Method and device for measuring calibration accuracy of camera external parameters, and electronic device

Abstract

The application is suitable for the field of camera calibration evaluation, and provides a camera extrinsic parameter calibration precision measurement method and device and electronic equipment, comprising: calibrating first intrinsic parameters and first distortion parameters of a first camera in a vision system, and calibrating second intrinsic parameters and second distortion parameters of a second camera in the vision system; calibrating extrinsic parameters between the first camera and the second camera; after the first camera and the second camera capture images of a same calibration board, pixel coordinates of corner points in each calibration board image are extracted respectively; according to the first intrinsic parameters, the first distortion parameters, the second intrinsic parameters, the second distortion parameters, the extrinsic parameters and the pixel coordinates of the corner points in the calibration board image, a spatial distance error between two pixel coordinates corresponding to a same corner point is calculated; and according to each spatial distance error and a preset spatial distance threshold, the calibration precision of the extrinsic parameters is measured. Through the above method, the accuracy of the measurement result can be improved.

Description

Technical Field

[0001] This application belongs to the field of camera calibration and evaluation technology, and in particular relates to methods, devices, electronic equipment, computer-readable storage media and computer program products for measuring the accuracy of camera extrinsic parameter calibration. Background Technology

[0002] In scenarios requiring depth perception or multi-view information, binocular or multi-view vision systems are typically used. To ensure the accuracy of spatial measurements by binocular or multi-view vision systems, extrinsic parameter calibration is required, which involves calibrating the rotation matrices and translation vectors of each camera in the system.

[0003] Currently, pixel distance error is usually calculated by relying on reprojection error, and then compared with a preset pixel error threshold. The comparison result is used to determine whether the camera's extrinsic calibration results meet the requirements.

[0004] However, when using this method to measure the accuracy of camera extrinsic calibration, the accuracy of the measurement results is low in some scenarios. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for measuring the calibration accuracy of camera extrinsic parameters, which can solve the problem of low measurement accuracy of existing extrinsic parameter calibration.

[0006] In a first aspect, embodiments of this application provide a method for measuring the accuracy of camera extrinsic parameter calibration, including: The first intrinsic parameter and the first distortion parameter of the first camera in the vision system are calibrated, and the second intrinsic parameter and the second distortion parameter of the second camera in the vision system are calibrated. Calibrate the extrinsic parameters between the first camera and the second camera; After the first camera and the second camera capture images of the same calibration board, the corner pixel coordinates of each calibration board image are extracted. Based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration board image, calculate the spatial distance error between the two corner pixel coordinates corresponding to the same corner point; The calibration accuracy of the external parameters is measured based on the spatial distance errors and the preset spatial distance threshold.

[0007] The beneficial effects of the embodiments in this application compared with the prior art are: In this embodiment, after the first and second cameras in the vision system capture images of the same calibration board, the corner pixel coordinates in each calibration board image are extracted. Based on the extracted corner pixel coordinates, the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter obtained from calibrating the first and second cameras, and the extrinsic parameters between the first and second cameras, the spatial distance error between the two corner pixel coordinates corresponding to the same corner point is calculated. Then, the calibration accuracy of the extrinsic parameters between the first and second cameras is measured based on each spatial distance error and a preset spatial distance threshold. Since the aforementioned spatial distance error is calculated jointly based on the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter, and the extrinsic parameters between the first and second cameras, and the probability of changes in the corresponding first intrinsic parameter, first distortion parameter, second intrinsic parameter, and second distortion parameter after the first and second cameras leave the factory is low, the aforementioned spatial distance error can reflect the calibration accuracy of the extrinsic parameters between the first and second cameras. Therefore, the calibration accuracy of the extrinsic parameters between the first and second cameras can be accurately measured based on each spatial distance error and the preset spatial distance threshold. Furthermore, since the aforementioned spatial distance error reflects the accuracy of the extrinsic parameters in real space, the measurement result of the calibration accuracy of the extrinsic parameters obtained based on the spatial distance error is not affected by different brands, resolutions, or optical parameters. That is, regardless of whether the cameras in the vision system are homogeneous or heterogeneous, the calibration accuracy of the extrinsic parameters can be accurately measured through spatial distance error.

[0008] Secondly, embodiments of this application provide a measuring device for camera extrinsic parameter calibration accuracy, comprising: An intrinsic parameter calibration module is used to calibrate the first intrinsic parameter and the first distortion parameter of the first camera in the vision system, and to calibrate the second intrinsic parameter and the second distortion parameter of the second camera in the vision system. An extrinsic parameter calibration module is used to calibrate the extrinsic parameters between the first camera and the second camera. The corner pixel coordinate extraction module is used to extract the corner pixel coordinates in each calibration board image after the first camera and the second camera take pictures of the same calibration board. The spatial distance error calculation module is used to calculate the spatial distance error between two corner pixel coordinates corresponding to the same corner point based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration board image. The calibration accuracy measurement module is used to measure the calibration accuracy of the external parameters based on the spatial distance errors and preset spatial distance thresholds.

[0009] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0010] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0011] Fifthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in the first aspect.

[0012] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0014] Figure 1 This is a flowchart illustrating a method for measuring the accuracy of camera extrinsic parameter calibration provided in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a camera extrinsic parameter calibration accuracy measurement device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0015] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0016] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0017] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0018] Furthermore, in the description of this application and the appended claims, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0020] Currently, before using a binocular or multi-view vision system for spatial measurement, it is necessary to calibrate the external parameters of the camera in the binocular or multi-view vision system.

[0021] After calibrating the camera's extrinsic parameters, it is necessary to measure the accuracy of the calibration. When measuring the accuracy of the extrinsic parameter calibration by means of reprojection error, it is usually done by calculating the pixel distance error at the same corner point and then comparing the pixel distance error with a preset pixel error threshold. If the pixel distance error is less than the preset pixel error threshold, the camera's extrinsic parameter calibration result is considered to meet the requirements; otherwise, the camera's extrinsic parameter calibration result is considered to not meet the requirements.

[0022] Because heterogeneous cameras may differ in resolution, brand, model, and other parameters, the physical size corresponding to a single pixel varies. This means that the same pixel error does not represent an equivalent physical distance on different cameras: on a low-resolution camera, a single pixel error may correspond to a larger physical distance; while on a high-resolution camera, a single pixel error may correspond to a smaller physical distance. Therefore, directly using pixel distance error to determine whether the extrinsic parameter calibration results of heterogeneous cameras meet the requirements cannot establish a unified evaluation standard, thus failing to fairly and consistently reflect the true physical space calibration accuracy, ultimately leading to reduced accuracy of the measurement results.

[0023] To improve the accuracy of measurement results, this application provides a method for measuring the accuracy of camera extrinsic parameter calibration.

[0024] In this measurement method, spatial distance error is used to measure the calibration accuracy of extrinsic parameters. Since spatial distance error can measure the spatial consistency of extrinsic parameter results, it is beneficial to improve the accuracy of the measurement results of extrinsic parameter calibration accuracy.

[0025] The method for measuring the accuracy of camera extrinsic parameter calibration provided in the embodiments of this application is described below with reference to the accompanying drawings.

[0026] Figure 1 This illustration shows a flowchart of a method for measuring the accuracy of camera extrinsic parameter calibration according to an embodiment of this application. This method can be applied to electronic devices, including mobile devices, workstations, computers, servers, etc. The method for measuring the accuracy of camera extrinsic parameter calibration provided in this embodiment is described in detail below: S11, calibrate the first intrinsic parameter and the first distortion parameter of the first camera in the vision system, and calibrate the second intrinsic parameter and the second distortion parameter of the second camera in the vision system.

[0027] The aforementioned vision system includes at least two cameras. When the vision system includes more than two cameras, the two cameras can be grouped together, and the two cameras in different groups can be calibrated separately.

[0028] To differentiate the cameras, the two calibrated cameras are named Camera 1 and Camera 2, respectively. The intrinsic parameters of Camera 1 and Camera 2 are named Intrinsic Parameter 1 and Intrinsic Parameter 2, respectively. The distortion parameters of Camera 1 and Camera 2 are named Distortion Parameter 1 and Distortion Parameter 2, respectively. These intrinsic parameters include the camera's focal length and principal point.

[0029] In this embodiment, the first intrinsic parameter of the first camera and the second intrinsic parameter of the second camera can be calibrated using methods such as checkerboard calibration, dot calibration, and Zhang Zhengyou calibration; the first distortion parameter of the first camera and the second distortion parameter of the second camera can be calibrated using models such as Brown-Conrady model, extended polynomial model, and B-spline distortion model, without limitation here.

[0030] S12, calibrate the external parameters between the first camera and the second camera.

[0031] The extrinsic parameters between the first camera and the second camera refer to the rigid body transformation relationship between the camera coordinate system where the first camera is located and the camera coordinate system where the second camera is located. It is usually described by a rotation matrix R and a translation vector t, which are used to represent the rotation and translation required to transform a point in one camera coordinate system to another camera coordinate system.

[0032] In this embodiment of the application, after the first camera and the second camera are installed and fixed according to their usage requirements, the external parameters between the first camera and the second camera can be calibrated using methods such as the Perspective-n-Point (PnP) method and the external parameter calibration method based on the calibration plate.

[0033] S13, after the first camera and the second camera take pictures of the same calibration board to obtain calibration board images, the corner pixel coordinates in each of the calibration board images are extracted respectively.

[0034] The aforementioned calibration plates include: checkerboard calibration plates, dot calibration plates, concentric ring calibration plates, stripe calibration plates, etc. In this embodiment, the corresponding calibration plate can be selected according to the actual situation.

[0035] In this embodiment, the clock signals of the first camera and the second camera can be synchronized first, and then the first camera and the second camera can be controlled to simultaneously take pictures of the same calibration board, so as to reduce the influence of environmental factors on the obtained calibration board image and thus improve the accuracy of the subsequent measurement results.

[0036] After capturing images of the calibration board, the electronic device can communicate with the vision system to acquire images of the calibration board captured by the first camera and the second camera, and extract the corner pixel coordinates from each calibration board image. These corner pixel coordinates are the pixel coordinates corresponding to corner points in the calibration board image. A corner point (or a corner on the calibration board) refers to a pixel location in the calibration board image where the grayscale value changes drastically. For example, in a dotted calibration board, the corner points are the boundary points between the circle and the background on the dotted calibration board.

[0037] In this embodiment, the electronic device can extract the corner pixel coordinates of each corner point in the calibration board image captured by the first camera, or it can extract the corner pixel coordinates of some corner points in the calibration board image captured by the first camera. Similarly, the electronic device can also extract the corner pixel coordinates of each corner point in the calibration board image captured by the second camera, or it can extract the corner pixel coordinates of some corner points in the calibration board image captured by the second camera.

[0038] In some embodiments, considering that the field of view of the first camera and the field of view of the second camera may not be exactly the same, that is, the corner points in the calibration board image captured by the first camera may not be exactly the same as the corner points in the calibration board image captured by the second camera, therefore, the common viewing area of ​​the first camera and the second camera can be determined first, and then the corner point pixel coordinates can be extracted from the area corresponding to the common viewing area in each calibration board image. At this time, the above-mentioned extraction of corner point pixel coordinates from each of the calibration board images after the first camera and the second camera capture images of the same calibration board includes: After the first camera and the second camera take pictures of the same calibration board to obtain two images of the calibration board, the common viewing area between the two calibration board images is determined. Extract the corner pixel coordinates corresponding to the common viewing area from the two calibration board images mentioned above.

[0039] The aforementioned common viewing area refers to the imaging area of ​​the calibration board region simultaneously located within the overlapping field of view of the first and second cameras, as shown in the two calibration board images. If corner A and corner B of the calibration board are both located within this overlapping field of view, then the common viewing area of ​​the two calibration board images respectively contains the pixel information corresponding to corner A and corner B.

[0040] In this embodiment of the application, the common viewing area between two calibration board images can be determined based on corner matching method, ArUco encoding and marking method, etc., which is not limited here.

[0041] After determining the common viewing area, corner point detection is performed within the region corresponding to the common viewing area in the calibration board image captured by the first camera, and the corner pixel coordinates of the detected corner points are determined. Similarly, the corner pixel coordinates of the corner points detected within the region corresponding to the common viewing area in the calibration board image captured by the second camera are determined.

[0042] Alternatively, the corner pixel coordinates can be determined in the following way: The common viewing area is cropped from the calibration board image to obtain the Region of Interest (ROI). A corner detection algorithm is used within the ROI to detect candidate corner positions. A sub-pixel precision algorithm is used to iteratively optimize each candidate corner position to improve the positioning accuracy to the sub-pixel level. Based on the iteratively optimized candidate corner positions and the offset of the ROI in the calibration board image, the iteratively optimized candidate corner positions are mapped back to the image coordinate system of the calibration board image to obtain the corner pixel coordinates.

[0043] Optionally, considering that corner detection is usually performed on grayscale images, the calibration board image can be converted to a grayscale image before cropping the common viewing area.

[0044] Because corner points are searched within a shared field of view, computational complexity is reduced, and search efficiency is improved. Simultaneously, searching within this region eliminates interfering background, textures, or irrelevant markers outside the shared field of view, thus enhancing detection reliability. Furthermore, since the shared field of view is the imaging region of the calibration board area within the overlapping field of view of the first and second cameras, corresponding to the areas in the two calibration board images, corner points extracted within this region are physically visible between the first and second cameras. This facilitates the subsequent establishment of the correspondence between corner points between the first and second cameras.

[0045] S14. Based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration plate image, calculate the spatial distance error between the two corner pixel coordinates corresponding to the same corner.

[0046] In this embodiment, the corner pixel coordinates can be converted from two-dimensional data in the image coordinate system to three-dimensional data in the camera coordinate system using intrinsic parameters and distortion parameters. Then, the two three-dimensional data are converted to the same camera coordinate system and subtracted according to extrinsic parameters to obtain the aforementioned spatial distance error.

[0047] Specifically: A1. Based on the first intrinsic parameter and the first distortion parameter, the pixel coordinates of the corner point corresponding to the first camera are back-projected onto the camera coordinate system of the first camera to obtain the first three-dimensional coordinates in the first camera coordinate system.

[0048] In this embodiment, it is assumed that the corner pixel coordinates corresponding to the first camera are (u, v). Based on the first distortion parameters of the first camera (including radial distortion and tangential distortion), the corner pixel coordinates (u, v) are subjected to distortion correction processing to obtain the distorted normalized corner pixel coordinates (u′, v′). Based on the first intrinsic parameters of the first camera, the distorted normalized corner pixel coordinates are converted into normalized imaging plane coordinates (u′, v′). , The calculation method is as follows: .

[0049] Given that the calibration plate is located on the same plane (e.g., the Z=0 plane) and the actual physical dimensions of the corner points are known, the first three-dimensional coordinates (X1, Y1, Z1) of the corner points in the calibration plate under the first camera coordinate system can be recovered by PnP solution or homography matrix decomposition.

[0050] Repeat the above steps for the extracted corner pixel coordinates of each first camera to obtain the three-dimensional point set of each corner pixel coordinate in the camera coordinate system.

[0051] A2. Based on the second intrinsic parameter and the second distortion parameter, the corner pixel coordinates corresponding to the second camera are back-projected onto the camera coordinate system of the second camera to obtain the second three-dimensional coordinates in the second camera coordinate system.

[0052] The process of determining the second three-dimensional coordinates is similar to that of determining the first three-dimensional coordinates, and will not be repeated here.

[0053] A3. Based on the above external parameters, determine the relative pose between the first camera and the second camera.

[0054] In this embodiment of the application, the relative pose between the first camera and the second camera is determined based on the rotation matrix and translation vector in the extrinsic parameters between the first camera and the second camera.

[0055] Optionally, the extrinsic parameters between the first camera and the second camera can be determined in the following way: Assume the first external parameter of the first camera relative to the world coordinate system is That is, a point in the world coordinate system Transform to the first camera coordinate system: .

[0056] Assume the second external parameter of the second camera relative to the world coordinate system is That is, a point in the world coordinate system Transform to the second camera coordinate system: .

[0057] in, and Let be a rotation matrix. and It is a translation vector.

[0058] according to and Constructing the homogeneous transformation matrix , used to represent the transformation from the world coordinate system to the first camera coordinate system, to obtain .

[0059] according to and Constructing the homogeneous transformation matrix , used to represent the transformation from the world coordinate system to the second camera coordinate system, to obtain .

[0060] According to the above formula, we can obtain: ,Should It is the extrinsic parameter between the first camera and the second camera (i.e., the extrinsic parameter from the second camera coordinate system to the first camera coordinate system), and it is also used to represent the relative pose between the first camera and the second camera.

[0061] A4. Based on the above relative pose, the first three-dimensional coordinates are transformed from the first camera coordinate system to the second camera coordinate system to obtain the third three-dimensional coordinates; and / or, based on the above relative pose, the second three-dimensional coordinates are transformed from the second camera coordinate system to the first camera coordinate system to obtain the fourth three-dimensional coordinates.

[0062] In this embodiment of the application, using It can transform each of the first three-dimensional coordinates to the second camera coordinate system, or it can transform each of the second three-dimensional coordinates to the first camera coordinate system.

[0063] A5. Based on the aforementioned third three-dimensional coordinates and the aforementioned second three-dimensional coordinates corresponding to the same corner point, determine the first spatial distance error of the aforementioned same corner point, and / or, based on the aforementioned first three-dimensional coordinates and the aforementioned fourth three-dimensional coordinates corresponding to the same corner point, determine the second spatial distance error of the aforementioned same corner point.

[0064] In this embodiment of the application, after calculating the three-dimensional coordinates of each corner point in the first camera coordinate system and the second camera coordinate system respectively, a correspondence relationship between the three-dimensional coordinates in the two coordinate systems is established. That is, the matching relationship between the two three-dimensional coordinates obtained after the same physical corner point is reconstructed by the first camera and the second camera respectively is determined. Then, the absolute value of the difference between each third three-dimensional coordinate and the corresponding second three-dimensional coordinate under the matching relationship is calculated to obtain each first spatial distance error; and / or, the absolute value of the difference between each first three-dimensional coordinate and the corresponding fourth three-dimensional coordinate under the matching relationship is calculated to obtain each second spatial distance error.

[0065] When only the first spatial distance error is calculated, the first spatial distance error is the spatial distance error between the pixel coordinates of the two corner points corresponding to the same corner point; when only the second spatial distance error is calculated, the second spatial distance error is the spatial distance error between the pixel coordinates of the two corner points corresponding to the same corner point; and when both the first spatial distance error and the second spatial distance error are calculated, the spatial distance error between the pixel coordinates of the two corner points corresponding to the same corner point is represented by the first spatial distance error and the second spatial distance error.

[0066] S15, the calibration accuracy of the above external parameters is measured based on each of the above spatial distance errors and the preset spatial distance threshold.

[0067] In this embodiment, at least one spatial evaluation value, such as the mean, maximum value, or standard deviation, can be determined based on each spatial distance error. The determined spatial evaluation value is then compared with a preset spatial distance threshold, and the calibration accuracy of the extrinsic parameters is measured based on the comparison result. It should be noted that the number of preset spatial distance thresholds is equal to the number of determined spatial evaluation values. For example, if the determined spatial evaluation values ​​include the mean and the maximum value, then the preset spatial distance thresholds include both the spatial distance threshold for the mean and the spatial distance threshold for the maximum value.

[0068] When measuring the calibration accuracy of extrinsic parameters based on the mean, the above-mentioned measurement of the calibration accuracy of extrinsic parameters based on each of the aforementioned spatial distance errors and a preset spatial distance threshold includes: Calculate the mean of the first spatial distance error for each of the above-mentioned corner points to obtain the first mean; And / or, Calculate the mean of the second spatial distance error for each of the above-mentioned corner points to obtain the second mean; The fusion mean is determined based on the first mean and / or the second mean mentioned above; The calibration accuracy of the aforementioned extrinsic parameters is measured based on the fusion mean and the preset spatial distance threshold.

[0069] In this embodiment, when the spatial distance error only includes the first spatial distance error, the mean of each first spatial distance error is calculated to obtain the first mean, and the fusion mean is determined based on the first mean, such as using the first mean as the fusion mean. When the spatial distance error only includes the second spatial distance error, the mean of each second spatial distance error is calculated to obtain the second mean, and the fusion mean is determined based on the second mean, such as using the second mean as the fusion mean. When the spatial distance error includes both the first and second spatial distance errors, the mean of each first spatial distance error and the mean of each second spatial distance error are calculated, and then the mean of the first and second means is calculated to obtain the fusion mean. If the fusion mean is less than a preset spatial distance threshold, it can be determined that the calibration accuracy of the extrinsic parameters between the first and second cameras is high; otherwise, it can be determined that the calibration accuracy between the first and second cameras is low.

[0070] When measuring the calibration accuracy of extrinsic parameters based on their maximum values, the above-mentioned measurement of the calibration accuracy of extrinsic parameters based on each of the aforementioned spatial distance errors and a preset spatial distance threshold includes: Determine the maximum value of the first spatial distance error for each of the above-mentioned corner points to obtain the first maximum value; And / or, Determine the maximum value of the second spatial distance error for each of the above-mentioned corner points to obtain the second maximum value; The maximum fusion value is determined based on the first maximum value and / or the second maximum value mentioned above; The calibration accuracy of the above extrinsic parameters is measured based on the maximum fusion value and the preset spatial distance threshold.

[0071] In this embodiment, when the spatial distance error only includes the first spatial distance error, the maximum value of each first spatial distance error is selected to obtain the first maximum value, and the fusion maximum value is determined based on the first maximum value, such as using the first maximum value as the fusion maximum value; when the spatial distance error only includes the second spatial distance error, the maximum value of each second spatial distance error is selected to obtain the second maximum value, and the fusion maximum value is determined based on the second maximum value, such as using the second maximum value as the fusion maximum value; and when the spatial distance error includes both the first and second spatial distance errors, the maximum values ​​of each first spatial distance error and each second spatial distance error are selected, and the larger of the first and second maximum values ​​is determined to obtain the fusion maximum value. If the fusion maximum value is less than a preset spatial distance threshold, it can be determined that the calibration accuracy of the extrinsic parameters between the first and second cameras is high; otherwise, it can be determined that the calibration accuracy of the extrinsic parameters between the first and second cameras is low.

[0072] In this embodiment of the application, the calibration accuracy of the extrinsic parameters can also be measured based on the mean and maximum values.

[0073] For example, suppose This represents the first three-dimensional coordinates corresponding to node i. Let i represent the fourth three-dimensional coordinates corresponding to node i. Then, the absolute value of the difference between the first three-dimensional coordinates and the fourth three-dimensional coordinates (i.e., the spatial distance error corresponding to corner point i) is the difference between the first three-dimensional coordinates and the fourth three-dimensional coordinates. for: Then, the following formulas can be used to calculate respectively. Mean and maximum values: , .

[0074] Where N represents The number of.

[0075] After obtaining the fusion mean and fusion maximum values, the fusion mean is compared with a preset spatial distance mean threshold, and the fusion maximum value is compared with a preset spatial distance maximum value threshold. If the fusion mean is less than the spatial distance mean threshold and the fusion maximum value is less than the spatial distance maximum value threshold, it can be determined that the calibration accuracy of the extrinsic parameters between the first camera and the second camera simultaneously meets the overall accuracy requirements and the local accuracy requirements, and the calibration result is reliable. If the fusion mean is greater than or equal to the spatial distance mean threshold, it is determined that the calibration accuracy is substandard and the extrinsic parameter calibration needs to be repeated. Alternatively, if the fusion maximum value is greater than or equal to the spatial distance maximum value threshold, it is determined that the calibration accuracy is substandard and the extrinsic parameter calibration needs to be repeated.

[0076] In this embodiment, the preset spatial distance threshold can be set to a fixed value or dynamically adjusted according to actual conditions. For example, the preset spatial distance threshold can be determined based on the spatial distance required by the project. In this case, before measuring the calibration accuracy of the extrinsic parameters between the first camera and the second camera based on the various spatial distance errors and the preset spatial distance threshold, the method further includes: Based on the requirements for spatial distance measurement accuracy of the project in which the aforementioned vision system is located, the aforementioned spatial distance threshold is determined.

[0077] The above-mentioned projects refer to the specific application scenarios in which the vision system is deployed. The above-mentioned requirements refer to the functional requirements of the vision system for spatial measurement accuracy in this scenario. The spatial distance threshold is a technical standard used to calibrate the accuracy assessment after quantifying the functional requirements.

[0078] For example, if the project is used for precision assembly (such as mobile phone parts assembly), and the required spatial positioning error is ≤0.1mm, then the above spatial distance threshold should be set to 0.1mm or more stringent.

[0079] Since different projects typically have different requirements for the spatial distance measurement accuracy of the vision system, determining the spatial distance threshold based on the spatial distance measurement accuracy requirements of the project in which the vision system is located is beneficial to improving the accuracy of the determined spatial distance threshold, thereby improving the accuracy of the measurement of the calibration accuracy of the extrinsic parameters between the first camera and the second camera.

[0080] The above description introduced how spatial distance error is used to measure the calibration accuracy of extrinsic parameters. In practice, pixel distance error can also be used to measure the calibration accuracy of extrinsic parameters. That is: In some embodiments, before measuring the calibration accuracy of the extrinsic parameters between the first camera and the second camera based on the aforementioned spatial distance errors and a preset spatial distance threshold, the method further includes: Determine the reprojection error at the same corner point; Correspondingly, the calibration accuracy of the extrinsic parameters between the first camera and the second camera, measured based on the aforementioned spatial distance errors and preset spatial distance thresholds, includes: The calibration accuracy of the extrinsic parameters between the first camera and the second camera is measured based on the aforementioned spatial distance errors, preset spatial distance thresholds, aforementioned reprojection errors, and preset pixel distance thresholds.

[0081] Optionally, the aforementioned pixel distance threshold can be determined based on user feedback. For example, if user feedback indicates that the results obtained after 3D reconstruction using calibration parameters with a reprojection error of 0.1 to 0.2 pixels meet the project's accuracy requirements, then the pixel distance threshold can be set to 0.2 pixels.

[0082] In this embodiment of the application, the reprojection error of the same corner point can be determined in the following way: The pixel coordinates of two corner points corresponding to the same corner point are extracted. Distortion correction is then applied to these two corner point pixel coordinates according to the first and second distortion parameters to obtain the normalized imaging plane coordinates after distortion correction. Triangulation is performed on the two corner points based on the first intrinsic parameter of the first camera, the second intrinsic parameter of the second camera, and the extrinsic parameter between the first and second cameras to reconstruct a three-dimensional coordinate of the corner point. These three-dimensional coordinates are then reprojected onto the corresponding image planes using the intrinsic and extrinsic parameters of the first and second cameras respectively, resulting in the reprojected pixel coordinates of the two corner points. The distances between the reprojected corner point pixel coordinates and the corresponding originally extracted corner point pixel coordinates are calculated to obtain the reprojection error of the same corner point under the first and second cameras.

[0083] After determining the reprojection error of each corner point under the first and second cameras, the calibration accuracy of the extrinsic parameters between the first and second cameras can be measured by combining the spatial distance error.

[0084] For example, pixel evaluation values ​​(such as the mean reprojection error, maximum reprojection error, or root mean square error of reprojection error, etc.) are calculated based on the reprojection error of each corner point under the first camera. These determined pixel evaluation values ​​are then compared with preset pixel distance thresholds, and the calibration accuracy of the extrinsic parameters is measured based on the comparison result. It should be noted that if the number of preset pixel distance thresholds is equal to the number of determined pixel evaluation values...

[0085] In this embodiment, a corresponding spatial evaluation value (such as the fusion mean and / or fusion maximum value) can be determined based on each spatial distance error. This spatial evaluation value is then compared with a preset spatial distance threshold, and the pixel evaluation value is compared with a preset pixel distance threshold. If both the spatial evaluation value and the pixel evaluation value are less than the preset spatial distance threshold, it can be determined that the calibration accuracy of the extrinsic parameters between the first and second cameras simultaneously meets both spatial accuracy and pixel accuracy requirements, and the calibration result is reliable. Otherwise, the calibration accuracy is deemed substandard, and extrinsic parameter calibration needs to be performed again. By measuring the calibration accuracy of the extrinsic parameters from both spatial and pixel dimensions, the accuracy of the measurement results can be improved.

[0086] In this embodiment, after the first and second cameras in the vision system capture images of the same calibration board, the corner pixel coordinates of each calibration board image are extracted. Based on the extracted corner pixel coordinates, the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter obtained from calibrating the first and second cameras, and the extrinsic parameters between the first and second cameras, the spatial distance error between the two corner pixel coordinates corresponding to the same corner point is calculated. Then, the calibration accuracy of the extrinsic parameters between the first and second cameras is measured based on each spatial distance error and a preset spatial distance threshold. Since the aforementioned spatial distance error is calculated jointly based on the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter, and the extrinsic parameters between the first and second cameras, and the probability of changes in the corresponding first intrinsic parameter, first distortion parameter, second intrinsic parameter, and second distortion parameter after the first and second cameras leave the factory is low, the aforementioned spatial distance error can reflect the calibration accuracy of the extrinsic parameters between the first and second cameras. Therefore, the calibration accuracy of the extrinsic parameters between the first and second cameras can be accurately measured based on each spatial distance error and the preset spatial distance threshold. Furthermore, since the aforementioned spatial distance error reflects the accuracy of the extrinsic parameters in real space, the measurement result of the calibration accuracy of the extrinsic parameters obtained based on the spatial distance error is not affected by different brands, resolutions, or optical parameters. That is, regardless of whether the cameras in the vision system are homogeneous or heterogeneous, the calibration accuracy of the extrinsic parameters can be accurately measured through spatial distance error.

[0087] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0088] Corresponding to the method for measuring the accuracy of camera extrinsic parameter calibration described in the above embodiments, Figure 2This diagram illustrates a structural block diagram of a camera extrinsic parameter calibration accuracy measurement device provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0089] Reference Figure 2 The camera extrinsic parameter calibration accuracy measurement device 2 is applied to electronic equipment and includes: an intrinsic parameter calibration module 21, an extrinsic parameter calibration module 22, a corner pixel coordinate extraction module 23, a spatial distance error calculation module 24, and a calibration accuracy measurement module 25. Wherein: The intrinsic parameter calibration module 21 is used to calibrate the first intrinsic parameter and the first distortion parameter of the first camera in the vision system, and to calibrate the second intrinsic parameter and the second distortion parameter of the second camera in the vision system. The extrinsic parameter calibration module 22 is used to calibrate the extrinsic parameters between the first camera and the second camera. The corner pixel coordinate extraction module 23 is used to extract the corner pixel coordinates in each of the calibration board images after the first camera and the second camera take pictures of the same calibration board. The spatial distance error calculation module 24 is used to calculate the spatial distance error between two corner pixel coordinates corresponding to the same corner point based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration plate image. The calibration accuracy measurement module 25 is used to measure the calibration accuracy of the above-mentioned external parameters based on the above-mentioned spatial distance errors and preset spatial distance thresholds.

[0090] In this embodiment, after the first and second cameras in the vision system capture images of the same calibration board, the corner pixel coordinates in each calibration board image are extracted. Based on the extracted corner pixel coordinates, the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter, and extrinsic parameter obtained from calibrating the first and second cameras, the spatial distance error between the two corner pixel coordinates corresponding to the same corner is calculated. Then, the calibration accuracy of the extrinsic parameter between the first and second cameras is measured based on each spatial distance error and a preset spatial distance threshold. Since the aforementioned spatial distance error is calculated jointly based on the first intrinsic parameter, first distortion parameter, second intrinsic parameter, second distortion parameter, and extrinsic parameter, and the probability of changes in the corresponding first intrinsic parameter, first distortion parameter, second intrinsic parameter, and second distortion parameter after the first and second cameras leave the factory is low, the aforementioned spatial distance error can reflect the calibration accuracy of the extrinsic parameter. Therefore, the calibration accuracy of the extrinsic parameter can be accurately measured based on each spatial distance error and the preset spatial distance threshold. Furthermore, since the aforementioned spatial distance error reflects the accuracy of the extrinsic parameters in real space, the measurement result of the calibration accuracy of the extrinsic parameters obtained based on the spatial distance error is not affected by different brands, resolutions, or optical parameters. That is, regardless of whether the cameras in the vision system are homogeneous or heterogeneous, the calibration accuracy of the extrinsic parameters can be accurately measured through spatial distance error.

[0091] Optionally, the corner pixel coordinate extraction module 23 described above is specifically used for: After the first camera and the second camera take pictures of the same calibration board to obtain two images of the calibration board, the common viewing area between the two calibration board images is determined. Extract the corner pixel coordinates corresponding to the common viewing area from the two calibration board images mentioned above.

[0092] Optionally, the aforementioned spatial distance error calculation module 24 is specifically used for: Based on the first intrinsic parameter and the first distortion parameter, the corner pixel coordinates corresponding to the first camera are back-projected onto the camera coordinate system of the first camera to obtain the first three-dimensional coordinates in the first camera coordinate system. Based on the aforementioned second intrinsic parameters and the aforementioned second distortion parameters, the aforementioned corner pixel coordinates corresponding to the aforementioned second camera are back-projected onto the camera coordinate system of the aforementioned second camera to obtain the second three-dimensional coordinates in the aforementioned second camera coordinate system. Based on the aforementioned external parameters, the relative pose between the first camera and the second camera is determined. Based on the aforementioned relative pose, the first three-dimensional coordinates are transformed from the first camera coordinate system to the second camera coordinate system to obtain the third three-dimensional coordinates; and / or, based on the aforementioned relative pose, the second three-dimensional coordinates are transformed from the second camera coordinate system to the first camera coordinate system to obtain the fourth three-dimensional coordinates. Based on the aforementioned third and second three-dimensional coordinates corresponding to the same corner point, the first spatial distance error of the same corner point is determined, and / or, based on the aforementioned first and fourth three-dimensional coordinates corresponding to the same corner point, the second spatial distance error of the same corner point is determined.

[0093] Optionally, the calibration accuracy measurement module 25 described above is specifically used for: Calculate the mean of the first spatial distance error for each of the above-mentioned corner points to obtain the first mean; And / or, Calculate the mean of the second spatial distance error for each of the above-mentioned corner points to obtain the second mean; The fusion mean is determined based on the first mean and / or the second mean mentioned above; The calibration accuracy of the aforementioned extrinsic parameters is measured based on the fusion mean and the preset spatial distance threshold.

[0094] Optionally, the calibration accuracy measurement module 25 described above is specifically used for: Determine the maximum value of the first spatial distance error for each of the above-mentioned corner points to obtain the first maximum value; And / or, Determine the maximum value of the second spatial distance error for each of the above-mentioned corner points to obtain the second maximum value; The maximum fusion value is determined based on the first maximum value and / or the second maximum value mentioned above; The calibration accuracy of the above extrinsic parameters is measured based on the maximum fusion value and the preset spatial distance threshold.

[0095] In some embodiments, the camera extrinsic parameter calibration accuracy measuring device 2 provided in this application further includes: The spatial distance threshold determination module is used to determine the spatial distance threshold according to the requirements of the project in which the vision system is located for spatial distance measurement accuracy before measuring the calibration accuracy of the external parameters based on the aforementioned spatial distance errors and preset spatial distance thresholds.

[0096] In some embodiments, the camera extrinsic parameter calibration accuracy measuring device 2 provided in this application further includes: The reprojection error determination module is used to determine the reprojection error of the same corner point before measuring the calibration accuracy of the external parameters based on the aforementioned spatial distance errors and preset spatial distance thresholds.

[0097] Correspondingly, the calibration accuracy measurement module 25 mentioned above is specifically used for: The calibration accuracy of the aforementioned extrinsic parameters is measured based on the aforementioned spatial distance errors, the preset spatial distance threshold, the aforementioned reprojection errors, and the preset pixel distance threshold.

[0098] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0099] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 As shown, the electronic device 3 of this embodiment includes: at least one processor 30 ( Figure 3 The diagram shows only one processor, a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, wherein the processor 30 executes the computer program 32 to implement the steps in any of the above method embodiments.

[0100] The electronic device 3 can be a desktop computer, laptop, handheld computer, or cloud server, etc. This electronic device may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0101] The processor 30 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0102] In some embodiments, the memory 31 may be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. In other embodiments, the memory 31 may be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. Furthermore, the memory 31 may include both internal and external storage units of the electronic device 3. The memory 31 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0103] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0104] This application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above method embodiments.

[0105] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method embodiments.

[0106] This application provides a computer program product that, when run on an electronic device, enables the electronic device to implement the steps described in the various method embodiments above.

[0107] If the integrated unit is implemented as 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, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0108] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0109] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0110] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0111] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

Claims

1. A method for measuring the accuracy of camera extrinsic parameter calibration, characterized in that, include: The first intrinsic parameter and the first distortion parameter of the first camera in the vision system are calibrated, and the second intrinsic parameter and the second distortion parameter of the second camera in the vision system are calibrated. Calibrate the extrinsic parameters between the first camera and the second camera; After the first camera and the second camera capture images of the same calibration board, the corner pixel coordinates of each calibration board image are extracted. Based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration board image, calculate the spatial distance error between the two corner pixel coordinates corresponding to the same corner point; The calibration accuracy of the external parameters is measured based on the spatial distance errors and the preset spatial distance threshold.

2. The method for measuring the accuracy of camera extrinsic parameter calibration as described in claim 1, characterized in that, After the first camera and the second camera capture images of the same calibration board, the corner pixel coordinates of each calibration board image are extracted, including: After the first camera and the second camera capture two images of the same calibration board, the common viewing area between the two calibration board images is determined. The corner pixel coordinates corresponding to the common viewing area are extracted from the two calibration board images respectively.

3. The method for measuring the accuracy of camera extrinsic parameter calibration as described in claim 1, characterized in that, The step of calculating the spatial distance error between two corner pixel coordinates corresponding to the same corner point based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration board image includes: Based on the first intrinsic parameter and the first distortion parameter, the corner point corresponding to the first camera is... The pixel coordinates are back-projected onto the camera coordinate system of the first camera to obtain the first three-dimensional coordinates in the first camera coordinate system; Based on the second intrinsic parameter and the second distortion parameter, the corner pixel coordinates corresponding to the second camera are back-projected onto the camera coordinate system of the second camera to obtain the second three-dimensional coordinates in the second camera coordinate system; The relative pose between the first camera and the second camera is determined based on the extrinsic parameters. Based on the relative pose, the first three-dimensional coordinates are transformed from the first camera coordinate system to the second camera coordinate system to obtain the third three-dimensional coordinates; and / or, based on the relative pose, the second three-dimensional coordinates are transformed from the second camera coordinate system to the first camera coordinate system to obtain the fourth three-dimensional coordinates; Based on the third three-dimensional coordinates and the second three-dimensional coordinates corresponding to the same corner point, a first spatial distance error of the same corner point is determined, and / or, based on the first three-dimensional coordinates and the fourth three-dimensional coordinates corresponding to the same corner point, a second spatial distance error of the same corner point is determined.

4. The method for measuring the accuracy of camera extrinsic parameter calibration as described in claim 3, characterized in that, The step of measuring the calibration accuracy of the extrinsic parameters based on each of the spatial distance errors and a preset spatial distance threshold includes: Calculate the mean of the first spatial distance error for each of the same corner points to obtain the first mean; And / or, Calculate the mean of the second spatial distance error for each of the same corner points to obtain the second mean; The fusion mean is determined based on the first mean and / or the second mean; The calibration accuracy of the extrinsic parameters is measured based on the fusion mean and a preset spatial distance threshold.

5. The method for measuring the accuracy of camera extrinsic parameter calibration as described in claim 3, characterized in that, The step of measuring the calibration accuracy of the extrinsic parameters based on each of the spatial distance errors and a preset spatial distance threshold includes: Determine the maximum value of the first spatial distance error for each of the aforementioned corner points to obtain the first maximum value; And / or, Determine the maximum value of the second spatial distance error for each of the aforementioned corner points to obtain the second maximum value; The maximum fusion value is determined based on the first maximum value and / or the second maximum value; The calibration accuracy of the extrinsic parameters is measured based on the maximum fusion value and the preset spatial distance threshold.

6. The method for measuring the accuracy of camera extrinsic parameter calibration as described in claim 1, characterized in that, Before measuring the calibration accuracy of the extrinsic parameters based on each of the spatial distance errors and a preset spatial distance threshold, the method further includes: The spatial distance threshold is determined based on the accuracy requirements for spatial distance measurement in the project in which the vision system is located.

7. The method for measuring the accuracy of camera extrinsic parameter calibration as described in any one of claims 1 to 6, characterized in that, Before measuring the calibration accuracy of the extrinsic parameters based on each of the spatial distance errors and a preset spatial distance threshold, the method further includes: Determine the reprojection error at the same corner point; The step of measuring the calibration accuracy of the extrinsic parameters based on each of the spatial distance errors and a preset spatial distance threshold includes: The calibration accuracy of the extrinsic parameters is measured based on the various spatial distance errors, the preset spatial distance threshold, the various reprojection errors, and the preset pixel distance threshold.

8. A measuring device for calibrating the accuracy of camera extrinsic parameters, characterized in that, include: An intrinsic parameter calibration module is used to calibrate the first intrinsic parameter and the first distortion parameter of the first camera in the vision system, and to calibrate the second intrinsic parameter and the second distortion parameter of the second camera in the vision system. An extrinsic parameter calibration module is used to calibrate the extrinsic parameters between the first camera and the second camera. The corner pixel coordinate extraction module is used to extract the corner pixel coordinates in each calibration board image after the first camera and the second camera take pictures of the same calibration board. The spatial distance error calculation module is used to calculate the spatial distance error between two corner pixel coordinates corresponding to the same corner point based on the first intrinsic parameter, the first distortion parameter, the second intrinsic parameter, the second distortion parameter, the extrinsic parameter, and the corner pixel coordinates in the calibration board image. The calibration accuracy measurement module is used to measure the calibration accuracy of the external parameters based on the spatial distance errors and preset spatial distance thresholds.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

11. A computer program product, characterized in that, Includes a computer program, which, when run, causes the electronic device to perform the method according to any one of claims 1 to 7.

Images