image and the reference image to obtain the first correction matrix.

[0021] In some technical solutions of the present application, the obtaining a to-be-corrected image according to the test image and the correction matrix to be adjusted includes:

[0022] Use a preset sampling algorithm to upsample the correction matrix to be adjusted to the size of the test image to obtain the to-be-corrected image;

[0023] The adjusting the initial correction matrix according to the loss calculation result between the to-be-corrected image and the reference image to obtain the first correction matrix includes:

[0024] Use a preset adjustment algorithm to adjust the initial correction matrix based on the loss calculation result to obtain the first correction matrix.

[0025] In some technical solutions of the present application, based on the second method, determining the second correction matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0026] Input the test image and the reference image into a preset prediction model to obtain a second calibration matrix of the target light field camera output by the prediction model.

[0027] In some technical solutions of the present application, the number of the above prediction models is multiple. The step of inputting the test image and the reference image into the preset prediction model to obtain the second calibration matrix of the target light field camera output by the prediction model includes:

[0028] Input the test image and the reference image into each of the prediction models to obtain candidate calibration matrices output by each of the prediction models;

[0029] Obtain the second calibration matrix of the target light field camera output by the prediction model according to the candidate calibration matrices.

[0030] In a second aspect, an embodiment of the present application provides an apparatus for calibrating a light field camera. The apparatus includes:

[0031] An acquisition module, configured to acquire a test image captured by a target light field camera in a test scenario and a reference image simulated according to the test image;

[0032] A determination module, configured to determine a target calibration matrix of the target light field camera according to a difference between the test image and the reference image;

[0033] A calibration module, configured to calibrate an initial image captured by the target light field camera in an actual scenario based on the target calibration matrix to obtain a calibrated target image.

[0034] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method for calibrating a light field camera are implemented.

[0035] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the steps of the above method for calibrating a light field camera are executed.

[0036] The technical solutions provided by the embodiments of the present application may include the following beneficial effects:

[0037] The method of the present application includes obtaining a test image captured by a target light field camera in a test scenario and a reference image simulated based on the test image; determining a target correction matrix of the target light field camera according to the difference between the test image and the reference image; and correcting an initial image captured by the target light field camera in an actual scenario based on the target correction matrix to obtain a corrected target image.

[0038] The present application constructs a reference image as a reference benchmark for correcting the test image, which can accurately reflect the pixel offset of the light field camera, and has the characteristics of high efficiency, accuracy, and high robustness. Moreover, it has a high degree of automation, can significantly reduce the labor cost and time cost of industrial applications, and provides a strong guarantee for the wide application of the light field camera.

[0039] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specifically gives preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0041] Figure 1 Shows a schematic flowchart of a method for correcting a light field camera provided by an embodiment of the present application;

[0042] Figure 2 Shows a schematic diagram of obtaining a first correction matrix provided by an embodiment of the present application;

[0043] Figure 3 Shows a schematic diagram of a device for correcting a light field camera provided by an embodiment of the present application;

[0044] Figure 4 Is a schematic structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. It should be understood that the accompanying drawings in this application are only for the purposes of illustration and description, and are not used to limit the protection scope of this application. Additionally, it should be understood that the schematic drawings are not drawn to actual scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowchart may not be implemented in sequence, and steps without a logical context relationship may be reversed or implemented simultaneously. Furthermore, those skilled in the art may add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of this application.

[0046] In addition, the described embodiments are only some embodiments of this application, rather than all of the embodiments. The components of the embodiments of this application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of this application claimed, but merely represents selected embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the protection scope of this application.

[0047] It should be noted that the term "including" will be used in the embodiments of this application to indicate the existence of the features stated thereafter, but does not exclude adding other features.

[0048] A light field camera is a camera that can record the direction and amount of light. It can not only capture the light amount information in the light field, but also record the direction information of the light, thereby achieving focusing on any depth of the scene. The light field camera adds an array full of micro lenses between the main lens and the sensor. Each micro lens receives light and projects it onto the sensor. This design enables the camera to record the landing point information of light at different distances and refocus digitally to generate clear photos.

[0049] Due to the optical imperfections of the main lens and micro lenses of the light field camera and the deviations during the assembly process, the captured light field images cannot present an ideal state, specifically manifested as pixel offsets in the images. This further affects the conversion process of light field information from the macro pixel space to the sub-aperture space and the epipolar plane space. To address this problem, the existing technology usually only processes the light field images as the object. This processing method not only requires a relatively large amount of manual participation, but also requires a large number of trial processes, with low correction efficiency and unsatisfactory results.

[0050] Based on this, the embodiments of the present application provide a method, an apparatus, an electronic device, and a storage medium for correcting a light field camera, which will be described below through embodiments.

[0051] Figure 1 The flowchart of a method for correcting a light field camera provided by an embodiment of the present application is shown. Among them, the method includes steps S101 - S103; specifically:

[0052] S101. Obtain a test image captured by a target light field camera in a test scenario and a reference image simulated based on the test image;

[0053] S102. Determine a target correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0054] S103. Correct an initial image captured by the target light field camera in an actual scenario based on the target correction matrix to obtain a corrected target image.

[0055] The present application constructs a reference image as a reference benchmark for correcting the test image, which can accurately reflect the pixel offset of the light field camera, and has the characteristics of high efficiency, accuracy, and robustness. Moreover, it has a high degree of automation and can significantly reduce the labor cost and time cost of industrial applications, providing a strong guarantee for the wide application of the light field camera.

[0056] Some embodiments of the present application will be described in detail below. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0057] For ease of description, the light field camera to be corrected in the embodiments of the present application is referred to as the target light field camera. In order to improve the accuracy of correcting the target light field camera, a reference image is set for the target light field camera in the embodiments of the present application, and then the target light field camera is corrected based on the captured image and the reference image of the target light field camera.

[0058] When correcting the target light field camera according to the target light field camera and the reference image, for the sake of improving efficiency, the embodiments of the present application place the target light field camera in a test scenario for shooting, and the image captured in the test scenario is called a test image. That is, the embodiments of the present application correct the target light field camera based on the test image and the reference image.

[0059] Further, to improve the calibration efficiency, the test image captured in the test scenario in the embodiments of the present application is a white image. Here, the white image is characterized as an image that does not contain any objects and is white. For example, it can be ensured that the camera lens is clean, the ambient light is uniform and is white light. The target light field camera is fixed on a tripod to avoid shaking, and the camera parameters (such as ISO, shutter speed, aperture size) are set. Then, it is aligned with the whiteboard for shooting to obtain a clear test image. After obtaining the test image, the embodiments of the present application also need to determine the reference image of the test image. The embodiments of the present application generate the reference image by simulation. When performing the simulation, if the test image contains any objects, it will increase the difficulty of the simulation. Specifically, the camera parameters of the target light field camera are obtained, and then the camera parameters are input into a preset optical simulation software to simulate the imaging system of the light field camera, and a white image is generated under preset ideal conditions, and the white image is used as the reference image. If there is no optical simulation software, the embodiments of the present application can also generate a white image by cropping the microlenses in the central field of view of the captured image as a template, and then use the white image as the reference image. Of course, if both of the above two methods can be implemented, the present application can also combine the above two methods. For example, select the white image with better effect from the first white image (simulated generated) and the second white image (generated with the microlens as a template) as the reference image; the first white image and the second white image can also be superimposed or fused as the reference image, etc.

[0060] After obtaining the test image and the reference image, the embodiments of the present application also need to determine the difference between the two, and then determine the target calibration matrix of the target light field camera according to the difference between the two; after determining the target calibration matrix, the initial image captured by the target light field camera in the actual scenario is calibrated based on the target calibration matrix to obtain the calibrated target image.

[0061] When determining the target calibration matrix, to improve the accuracy, the embodiments of the present application use multiple different methods. Specifically: Based on the first method, according to the difference between the test image and the reference image, determine the first calibration matrix of the target light field camera; based on the second method, according to the difference between the test image and the reference image, determine the second calibration matrix of the target light field camera; according to the first calibration matrix and the second calibration matrix, obtain the target calibration matrix of the target light field camera.

[0062] For the first method: As Figure 2As shown, in order to obtain the first correction matrix, the embodiment of the present application needs to first generate an initial correction matrix based on the test image (the elements are set to zero, and the physical meaning is the offset of the pixel), and then adjust the initial correction matrix according to the difference between the test image and the reference image. After adjustment, the first correction matrix is obtained. When generating the initial correction matrix, the generation basis here is the size of the test image. That is, the initial correction matrix generated by the present application needs to correspond to the size of the test image. For example, if the size of the test image is 1000 pixels × 800 pixels, the initial correction matrix generated by the present application is 1000 × 800.

[0063] After obtaining the initial correction matrix, if the initial correction matrix is directly stored in the target light field camera for use, it will occupy more memory and consume a large amount of storage time, affecting the correction efficiency. Based on this, after obtaining the initial correction matrix, the embodiment of the present application performs a shrinking process on the initial correction matrix to obtain the adjusted correction matrix to be adjusted. The embodiment of the present application stores the shrunk correction matrix to be adjusted in the target light field camera for subsequent correction processing. In specific implementation, when shrinking the initial correction matrix, it needs to be shrunk according to a preset magnification. The preset magnification here can be determined according to the number of microlenses. For example, the initial correction matrix is divided by the number of microlenses to obtain the correction matrix to be adjusted.

[0064] After obtaining the correction matrix to be adjusted, the test image is deformed based on the correction matrix to be adjusted to obtain the image to be corrected. Then, according to the loss calculation result between the image to be corrected and the reference image, the initial correction matrix is adjusted to obtain the first correction matrix. When obtaining the image to be corrected, the embodiment of the present application can use a preset sampling algorithm, and the sampling algorithm here can be an interpolation algorithm, a bicubic interpolation algorithm, etc. By using the bicubic interpolation algorithm, the correction matrix to be adjusted is upsampled to the same size as the test image; the purpose of this step is to increase the resolution of the correction matrix to be adjusted to the same as that of the test image for pixel-level alignment and correction. Then, according to the loss calculation result between the image to be corrected and the reference image, the initial correction matrix is adjusted to obtain the first correction matrix. Specifically, the gradient descent method can be used to iteratively update the elements of the correction matrix to be adjusted, and then the difference (such as the root mean square error) between the image to be corrected and the reference image is obtained; the correction matrix to be adjusted is adjusted according to the difference until the difference is minimized. This step needs to set an appropriate number of iterations to avoid overfitting or slow convergence. After completing the above steps, the final first correction matrix is output and saved.

[0065] For the second method: To obtain the second correction matrix, the embodiments of the present application need to pre-train a prediction model. For example, a zero-shot training method is adopted, or a large-scale simulation dataset can be used, and a supervised neural network method can also be used. To improve the accuracy, the embodiments of the present application obtain multiple prediction models through the above different methods, input the test image and the reference image into each prediction model respectively, and then obtain the candidate correction matrices output by each prediction model. The second correction matrix is obtained by evaluating the effects of each candidate correction matrix or by fusing multiple candidate correction matrices.

[0066] After obtaining the first correction matrix and the second correction matrix, the target correction matrix can be selected from the first correction matrix and the second correction matrix, or the first correction matrix and the second correction matrix can be fused into the target correction matrix. This target matrix can be used as the light field correction matrix of the camera, and subsequent images captured by the camera can be corrected through this matrix.

[0067] The present application can accurately and adaptively correct the imperfections existing in the imaging process of both focused and non-focused light field cameras, effectively solving the problems of the traditional manual correction method being cumbersome, time-consuming, laborious, and difficult to guarantee the correction accuracy. Compared with the traditional method, this method not only has higher correction accuracy, can adapt to the significant difference in imaging quality between the central field of view and the edge field of view commonly existing in optical lenses, can flexibly cope with light field cameras with various different parameters, and greatly improves the efficiency and accuracy of the light field camera during the calibration process.

[0068] Figure 3 The structural schematic diagram of a device for correcting a light field camera provided by the embodiments of the present application is shown. The device includes:

[0069] An acquisition module, configured to acquire a test image captured by a target light field camera in a test scenario and a reference image simulated according to the test image;

[0070] A determination module, configured to determine the target correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0071] A correction module, configured to correct an initial image captured by the target light field camera in an actual scenario based on the target correction matrix to obtain a corrected target image.

[0072] Determining the target correction matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0073] Based on the first method, determining the first correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0074] Based on the second method, determine the second correction matrix of the target light field camera according to the difference between the test image and the reference image;

[0075] Obtain the target correction matrix of the target light field camera according to the first correction matrix and the second correction matrix.

[0076] The method based on the first method for determining the first correction matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0077] Generate an initial correction matrix corresponding to the size of the test image according to the test image;

[0078] Adjust the initial correction matrix according to the difference between the test image and the reference image to obtain the first correction matrix.

[0079] After generating the initial correction matrix, it further includes:

[0080] Shrink the initial correction matrix according to a preset magnification factor to obtain a processed correction matrix to be adjusted;

[0081] The adjusting the initial correction matrix according to the difference between the test image and the reference image to obtain the first correction matrix includes:

[0082] Obtain a to-be-corrected image according to the test image and the correction matrix to be adjusted;

[0083] Adjust the initial correction matrix according to the loss calculation result between the to-be-corrected image and the reference image to obtain the first correction matrix.

[0084] The obtaining a to-be-corrected image according to the test image and the correction matrix to be adjusted includes:

[0085] Use a preset sampling algorithm to upsample the correction matrix to be adjusted to the size of the test image to obtain the to-be-corrected image;

[0086] The adjusting the initial correction matrix according to the loss calculation result between the to-be-corrected image and the reference image to obtain the first correction matrix includes:

[0087] Use a preset adjustment algorithm to adjust the initial correction matrix based on the loss calculation result to obtain the first correction matrix.

[0088] Based on the second method, determining a second calibration matrix of the target light field camera according to the difference between the test image and the reference image includes:

[0089] Inputting the test image and the reference image into a preset prediction model, and obtaining the second calibration matrix of the target light field camera output by the prediction model.

[0090] There are multiple prediction models. The step of inputting the test image and the reference image into a preset prediction model to obtain the second calibration matrix of the target light field camera output by the prediction model includes:

[0091] Inputting the test image and the reference image into each of the prediction models to obtain candidate calibration matrices output by each of the prediction models;

[0092] Obtaining the second calibration matrix of the target light field camera output by the prediction model according to the candidate calibration matrices.

[0093] As Figure 4 shown, an embodiment of the present application provides an electronic device for executing the method for calibrating a light field camera in the present application. The device includes a memory, a processor, a bus, and a computer program stored on the memory and executable on the processor. When the above-mentioned processor executes the above-mentioned computer program, the steps of the above-mentioned method for calibrating a light field camera are implemented.

[0094] Specifically, the above-mentioned memory and processor can be general-purpose memory and processor, and no specific limitation is made here. When the processor runs the computer program stored in the memory, it can execute the above-mentioned method for calibrating a light field camera.

[0095] Corresponding to the method for calibrating a light field camera in the present application, an embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the above-mentioned method for calibrating a light field camera are executed.

[0096] Specifically, the storage medium can be a general-purpose storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, it can execute the above-mentioned method for calibrating a light field camera.

[0097] In the embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed among each other can be through some communication interfaces. The indirect couplings or communication connections of the systems or units can be in electrical, mechanical or other forms.

[0098] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0099] In addition, each functional unit in the embodiments provided in the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0100] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0101] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0102] Finally, it should be noted that the above-described embodiments are only specific implementation manners of the present application, used to illustrate the technical solutions of the present application, rather than limiting it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions recorded in the foregoing embodiments or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for light field camera correction, characterized in that: The method comprises: Acquire a test image captured by a target light field camera in a test scene and a reference image simulated and generated according to the test image; determining a target correction matrix for the target light field camera according to a difference between the test image and the reference image; The initial image captured by the target light field camera in the actual scene is corrected based on the target correction matrix to obtain a corrected target image.

2. The method according to claim 1, characterized in that The step of determining a target correction matrix of the target light field camera according to a difference between the test image and the reference image comprises: Based on the first method, determining a first correction matrix of the target light field camera according to a difference between the test image and the reference image; Based on the second method, determining a second correction matrix of the target light field camera according to the difference between the test image and the reference image; A target correction matrix of the target light field camera is obtained according to the first correction matrix and the second correction matrix.

3. The method according to claim 2, characterized in that The method of determining a first correction matrix of the target light field camera based on the first method according to the difference between the test image and the reference image includes: According to the test image, generating an initial correction matrix corresponding to the size of the test image; The initial correction matrix is ​​adjusted according to the difference between the test image and the reference image to obtain the first correction matrix.

4. The method according to claim 3, characterized in that After generating the initial correction matrix, the method further includes: The initial correction matrix is ​​reduced according to a preset magnification to obtain a processed correction matrix to be adjusted; The step of adjusting the initial correction matrix according to the difference between the test image and the reference image to obtain the first correction matrix includes: Obtaining an image to be corrected according to the test image and the correction matrix to be adjusted; The initial correction matrix is ​​adjusted according to the loss calculation result between the image to be corrected and the reference image to obtain the first correction matrix.

5. The method according to claim 4, characterized in that The step of obtaining the image to be corrected according to the test image and the correction matrix to be adjusted includes: Using a preset sampling algorithm, up-sampling the correction matrix to be adjusted to the size of the test image to obtain the image to be corrected; The step of adjusting the initial correction matrix according to the loss calculation result between the image to be corrected and the reference image to obtain the first correction matrix includes: The initial correction matrix is ​​adjusted based on the loss calculation result using a preset adjustment algorithm to obtain the first correction matrix.

6. The method according to claim 2, characterized in that The method of determining a second correction matrix of the target light field camera based on the difference between the test image and the reference image based on the second method includes: The test image and the reference image are input into a preset prediction model to obtain a second correction matrix of the target light field camera output by the prediction model.

7. The method according to claim 6, characterized in that The number of the prediction models is multiple, and the test image and the reference image are input into a preset prediction model to obtain a second correction matrix of the target light field camera output by the prediction model, including: Inputting the test image and the reference image into each of the prediction models to obtain a correction matrix to be selected output by each of the prediction models; According to the selected correction matrix, a second correction matrix of the target light field camera output by the prediction model is obtained.

8. A device for light field camera correction, characterized in that: The device comprises: An acquisition module is used to acquire a test image captured by a target light field camera in a test scene and a reference image simulated and generated according to the test image; a determination module, configured to determine a target correction matrix of the target light field camera according to a difference between the test image and the reference image; A correction module is used to correct the initial image taken by the target light field camera in the actual scene based on the target correction matrix to obtain a corrected target image.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for light field camera correction according to any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for correcting a light field camera according to any one of claims 1 to 7 are executed.

Images